A deep learning method with a convolutional neural network (CNN) can support the evaluation of small solid renal masses in dynamic CT images with acceptable diagnostic performance. Between 2012 and 2016, researchers at Japan’s Okayama University studied 1,807 image sets from 168 pathologically diagnosed small (≤ 4 cm) solid renal masses with four CT phases—unenhanced, corticomedullary, nephrogenic, and excretory—in 159 patients. Masses were classified as malignant (n = 136) or benign (n = 32) using a 5-point scale, and this dataset was then randomly divided into five subsets. As first AJR author, Takashi Tanaka explained, “four were used for augmentation and supervised training (48,832 images), and one was used for testing (281 images).” Utilizing the Inception-v3 architecture CNN model, the AUC for malignancy and accuracy at optimal cutoff values of output data were evaluated in six different CNN models. Finding no significant size difference between malignant and benign lesions, Tanaka’s team did find that the AUC value of the corticomedullary phase was higher than that of other phases (corticomedullary vs excretory, p = 0.022). Additionally, the highest accuracy (88%) was achieved in the corticomedullary phase images. Multivariate analysis revealed that the CNN model of corticomedullary phase was a significant predictor for malignancy, “compared with other CNN models, age, sex, and lesion size.”
A Stanford University team has developed a quantitative framework able to sonographically differentiate between benign and malignant thyroid nodules at a level comparable to that of expert radiologists, which may prove useful for establishing a fully automated system of thyroid nodule triage. Alfiia Galimzianova et al. retrospectively collected ultrasound images of 92 biopsy-confirmed nodules, which were annotated by two expert radiologists using the American College of Radiology’s Thyroid Imaging Reporting and Data System (TI-RADS). In the researchers’ framework, nodule features of echogenicity, texture, edge sharpness, and margin curvature properties were analyzed in a regularized logistic regression model to predict nodule malignancy. Authenticating their method with leave-one-out cross-validation, the Stanford team used ROC AUC, sensitivity, and specificity to compare the framework’s results with those obtained by six expert annotation-based classifiers. Galimzianova et al. noted that the AUC of the proposed framework measured 0.828 (95% CI, 0.715–0.942)—“greater than or comparable to that of the expert classifiers”—whose AUC values ranged from 0.299 to 0.829 (p = 0.99). Additionally, in a curative strategy at sensitivity of 1, use of the framework could have avoided biopsy in 20 of 46 benign nodules—statistically significantly higher than three expert classifiers. In a conservative strategy at specificity of 1, the framework could have helped to identify 10 of 46 malignancies— statistically significantly higher than five expert classifiers. “Our results confirm the ultimate feasibility of computer-aided diagnostic systems for thyroid cancer risk estimation,” concluded Galimzianova. “Such systems could provide second-opinion malignancy risk estimation to clinicians and ultimately help decrease the number of unnecessary biopsies and surgical procedures.”
An AJR article reviewing various techniques and clinical management paradigms to treat severe frostbite injuries— especially relevant for interventional radiologists showed promising results using both intraarterial (IA) and IV tPA (tissue plasminogen activator) to reduce amputation. “Severe frostbite injuries can lead to devastating outcomes with loss of limbs and digits, yet clinical management continues to consist primarily of tissue rewarming, prolonged watchful waiting, and often delayed amputation,” wrote Boston Medical Center radiologists John Lee and Mikhail Higgins. A search of the literature by Lee and Higgins yielded 157 publications. After manually screening for inclusion criteria of case reports, case series, cohort studies, and randomized prospective studies that reported the use of tPA to treat severe frostbite injuries, 16 qualified for review. Lee and Higgins’ analyzed series included 209 patients with 1,109 digits at risk of amputation treated with IA or IV tPA—116 and 77 patients, respectively. A total 926 at-risk digits were treated with IA tPA and resulted in amputation of 222 digits, for a salvage rate of 76%. Twenty-four of 63 patients underwent amputation after IV tPA, resulting in a 62% salvage rate. Both digital subtraction angiography and triple-phase bone scan were utilized for initial imaging evaluation of patients with severe frostbite injuries. Additional concurrent treatment included therapeutic heparin at 500 U/h, warfarin with target international normalized ratio of 2:3, nonsteroidal antiinflammatory drugs, pain management, and light dressings with topical antimicrobial agents. “For many years,” Lee and Higgins concluded, “the axiom ‘frostbite in January, amputate in July’ was an accurate description of the common outcome in frostbite injuries. Through a meta-analysis of thrombolytic therapy in the management of severe frostbite, this article provides a useful guideline for interventional radiologists, including a suggested protocol, inclusion and exclusion criteria, and potential complications.”
Mobile devices proved both reliable and accurate for the clinical decision to administer IV thrombolysis in patients with acute stroke, according to an ahead-of-print article in the April issue of AJR. To assess reliability and accuracy of IV thrombolysis recommendations made after interpretation of head CT images of patients with acute stroke symptoms displayed on smartphone or laptop reading systems—compared with those made after interpretation of images displayed on a medical workstation monitor—Antonio J. Salazar at the University of Los Andes in Bogotá, Columbia utilized a factorial design with 188 patients, four radiologists, and three reading systems to produce 2,256 interpretations. To evaluate reliability, Salazar and colleagues calculated the intraobserver and interobserver agreements using the intraclass correlation coefficient (ICC) and five interpretation variables: hemorrhagic lesions, intraaxial neoplasm, stroke dating (acute, subacute, chronic), hyperdense arteries, and infarct size assessment. Accuracy equivalence tests were also performed for the IV thrombolysis recommendation; for this variable, sensitivity, specificity, and ROC curves were evaluated. Good or very good intraobserver agreements were observed for all the variables. Specifically, for those variables required to establish contraindications for IV thrombolysis, the agreements were ranked as very good. “This finding is important,” wrote Salazar et al., “because it reflects the good performance of mobile devices to evaluate the most significant imaging variables for clinical decisions.” For IV thrombolysis recommendation, the main subject of this evaluation, the interobserver agreements for the three reading systems were ranked as very good (ICC > 0.88). Similarly, very good intraobserver agreements were observed for all comparisons (ICC > 0.84). The AUC values (0.83–0.84) and sensitivities (0.94–0.95) for IV thrombolysis recommendation were equivalent among all the reading systems at a 5% equivalent threshold. As a unique assessment of imaging-based recommendations for the administration of IV recombinant tissue plasminogen activator based on unenhanced brain CT scans, Salazar also noted: “These results constitute a strong foundation for the development of mobilebased telestroke services because they increase neuroradiologist availability and the possibility of using reperfusion therapies in resource-limited countries.”
Clinical Associate Professor, Trauma & Emergency Radiology NYU Langone Health, Bellevue Hospital
Eric A. Roberge
Assistant Professor of Radiology Uniformed Services University of the Health Sciences
Suzanne Chong
Associate Professor, Emergency Radiology Division, Radiology and Imaging Sciences Department Indiana University Health
Published April 20, 2020
Mark P. Bernstein’s “Mass Casualty Incidents: An Introduction for Imagers” was published in the Winter 2020 issue of ARRS’ InPractice magazine. Below, Bernstein et al. provide a primer for hospitals and health care systems responding to the disaster surge of COVID-19.
The coronavirus disease (COVID-19) pandemic has created a mass casualty disaster of staggering proportions. By April 2020, the novel coronavirus responsible for COVID-19 had forced many parts of the United States into crisis mode, while others race to prepare for the inevitable. In regions where the case numbers have not yet begun to climb, disaster planning teams have time to prepare for a crisis response and implement lessons learned from those who were impacted earlier. The goal is the greatest good for the greatest number of people, so hospitals and health care systems are turning the focus from individual health to population health in their disaster surge response to save as many lives as possible.
Mass casualty incidents (MCIs) can be man-made acts of violence, such as mass shootings, bioterrorism, or exploding bridges, or natural disasters in the form of earthquakes, tornados, tsunamis, and pandemics. Tragedies of intentional violence or infrastructure disasters create a sudden surge, demanding a rapid shift in a hospital’s daily routine, and are usually limited geographically—for example, the site of an active shooter or a train derailment. Natural disasters, however, cover much larger regions (i.e., the path of a tornado), whereas, by definition, pandemics know no boundaries.
One key variable in these disasters is time. Time, in most cases, determines our ability to prepare for and maintain a disaster response. In trauma MCIs, there is a window of time when patients arrive to local hospitals, which is often measured in minutes to hours. In the case of bioterrorism or pandemics, timelines are prolonged, measured in days to weeks. Regarding the ongoing COVID-19 pandemic, the window of time is indefinite and unknown. The disruption of a hospital’s daily routine for prolonged periods of time and the need for resources beyond those available, or worse, outstrips the supply chain, placing severe strain on the health care system. Our best tools to manage these challenges are preparation, planning, and practice.
Preparation and planning take place from the federal and state levels to the community and local health care facility levels. Community planning should be coordinated with local governmental agencies, in accordance with state and federal disaster planning efforts, and integrated with local public health and emergency medical services. With respect to pandemics, community strategies must make every effort to “flatten the curve” in order to break the chain of transmission and slow the spread of infections. At the same time, hospital system strategies “raise the roof” of surge response by increasing health care system capacity (Fig. 1) through predesigned efforts focused on three factors: space, staff, and supplies. The hospital system is the backbone of these three elements.
Figure 1 – Community efforts to “flatten the curve” of coronavirus infections often intersect with health care system strategies to “raise the roof” for patient capacity (modified from Disaster Med Public Health Prep with permission from the Society for Disaster Medicine and Public Health).
Strategies for increasing health care system capacity will include conservation and substitution during a conventional response, adaptation and recycling during a contingency response, and, finally, reallocation of resources during a crisis response—essentially, withholding resources from one patient population to use them more effectively on another patient population. These “raise the roof” strategies involve nuanced ethical and legal considerations that must be addressed in advance, authorized by hospital leadership, and communicated clearly to frontline health care workers.
System
Ultimately, the hospital system component directs the response that determines the allocation of the three critical resources of space, staff, and stuff, which are based on supply and demand.
A robust hospital incident command system provides broad management for a multitude of issues, including: hospital controls (facility access, ventilation), communication (internal and external), community coordination (health care facilities, state and federal agencies, as well as utilities and supply chains), and continuity of emergency health care operation (vis-à-vis utility or other system failures). The hospital incident command should also determine and communicate which disaster response is being utilized. Disaster response can be described, in escalating intensity, as conventional, contingency, and crisis, dependent on surge severity and resource availability. The more severe the surge, the fewer the resources; the lower the hospital’s capacity to take care of victims, the more quickly the disaster response must shift into a higher mode (Fig. 2).
Space
Upon declaration of an MCI, efforts must be made to free up physical space for patients. The size and nature of the disaster will dictate the scope and speed necessary.
Figure 2 – As the hospital incident command system escalates the intensity of disaster response—from conventional to contingency to MCI—the minimum acceptable standard of care for patients is diminished (modified from Disaster Med Public Health Prep with permission from the Society for Disaster Medicine and Public Health).
The conventional response is for surges causing a 20% increase in patients beyond normal capacity. In this situation, all staffed beds are made available and filled. Elective procedures are postponed or cancelled, and patient discharge plans are activated to dedicate more space and empty beds to the surge.
A contingency response is used for surges that are twice a hospital’s capacity and demands more aggressive actions. As the numbers of patients greatly exceed the available hospital and critical care beds, hospital spaces designed for other purposes, including step-down units, observation units, and procedure suites, can be repurposed to recruit more space to bed patients. Transferring patients to other available facilities for ongoing, nonemergent care can be initiated.
A crisis situation completely overwhelms a health care facility. Patients fill hallways, and makeshift spaces, such as tents and offices, need to be devised. Erecting tent hospitals with intensive care units in city parks, converting convention centers into field hospitals, and docking of the United States Naval Ship (USNS) Comfort in Manhattan and USNS Mercy in Los Angeles are evidence that our nation is in crisis because of the COVID-19 pandemic.
Staff
As more space becomes available, achieving appropriate staffing and obtaining adequate supplies for the surge of patients is vital. The hospital incident command system should be convened for action as soon as a disaster is declared to urgently alert and mobilize necessary staff. The type of injuries that are expected (e.g., blunt trauma, penetrating trauma, or biological agent) will determine the type of staff best suited to respond. If staffing levels are insufficient, measures to increase staffing may be warranted, including expanding the scope of responsibilities, lengthening shifts, and enlarging patient-to-nurse ratios.
In a conventional response, trained and credentialed staff are able to care for patients with minor modifications, while maintaining usual standards of care.
The standard of care is challenged in a contingency response, as adequately trained staff must train and supervise off-service staff to safely provide care. Bringing in additional staff should be considered, and outside staff need to be given emergency privileges and credentialing.
A crisis response demands staff to perform clinical functions outside their usual domain. Aggressive staff recruitment and rapid training are necessary to meet the patient care demands and volume. During crisis mode, triage becomes necessary to ensure that acceptable care is provided for the largest number of people. Over- and under-triage can result in higher mortality rates.
Supplies (“Stuff”)
Supplies include medications, medical equipment, and personal protective equipment (PPE). Considerations must also be made for laboratory reagents, diagnostic testing, as well as for food, water, and linens.
The hospital system must be aware of onsite and offsite supply storage and availability through supply chains. The ability to adapt, reuse, and reallocate becomes necessary in both contingency and crisis situations.
In the current COVID-19 pandemic, we are witnessing contingency and crisis responses. Hospitals are experiencing severe shortages of ventilators and PPE, meaning patients may be deprived of life-saving care and health care providers are likely to be infected with dire, cascading ramifications.
Radiology Department Response
A departmental incident command team should be in place to implement a disaster management plan and engage in clear and consistent communication. The radiology department must have containment and mitigation strategies that ensure the safety of all staff and patients being imaged. For COVID-19, these measures include ensuring adequate PPE, especially for frontline technicians performing imaging studies, enforcing physical distancing, and limiting in-person interactions. Remote reading should be instituted, where possible. Decontamination protocols must be defined and executed. Nonemergent studies should be halted, including interventional procedures, to preserve PPE and limit exposure.
All real-time changes to address incident-specific issues should be frequently updated and communicated. Implementing these types of measures allows radiology departments to provide safe and appropriate care during surges and helps to ensure sustainable operations.
The lessons we learn from responses nationally and internationally should be incorporated into our hospital and departmental MCI and disaster planning process. Our ability to plan and prepare by focusing on system, space, staff, and stuff will make all the difference in the number of lives saved.
Suggested Reading
Christian MD, Devereaux AV, Dichter JR, Rubinson L, Kissoon N. Introduction and executive summary: care of the critically ill and injured during pandemics and disasters: CHEST consensus statement. Chest 2014; 146:8S–34S
Institute of Medicine (US) Committee on Guidance for Establishing Crisis Standards of Care for Use in Disaster Situations; Altevogt BM, Stroud C, Hanson SL, Hanfling D, Gostin LO, eds. Guidance for establishing crisis standards of care for use in disaster situations: a letter report. Washington, DC: The National Academies Press, 2009
Institute of Medicine (US) Committee on Guidance for Establishing Crisis Standards of Care for Use in Disaster Situations; Hanfling D, Altevogt BM, Viswanathan K, Gostin LO, eds. Crisis standards of care: a systems framework for catastrophic disaster response. Washington, DC: The National Academies Press, 2012
Institute of Medicine (US) Committee on Crisis Standards of Care: A Toolkit for Indicators and Triggers; Hanfling D, Hick J, Stroud C, eds. Crisis standards of care: a toolkit for indicators and triggers. Washington, DC: The National Academies Press (US), 2013
The views expressed are those of the authors and do not reflect the official policy of the Department of the Army, the Department of Defense, or the U.S. Government.
Associate Executive Director Diagnostic Radiology American Board of Radiology
Published April 10, 2020
Along with the Radiological Society of North America, American College of Radiology, American Radium Society, and American Medical Association Section on Radiology, the American Roentgen Ray Society (ARRS) co-sponsored the founding of the American Board of Radiology (ABR) in 1934. The mission of the ABR is to certify that our diplomates have demonstrated and maintained the requisite knowledge, skill, and understanding of their disciplines for the benefit of their patients.
Board certification serves as an important marker for the highest standard of care. It reflects the critical core values of compassion, patient-centeredness, and a commitment to life-long learning. Patients, physicians, medical physicists, health care providers, insurers, and quality organizations look for board certification as the best measure of a physician’s or medical physicist’s knowledge, experience, and skills to provide quality care within a given specialty.
Board certification and participation in a maintenance of certification (MOC) program has many benefits. It assures patients, privileging committees, payers, and regulators that the physician has successfully completed a training program and continues to expand his or her medical knowledge, which leads to improvements in their practice and patient safety. In 2007, the ABR instituted a requirement for practice quality improvement projects, which must be relevant to one’s practice, achievable, provide measurable results, and likely to improve quality. This remains a major component of the MOC program. Over the years, the number of qualified projects and participatory activities has been greatly expanded to reflect the integration of radiologists into the health care system.
Meet your MOC Requirements Leverage these member-exclusive benefits to meet your educational requirements with the ABR.
Since the founding of the ABR in 1934, the field of radiology has grown dramatically, and it became increasingly difficult to master the entire field. Thus, separate residency training programs were developed for diagnostic radiology and radiation oncology. (Most recently, a primary residency for interventional radiology has been approved.) Continued advances and the development of new imaging modalities resulted in many diagnostic radiologists restricting their practice domains to some extent. The ABR responded by providing subspecialty certification to reflect the importance of subspecialization. Subspecialty certification was offered for pediatric radiology and vascular and interventional radiology in 1994, for neuroradiology in 1995, and nuclear radiology in 1999. Given the speed with which these many advances in medical science changed the field of radiology, it became apparent that remote board certification was no longer pertinent. Something was needed to assure the public that physicians were keeping up with these new developments. The four subspecialty certificates offered by the ABR were timelimited from their inception, and the last lifetime primary certificates issued by the ABR were given in 2001. MOC would now be required to maintain ABR certification for all but lifetime certificate holders.
The four components of MOC are:
professionalism and professional standing
life-long learning and self-assessment
assessment of knowledge, judgment, and skills and
improvement in medical practice.
Originally, these requirements were met by maintaining an unrestricted state medical license in each state of practice, participating in continuing medical education that includes self-assessment, a cognitive exam, and participation in quality improvement projects. The ABR MOC requirements have been modified over the years, based on feedback from our diplomates.
Initially, a cognitive exam was required to satisfy Part 3 for MOC participants. However, this required radiologists to take time away from their practices and to pay for expenses to travel to a testing center. Furthermore, the cognitive assessment was required only every 10 years, an interval that many considered too long. An improved program was needed that would be meaningful, but not onerous for the diplomates.
The ABR Online Longitudinal Assessment (OLA) was introduced for diagnostic radiology in 2019, and for interventional radiology, radiation oncology, and medical physics in 2020. Each week, participating diplomates receive an email giving them the opportunity to answer one or two questions. Most diplomates are required to answer 52 questions a year. (Some with multiple certificates are required to answer more questions.) These questions were designed to test “walking around knowledge”—information diplomates should know “off the top of their head,” if asked by a colleague, resident, or patient. Furthermore, it is a learning experience, as the rationale for the correct answers and a reference is provided immediately.
Reaction to OLA has largely been positive. Many radiologists enjoy receiving two questions every week in their selected areas of practice. Since the “shelf life” of a question is four weeks, diplomates can elect to answer eight questions every four weeks, if they prefer “batching” the questions rather than answering two questions every week. Most radiologists have enjoyed participating in OLA, as it takes only a few minutes each week and does not require travel. Many continue to answer the weekly questions even after completing their yearly requirement of 52 items. More than 20,000 radiologists are now actively participating in MOC.
Questions for all ABR examinations are written by volunteers and reviewed by a subspecialty committee, before being submitted to be included in the cognitive assessment. The next step is the test assembly meetings, where all questions are again reviewed. Despite this rigorous process, an occasional problematic question may appear on an examination. These are picked up when ABR staff review the results of the exam. The ABR is fortunate to have two psychometricians and multiple experienced exam developers on staff, who review any potentially questionable item. Often, problematic questions are referred to a radiologist member of the Board of Trustees or the appropriate Committee Chair to participate in the decision whether to keep or remove the item from examination scoring.
Hear directly from ABR leadership as they provide an overview of the new ABR OLA program, which replaces the every ten-year MOC exam.
The ABR is a non-profit organization, which is highly dependent upon its many volunteers. The ABR has more than 900 diagnostic radiologists serving as volunteers on 68 different committees. Most of the paid office staff live in the Tucson, Arizona area and work at the ABR office building. Volunteers do much of their work electronically, but they do have periodic committee meetings in the Chicago testing center, near O’Hare airport, or at ABR headquarters in Tucson.
The volunteers contribute their time and expertise in writing questions, reviewing them for image quality and appropriateness as well as for constructing examinations. Before an examination is administered, another group of volunteers sets the passing standard (cutscore) using the Angoff Method. The Angoff Method is done by having a group of subject matter experts— many of whom are residency program directors—evaluate each item to estimate the proportion of minimally competent candidates who would correctly answer the item. The cutscore is the score that the panel estimates a minimally qualified candidate would receive. This is the legally defensible method used for many high-stakes examinations in the United States.
The goal of the ABR is to conduct examinations in which the candidates are comfortable and can do their best in demonstrating their knowledge. Thus, videos have been created to demonstrate the examination experience in both Chicago and Tucson. The ABR also communicates with their diplomates, candidates for certification, and the public through a variety of other means. The ABR has a booth at several of the larger radiology meetings to provide in-person answers and advice to attendees. The BEAM is the ABR’s newsletter that has recently increased from three to six issues a year. The ABR’s blog received more than 23,000 views last year. Additional communication efforts, which began in November 2018, include the social media outlets of Facebook, Twitter, Instagram, and LinkedIn.
The American Board of Radiology, along with the other 23 American Board of Medical Specialty member boards, strives to advance our field, improve patient care, and protect the public by assuring that our diplomates have acquired and maintained the requisite knowledge and skills to be effective practitioners. Board certification is an important marker of those attributes.
Associate Professor, Emergency Radiology University of Washington
Published April 23, 2020
Advancements in the accessibility, speed, and image quality of MDCT in the last 20 years have guaranteed that MDCT is the preferred imaging modality for evaluation of most conditions presenting in the emergency department, and this is particularly true for imaging after trauma. Vascular injuries, including those involving the thoracic and abdominal aorta, abdominal mesentery, pelvis, cervical vessels, and upper and lower extremities, are an uncommon but potentially lethal outcome of both penetrating and blunt trauma. Historically, diagnosis of vascular injuries relied on open exploration or conventional catheter-based angiography (CTA), both of which are invasive, time-consuming, and not broadly accessible. Today, however, the diagnosis or exclusion of many of these injuries is made on MDCT, oftentimes obviating the need for more invasive techniques. The timing and type of endovascular repair, particularly for aortic injuries, has also evolved.
Most deaths from traumatic aortic injuries still occur at the scene or before the patient reaches the hospital. However, those who reach the emergency department and undergo imaging can now be treated with a high rate of success. The majority of traumatic thoracic aortic injuries (TTAI) are found at the aortic isthmus. Injuries are found less commonly at the aortic root, ascending aorta, distal descending aorta, or at branch vessel origins, with multifocal injuries in up to 18% of patients. Although most aortic injuries are associated with high-energy trauma, there is no consensus as to the precise definition of “high-energy.” Furthermore, direct chest trauma or visible external signs of chest trauma are not necessary for the diagnosis. Therefore, liberal screening with chest CTA is encouraged in any patient with more than minimal deceleration injury.
Initial screening for mediastinal hematoma may be performed with a portable chest radiograph in many institutions. Signs suggesting mediastinal hematoma, and thus raising the possibility of a surgically relevant aortic injury, include right paratracheal stripe thickening, superior mediastinal widening, aortic arch enlargement or irregularity, opacification of the aortopulmonary window, rightward displacement of the trachea or enteric tube, inferior displacement of the left mainstem bronchus, obscuration of the descending aorta, widening of the paraspinal lines, or apical capping. These radiographic features are neither sensitive nor specific, and absence should not preclude chest CTA in high-risk trauma victims. Contrast-enhanced chest MDCT has very high sensitivity and a negative predictive value for acute aortic injuries, and it should be obtained in all at risk patients. Thoracic CTA with cardiac gating or ultrahigh pitch is the diagnostic study of choice, but nongated CTA is sufficient in most patients. Intimal flaps, intraluminal thrombus, intramural hematoma, irregular external aortic contour, focal luminal dilation or saccular outpouching (also known as pseudoaneurysm), or active extravasation are direct signs of aortic injury. Injuries can be graded according to the Society of Vascular Surgery (SVS) system or using a newer system of minimal/moderate/severe injuries that more directly guides management. Minimal aortic injuries have no external aortic contour deformity and intimal tear or intraluminal thrombus less than 10 mm in size (equivalent to SVS grade 1); these injuries do not require operative intervention and instead receive antiplatelet therapy for 4–6 weeks, with optional follow-up imaging. Moderate and severe TTAIs require surgical intervention.
Clinical Case-Based Review of Vascular and Interventional Imaging
Only about 5% of blunt aortic injuries involve the abdominal aorta. Sufficiently rare that many radiologists may never encounter one during their careers, this injury should be specifically sought when patients present with blunt abdominal trauma, such as a seatbelt sign or abdominal impact on the steering wheel, and have spinal fractures (particularly flexion-distraction injuries), duodenal or small bowel injuries, or pancreatic injuries. Isolated blunt abdominal aortic injuries (BAAI) are also rare. Two-thirds of BAAIs occur between the renal arteries and the aortic bifurcation, and up to one-quarter also have injuries involving the thoracic aorta. Abdominal CTA is the diagnostic study of choice, though venous-phase abdominal MDCT is sufficient for diagnosis and preintervention planning in most cases. In patients stable enough to be evaluated on MDCT, the most common appearance of BAAI is intimal flaps or intimal thrombi without external aortic contour deformity. Pseudoaneurysms of the abdominal aorta are only seen in 16% but require repair, either open or endovascular, depending on location. In the absence of external contour abnormality, BAAI can be managed nonoperatively with antiplatelet therapy and beta blockers. It is important to note that neither TTAI or BAAI can be excluded on a unenhanced CT because injuries, particularly intraluminal thrombi and intimal flaps, may not be accompanied by periaortic hematoma or stranding.
Injuries of the mesenteric vasculature are also uncommon in blunt trauma patients, more often resulting from penetrating trauma. Unfortunately, these are frequently lethal due to exsanguination, reflecting the difficulty in obtaining control of the proximal superior mesenteric artery (SMA), as well as back-bleeding from the valveless portomesenteric venous system. Though uncommon at initial laparotomy, bowel infarction and subsequent sepsis and multiple organ system failure are responsible for the bulk of delayed deaths from mesenteric vascular injury. Classification systems by the American Association for the Surgery of Trauma-Organ Injury Scale (AAST-OIS) and by Fullen et al. are both anatomy-based, reflecting the greater surgical difficulty and poorer outcomes associated with more proximal mesenteric arterial or venous injuries. Although immediate operative evaluation is appropriate in any patient with penetrating trauma to the peritoneum or with blunt trauma in extremis, patients with hemodynamic stability following blunt abdominal trauma can be imaged with contrast-enhanced MDCT. On MDCT, direct signs of surgically important mesenteric vascular injuries include mesenteric vascular beading, abrupt termination, or active extravasation. Intraperitoneal low- or intermediate-density free fluid is highly sensitive for either bowel or mesenteric injury, as is abnormal bowel wall thickening or enhancement, and surgical exploration is appropriate when any of these are found on MDCT. The absence of intraperitoneal free fluid has a high negative predictive value for surgically important mesenteric or bowel injury. Isolated mesenteric stranding or hematoma without active extravasation does not necessarily need surgical exploration, but these patients should be monitored carefully for delayed presentation of CT-occult bowel injury or mesenteric injury resulting in bowel ischemia.
Hemorrhage from pelvic ring injuries can be significant and life-threatening. Arterial hemorrhage accounts for 15–20% of pelvic bleeding, and low-pressure bleeding from venous structures or fractured edges of cancellous bone account for the remainder. These low-pressure bleeding sites are usually controlled by pelvic sheeting, external fixation, or internal pelvic packing, whereas arterial hemorrhage is amenable to endovascular control. Early triage to angiography may be considered for those patients with obturator ring fractures displaced at least 1 cm or pubic symphyseal diastasis of at least 1 cm, as these are independent predictors of major hemorrhage. If a patient with pelvic ring injuries is hemodynamically stable, multiphase MDCT can improve the sensitivity and specificity of detection of pelvic bleeding. Ideally, an arterial phase is obtained to identify arterial injury, as opposed to venous injury, and an additional phase differentiates active bleeding from pseudoaneurysm. Unenhanced MDCT or dual-energy CT with virtual unenhanced images may be necessary to identify bone fragments that mimic pseudoaneurysm or active extravasation. Absence of contrast extravasation on MDCT has a high negative predictive value for clinically significant pelvic bleeding. When conventional catheter-based pelvic angiography is performed, whether before or after MDCT, injection of the bilateral internal iliac veins and the bilateral external iliac veins should be performed.
Lower extremity CTA is the diagnostic study of choice for noninvasive evaluation of lower extremity vascular trauma. Any patient with hard signs of vascular trauma, including active hemorrhage, an expanding or pulsatile hematoma, a wound with bruit or thrill, a distal pulse deficit, or distal ischemic changes, should undergo CTA, unless emergency surgical intervention is necessary. Even those with lower extremity injuries, without hard signs of vascular injury, may still benefit from lower extremity CTA if the ankle-brachial index (ABI) is reduced below 0.9 (sensitivity 87–100%, specificity 80–100%). For those with ABI above 0.9, the likelihood of vascular injury requiring surgery is low, though these patients may still be observed with serial exams for 24–48 hours. One important protocol issue with lower extremity CTA on newer ultrafast scanners is that the scan may “outrun” the contrast bolus in the distal lower extremity, particularly in patients with lower cardiac output. For this reason, at my institution, our lower extremity CTA protocol includes the arterial phase from abdomen or pelvis through the toes, followed 7 seconds later by an immediate delayed (late arterial) phase from knees to toes. Inclusion of both lower extremities in the reconstructed FOV is helpful, even if the injury is unilateral, to provide internal comparison.
Upper extremity CTA is less commonly performed and is more variable in technique. If the upper extremity CTA is performed in isolation, the arm may be positioned above the patient’s head to improve image quality and reduce radiation dose, as long as the patient’s injuries permit such positioning, whereas if the CTA is performed concurrent with a chest CTA, the arm can be positioned at the patient’s side. Always consider contrast injection contralateral to the injured arm to avoid a nondiagnostic scan because of venous extravasation or extensive streak artifact. If both upper extremities require evaluation, a central line should be used for contrast injection. Furthermore, the suspected location of vascular injury may affect the scan range. Proximal injuries, such as those from scapulothoracic dissociation, may only require evaluation of the upper arm to the level of the elbow. If more distal evaluation of the forearm, hand, or fingers is required, some advocate a two-part CTA protocol with different energies and fields of view (100–120kV for aortic arch to elbow, small field of view and 80–100kV for elbow to fingertips, when elbow is positioned above the head).
CTA findings of vascular injury in the upper and lower extremities requiring intervention include vascular occlusion, dissection, extravasation, transection, pseudoaneurysm, and arteriovenous fistula. Differential considerations include preexisting peripheral arterial atherosclerotic disease, nonocclusive vascular spasm, extrinsic compression from adjacent bone fragments or compartment syndrome, or acute embolic occlusion, such as from proximal aortic injury.
Vascular trauma requires prompt recognition and appropriate treatment to prevent significant mortality or morbidity. Today, MDCT is by far the most common technique by which these injuries are diagnosed following trauma. Since vascular injuries are uncommon, many radiologists might not feel adept at imaging them, recognizing them, and characterizing them. It is imperative that an arterial phase MDCT protocol be developed for use in high-risk patients, that intravenous contrast be used in all cases, and that suspicious imaging findings are conveyed to the trauma team appropriately and urgently. These patients may benefit from referral to a level 1 trauma center for definitive treatment.
The opinions expressed in InPractice magazine are those of the author(s); they do not necessarily reflect the viewpoint or position of the editors, reviewers, or publisher.
Senior Faculty, Icahn School of Medicine Mount Sinai Hospital
Leonie Gordon
Vice Chair of Education Professor of Radiology and Nuclear Medicine Medical University of South Carolina
Don C. Yoo
Professor of Diagnostic Imaging, Clinician Educator, Warren Alpert Medical School of Brown University Director of Nuclear Medicine, Miriam Hospital
Esma Akin
Associate Professor of Radiology, Chief of Division of Nuclear Medicine George Washington University Medical Center
Katherine Zukotynski
Departments of Medicine and Radiology McMaster University
Published March 23, 2020
Theranostics is an exciting, emerging field where imaging and therapy are intimately tied together, as the same chemical molecule is used for imaging and therapy with a different radionuclide. Theranostics is a relatively new word, but radioiodine ablation with Iodine-131 (I-131) for imaging and treatment of thyroid cancer is one of the earliest examples of theranostics. Although radioactive iodine (RAI) has been used in clinical practice since the second world war, advances in imaging, therapeutic agents, and our understanding of the molecular basis of disease has slowly led to change. Developing a standardized approach to the management of thyroid cancer has been challenging and controversial. The American Thyroid Association (ATA) and the National Comprehensive Cancer Network (NCCN) guidelines for the management of thyroid cancer have undergone several iterations, some of which have been controversial, as conclusive data on the use of imaging and management strategies is often limited. Recently, however, discussion has led to the publication of the Martinique principles—setting the stage for increased interdisciplinary communication in an attempt to establish a set of recommendations based on the existing data and our wealth of accumulated experience.
Today, when a patient is diagnosed with thyroid cancer, they are assessed clinically and stratified as low, intermediate, or high risk. This is commonly done using the ATA or NCCN risk stratification system for recurrence, and the American Joint Committee on Cancer staging system predictions for long-term outcomes. Based on the risk stratification, a management plan is devised. Often, this includes pre-therapy imaging, therapy, and post-therapy imaging. There is a host of imaging that may be performed before and after therapy. Most commonly, ultrasound, CT, and planar imaging using RAI are performed with a gamma camera; however, SPECT and SPECT combined with CT may improve the sensitivity and specificity for the detection of disease compared with planar imaging alone. PET either with CT or MRI has a role for patients suspected of having thyroid cancer recurrence with rising thyroglobulin and negative diagnostic thyroid scans. There are several radioactive agents to choose from for imaging purposes. Although I-131 is typically less expensive and may be used for both imaging and therapy, Iodine-123 (I-123) has better imaging characteristics and is often used for diagnostic scans, especially in low- or intermediate-risk patients; however, I-123 cannot be used for therapy. Additionally, I-124 (although not routinely clinically used) and 18F-FDG may be helpful in certain situations, but they require access to PET.
ARRS Quick Bytes is a member-exclusive benefit that provides 20-minute videos on emerging topics in radiology for CME on the go.
There are several controversies regarding imaging and therapy of patients with thyroid cancer. For example, there is significant debate about the need for imaging before and after therapy using radioactive iodine vs other modalities, such as ultrasound and CT. Also, whereas RAI therapy has been a mainstay in thyroid cancer treatment for years, there have been recent changes in how this therapy is done. Historically, the amount of RAI to give for the purposes of therapy was based on disease extent, so those patients with distant metastases received a higher empiric amount of RAI than those with localized disease. Furthermore, pediatric patients were treated similarly to adults. Recently, however, we have tried to lower the amount of RAI administered based on the patient’s risk stratification and age to improve long term outcome and minimize radiation exposure, where possible. There is also a recognition that pediatric patients have some different clinical issues; thus, their approach to RAI has some key differences compared to adults.
While the use of and approach to theranostics in thyroid cancer is evolving, a few constants remain. Specifi cally, the experience and expertise of the multidisciplinary care team, as well as the desires of the patient, all need to be considered when making decisions about how to proceed. Imagers must recognize which information they can glean from their scans will best assist in determining the optimal course for treatment. When deciding on therapy, the amount and type of therapy to be given is based not only on pathology, but also on the medical team and patient’s wishes for short- and long-term follow up. Ultimately, as physicians who are intimately involved with both imaging and therapy, our insight can offer a lot to the overall care of the patient with thyroid cancer.
With technological innovation teeming and networks more globalized than ever before, teleradiology—an imaging practice heavily reliant upon both—would seem uniquely moored to rise with these two tides. As advancement often begets acquisition, and especially as radiology practices continue to amalgamate, some teleradiology experts are floating a seemingly counterintuitive notion: the teleradiology wave could be starting to crest.
To be sure, long-distance diagnosis has been dialed into imaging for some 30 years.
In late 1991, University of Kansas researcher Arch W. Templeton boasted in AJRthat more than 1,000 cases had been “digitized, transmitted, and printed on our teleradiology system.” Four years later for the journal, Douglas R. DeCorato detailed his after-dark developments, writing “all radiologic studies performed at Roosevelt Hospital between the hours of midnight and 8 A.M. were digitized and then transmitted over a T1 fiberoptic link to the radiology department of St. Luke’s Hospital, 4.8 kilometers away.”
Transmissions from individual institutions coming in loud and clear, by the summer of 2005, David B. Larson and colleagues had painted the first “comprehensive portrait of teleradiology in radiology practices.” Based upon a 66% response rate from the 970 practices that the American College of Radiology (ACR) surveyed in 1999, as Larson confirmed in AJR, “Seventy-one percent of multiradiologist practices had teleradiology systems in place, using them to interpret 5% of their studies. For solo practices, corresponding statistics were 30% and 14%.”
Flash forward a scant two years, when Todd L. Ebbert sought to capture and communicate just how big teleradiology had become. His 2007 AJR web exclusive had two distinct objectives: “to describe in detail the use of teleradiology in 2003 and to report on changes since 1999 in this rapidly evolving field.” Armed with the ACR’s Survey of Radiologists from 2003 (sent by mail, ironically), as well as its 1999 Survey of Practices, Ebbert et al. verified that 67% of radiology practices in the United States, “which included 78% of all U.S. radiologists,” had performed teleradiology.
Almost a decade’s worth of telemedicine would be performed until the first nationally representative approximation of telehealth practices across all medical specialties was published.
Now, nearly three decades removed from AJR’s initial frontline reporting, is the teleradiology revolution running out of steam? Well, the business answer at least is rather hazy.
Echoing Muroff’s sentiments, Elizabeth Krupinski, professor and vice chair for research in the department of radiology and imaging sciences at Emory University School of Medicine, indicated a swing allied with artificial intelligence, saying “I can see teleradiology changing as a byproduct of the overall industry shift.”
And already, less than a year later, as Muroff told Diagnostic Imaging this past September: “Teleradiology is a saturated, mature market that is no longer growing.” “If anything,” he compounded, “it’s shrinking somewhat because, as practices get larger, they have a greater capability of providing comprehensive call themselves.”
Corporate takeovers of independent radiology practices—what AuntMinnie flagged as one of 2019’s “biggest threats to radiology”—has taken on teleradiology, too. As reporters Brian Casey and Erik Ridley explained: “The rise of corporate radiology companies— which have oftentimes grown by acquiring smaller groups— is turning many radiologists from entrepreneurs into employees. Meanwhile, hospitals continue to expand by swallowing up outpatient centers that once operated independently.”
The bigger they are, the louder their call, indeed.
A newer survey from KPMG tapped 330 corporate, private equity, and investment banking executives in the life science industries for their two cents. And what did this “Big Four” accounting organization find? This year, the health care sector will endure even more absorption than it did in 2019.
Of course, suffering the slings and arrows of more and more mergers and acquisitions (M&A) doesn’t have to necessarily stymie teleradiology’s forward march.
In this past November’s issue of AJR, Michael A. Bruno and team pointed out that “in a fully integrated practice model a single group of subspecialist radiologists would provide care seamlessly at all practice sites, either on a rotational basis or by sharing cases through teleradiology or shared PACS systems, across the full spectrum of care.”
Ultimately, the truest test of tele-harmony writ large depends on who, how, and expressly where you ask.
In light of all the headline legislation, relentless coups, and exaggerated projections implying the demise of enterprise teleradiology, according to its most thorough clinical evaluation to date, the actual practice of teleradiology is very much alive and well.
Defining teleradiology as “the interpretation of medical imaging examinations at a separate facility from where said examination was performed,” Rosenkrantz and his colleagues solicited responses (appropriately enough, via email) from a random sample of 936 ACR members. While a clear majority, 731 respondents, designated their main work setting as non-teleradiology, 85.6% of that cohort indicated they had practiced teleradiology within the past 10 years. Furthermore, 25.4% stated teleradiology comprised a majority of their annual imaging volumes.
No longer the realm of nighthawks, a staggering 91.3% of respondents said that they had implemented teleradiology during normal business hours, while 44.5% to 79.6% said they had implemented teleradiology over evening, overnight, and weekend shifts.
In rural areas, 46.2% of American radiologists surveyed by Rosenkrantz reported performing teleradiology, and 37.2% reported performing teleradiology in critical access hospitals.
Helping working radiologists realize after-hours success and expand coverage for underserved patients, “despite historic concerns,” Rosenkrantz reassured, “teleradiology is widespread throughout modern radiology practice.”
Like most cutting-edge revolutions, efficacious telemedicine continues to spread.
A pilot study just published ahead-of-print in AJRby a team from Germany’s second-largest city and Europe’s third-largest port, Hamburg, christened the concept of maritime telemedicine with the inauguration of a PACS-centered service staffed 24/7 by specialized radiologists at a tertiary hospital on shore.
So, what do the trade machinations of tomorrow or the next administration’s regulations portend for teleradiology’s next wave, the clinical and the commercial?
Fresh out of stealth mode and buttressed by $16.5 million in Series A cash, the mission of this teleradiology company is decidedly asynchronous—assist radiologists in triaging head CT scans through machine learning.
Associate Professor of Clinical Radiology University of California, San Francisco
Published January 10, 2020
The latter half of 2019 saw the identification of an entirely new respiratory illness and introduction of a new diagnosis— e-cigarette, or vaping, product use associated lung injury (EVALI)—into the medical lexicon. Based on current understanding, EVALI is an acute or subacute respiratory illness that is often severe and in some cases fatal, purported to be a chemical pneumonitis resulting from inhalation of one or more toxic substances. As of November 5, the Centers for Disease Control (CDC) reports 2,051 cases of EVALI in the United States, including every state except Alaska, plus the District of Columbia, with 39 deaths in 24 states. The reported cases have affected patients as young as 13 and as old as 75, with males about twice as common as females. Nearly all of these patients have presented with acute or subacute respiratory symptoms, and not surprisingly, thoracic imaging (chest radiograph and CT) has become central to the diagnosis of these patients. Knowledge about these diseases is changing every day, and it is imperative that diagnostic radiologists have a general understanding of what vaping is, how it looks on imaging, and what we know so far.
What Is an E-cigarette, and What Is Vaping?
An e-cigarette is an electronic device that is designed to simulate traditional smoking. Instead of the combustion of tobacco (or, more recently, marijuana), e-cigarettes heat a substance (usually liquid, oil, or wax) to create a vapor that is inhaled, hence the term “vaping.” E-cigarettes were invented in 2003 and introduced in the U.S. around 2007.
Most devices have three main components: a chamber or cartridge that contains the substance to be heated and vaporized (also referred to as the e-liquid); an atomizer or heating element that vaporizes the substance in the cartridge, so that it can be inhaled; and a battery to power the heating source.
Like any piece of technology, these devices have evolved and are becoming increasingly sophisticated. Early devices tended to mimic tobacco cigarettes in shape, but now admittedly look outdated, compared to the most recent generations of products that are smaller, sleeker, and more easily concealed. Some bear more of a resemblance to USB thumb drives than smoking devices, and a few even have Bluetooth connectivity to track how much one vapes.
What Substances Do People Vape?
The substances that patients vape are almost limitless, and this variability is one of the main reasons it has been so difficult to pinpoint the recent surge of cases on one specific cause. A majority of recent cases have been associated with vaping tetrahydrocannabinol (THC) or other marijuana derivatives, and there is mounting evidence for vitamin E acetate as one of the main culprits. While a majority of patients with EVALI report vaping both THC and nicotine, some patients report vaping exclusively nicotine, so it is possibly not just the vitamin E acetate, a thickening agent in THC-containing vaping products, that is to blame.
Given the lack of regulation, there are many ways that the substances are stored, filled, and refilled, as well as many suppliers from where patients get their vaping substances. Some are refillable cartridges, whereas others are disposable pods; some people create their own “home brews” or buy products off the street that may not be sterile and may be adulterated. It is suspected that some of the aftermarket or “off-the-street” products may be more likely to cause injury, and the CDC advises against their use.
Nicotine is often mixed with flavoring agents or “vape juice,” and there are more than 15,000 different flavors. Adults may prefer more traditional flavors such as tobacco, mint, or menthol that try to mimic the taste of cigarette smoking. But other flavors more unabashedly appeal to teenagers and adolescents—many of whom were never smokers prior to vaping. A study in Pediatrics found that adolescents who vaped these nontraditional flavors (including fruit, candy, sweet or dessert, buttery, or other blends not including traditional flavors) were more likely to continue vaping at six months and take more puffs per occasion. The use of these flavors resulted in greater self-reported addiction and satisfaction in another study of young adults.
Why the Recent Rise in Vaping Cases?
Vaping is becoming more popular, particularly among adolescents and young adults; the variety of substances that can be consumed has expanded; and e-cigarette companies have increased the marketing of their products, just to name a few. But with the constant media coverage, everyone thinks about vaping as a cause of lung injury, and that has led to increased recognition by physicians, including radiologists.
The first case we suspected to be lung disease due to vaping was in 2014 in an adult male patient with ground-glass opacity (GGO) on CT, although this case could never be proven, as there wasn’t even a name for this disease then. The first confirmed case we saw was in 2017 in a female who was vaping THC to help her sleep. Case reports of EVALI date back to 2012, but our original article in AJR, “Imaging Findings of Vaping-Associated Lung Injury,” is the first to review and present all of the different imaging patterns that we have encountered so far. The varied appearances underscore the confusion and difficulty in these cases, arguing in support of a multifactorial cause.
What’s the Bare Minimum Radiologists Should Know About Vaping and EVALI?
using an e-cigarette or dabbing (i.e., heating concentrated cannabis oil or wax and inhaling the vapors) in 90 days prior to symptom onset
abnormalities on either chest radiograph or CT
negative infectious workup
no alternative plausible diagnosis (e.g., cardiac, rheumatalogic, or neoplastic)
A probable case of EVALI is similar—the one distinction that either an infection was detected by culture or polymerase chain reaction but not suspected of being the sole cause of lung injury, or minimum testing to exclude infection was not performed.
Imaging is part of the case definition, and as such, radiologists are critical to the diagnosis. The CDC definition verbatim is “pulmonary infiltrate, such as opacities, on plain film chest radiograph or ground-glass opacities on chest CT,” but to any radiologist, this sounds like a vague and generic explanation. One can review the many different patterns of lung injury in our AJR paper, but the one commonality from the cases we present, our review of the literature, and those cases we’ve encountered since is that these patients almost universally present with bilateral opacities that look like acute lung injury and/or organizing pneumonia. Cases may be diffuse, upper- or lower-lobe predominant.
In the appropriate clinical setting, it is arguable that chest radiograph should be sufficient for the diagnosis if bilateral opacities are present, although clinicians often order CT to evaluate for alternative causes, such as pulmonary embolism. Patients who present with acute illness may require ventilatory support. Unfortunately, some patients have died. Treatment with corticosteroids seems to be effective. While most patients completely heal, there is little data on the long-term appearance of survivors of EVALI.
Lung Pathology of EVALI
For most cases of EVALI, obtaining lung tissue is unnecessary for establishing the diagnosis, although some literature on pathology now exists. The largest series of cases where pathologic specimens were available was recently published and concluded that EVALI is “a form of airway-centered chemical pneumonitis from one or more toxic substances” in the aerosolized vapor. The presence of lipid-laden macrophages and positive oil-red-O stains has raised the possibility of exogenous lipoid pneumonia due to vitamin E acetate. Regardless of the underlying pathology, macroscopic fat has not been observed on CT imaging.
What Does the Future Hold for Vaping and EVALI?
The recent illnesses and deaths represent a grave tragedy and public health crisis with little precedent. However, it is important for physicians to not focus exclusively on the negative press. Lost in the daily media shuffle is the fact that for some patients, e-cigarettes may be an effective tool for smoking cessation. Recent data from a randomized control trial in the United Kingdom found that nicotine-containing e-cigarettes were almost twice as effective for smoking cessation at one year compared to other forms of nicotine replacement, but the abstinence rate was still only 18.0% (vs 9.9%). Patients in the e-cigarette group also experienced a greater reduction in cough and phlegm production. On the other hand, can patients stop vaping once they have started? At the one-year mark, 80% of the patients using e-cigarettes in this group were still using. Many health officials are worried that it is just replacing one habit (smoking) with a slightly less bad habit (vaping). Vaping might be safer than traditional cigarette use, but at this point, we just don’t know.
Hopefully, the agents responsible for cases of EVALI will be discovered, and cases of acute lung injury will subside, but what are the long-term effects of vaping? Vaping is a new practice that has been around for barely more than a decade, and it is certainly a worry that long-term vaping could lead to more chronic lung disease and fibrosis. We have seen some indication that vaping may progress to fibrosis, although this realm is largely unexplored and ripe for imaging research.
The opinions expressed in InPractice magazine are those of the author(s); they do not necessarily reflect the viewpoint or position of the editors, reviewers, or publisher.
Writing on MRS in the October 2004 issue of the Journal ofNeuroscience, Matthew K. Belmonte from the Autism Research Centre at the University of Cambridge duly noted: “It has been said that people with autism suffer from a lack of ‘central coherence,’ the cognitive ability to bind together a jumble of separate features into a single, coherent object or concept. Ironically, the same can be said of the field of autism research, which all too often seems a fragmented tapestry stitched from differing analytical threads and theoretical patterns”.
Fifteen years removed, while ASD remains very much an heterogeneous disorder of multifactorial etiology, evidencing an expansive range of symptoms and severities alike, radiology is in the process of reconciling so many image threads. True, bereft of a priori behavioral phenotyping (e.g., Autism Diagnostic Observation Schedule [ADOS], Social Responsiveness Scale, Kaufman Brief Intelligence Test, composite IQ score), right now, radiology alone still cannot definitively diagnose ASD in anyone, child or adult. There is good news, though. The radiology research paradigm is shifting—away from mere aberration identification to clinical diagnosis.
The sands underneath it all first loosened in 2014, when University of Pittsburgh and Carnegie Mellon researchers utilized machine-learning algorithms to grade 34 young adults as either autistic or control with > 97% accuracy based upon fMRI neurocognitive markers for eight social interaction verbs: compliment, insult, adore, hate, hug, kick, encourage, and humiliate. Moving quickly, one year later, Virginia Tech Carilion Research Institute professor P. Read Montague synthesized nine years’ worth of previous trials to announce in ClinicalPsychological Science that his team had developed an even more efficient technique to diagnose children with ASD in under two minutes: single-stimulus fMRI. Subjects were shown 15 images of themselves and 15 images of another child, matched according to age and gender, for four seconds per image in randomized order. Like the control adults in Montague’s earlier experiments with imaging for ASD, when viewing their own pictures, the control children had a high response in the middle cingulate cortex; by contrast, children with ASD showed an appreciably diminished reaction. Notably, Montague et al. could detect this disparity using one, solitary image.
This May, much of Montague’s same colleagues, including principal investigator, Kenneth Kishida of the Wake Forest School of Medicine, made headlines for a Biological Psychology article demonstrating that a single stimulus and < 30 seconds of fMRI data were sufficient to differentiate ASD children from their typically developing (TD) peers. To test a hypothesis that responsiveness of the brain’s ventral medial prefrontal cortex (vmPFC) in children diagnosed with ASD is diminished for visual cues, denoting high-value social interaction, 40 participants (of which 12 had ASD and 28 were TD), aged 6–18 years old, were prompted to observe images of four faces and four objects, which were projected onto a screen and viewed through a mirror during fMRI scanning. With each image characterized as favorite, pleasant, neutral, or unpleasant, the favorite images depicted each of the participants’ self-selected favored face and object, and the remaining images were selected from the International Affective Picture System (IAPS) database. Each of the eight images was then displayed only once for five seconds during a block that repeated six times. Following the completion of 12- to 15-minute MRI scans, participants were shown the identical set of images on a computer screen, ranking them in order, from pleasant to unpleasant, with a self-assessing sliding scale. Results showed that the average response of vmPFC was significantly lower in the ASD cohort, compared to the TD cohort.
“How the brain responded to these pictures is consistent with our hypothesis that the brains of children with autism do not encode the value of social exchange in the same way as typically developing children,” Kishida said in a prepared statement. “Based on our study,” he continued, “we envision a test for autism in which a child could simply get into a scanner, be shown a set of pictures, and within 30 seconds, have an objective measurement that indicates if their brain responds to social stimulus and non-social stimuli.”
There are limitations here. Because these 40 children were permitted to specify favored objects and people, reasonably assuming that there were distinct visual differences between these non-IAPS images and that canonical cache, Kishida conceded the possibility that at least some of the reported response differential could simply be due to known vs. novel. Moreover, since ASD disproportionately affects male patients—i.e., four times more common among boys than girls, the CDC maintains—he acknowledged an optimal design could be updated to investigate the gender divide between the ASD and the TD children more thoroughly.
“Based on our study, we envision a test for autism in which a child could simply get into a scanner, be shown a set of pictures, and within 30 seconds, have an objective measurement that indicates if their brain responds to social stimulus and non-social stimuli.”
—Kenneth Kishida
Another Wake Forest faculty member, Christopher T. Whitlow, has been presenting related research on ASD imaging since 2014. As his studies have surveyed patterns of joint variability in severely preterm infants, might we see an eventual diagnostic environment where Whitlow’s voxel-based morphometry informs Kishida and Montague’s single-stimulus exemplar to evidence brain dysfunction in patients younger than the age-six threshold?
Although reproductive stoppage (i.e., the tendency for arrested propagation after diagnosis of an affected child) can lead to underestimates of sibling recurrence risk for ASD, with ascertainment biases and overreporting often pointing to its inflation, we should focus on the family first. In 2011, the multisite international network, Baby Siblings Research Consortium, conducted a prospective longitudinal study of 664 infants who had an older biological sibling with ASD, monitoring them from early life to 36 months, when they were classified as having or not having ASD—an ASD taxonomy requiring exceeding the ADOS cut-off, as well as an expert’s diagnosis. In total, 18.7% of infants developed ASD. Whereas infant age at enrollment, gender and functioning level of the infant’s older sibling, and other demographic circumstances did not predict ASD outcome, infant gender and the presence of > 1 older affected sibling were significant forecasters. Again, there was a nearly threefold risk escalation for male subjects and an additional twofold increase in risk if there was > 1 older affected sibling.
Family history, meet deep learning. Recent findings published in Science Translational Medicine by University of North Carolina at Chapel Hill researchers revealed that when applied to functional connectivity MRI (fcMRI) data at six months of age in infants with high familial risk for ASD, a nested, cross-validated machine-learning algorithm predicted an ASD diagnosis with > 96% accuracy at 24 months. Citing severalbrainvariances—both morphological and electro-physiological—members of his team had documented as early as six months in infants later diagnosed with ASD, “Given the complexity and heterogeneity of ASD,” lead author Robert W. Emerson surmised, “methods for the early detection of ASD using brain metrics will likely require information that is multivariate, complex, and developmentally sensitive.” Apropos, Emerson et al. employed an array of 230 regions of interest (ROI) previously defined across the entire brain to create functional connectivity matrices from the fMRI scans of 59 at-risk infants (11 diagnosed with ASD at 24 months, 48 who did not have ASD at 24 months) during natural sleep without sedation at their six-month visit. “Our logic was that these regions would be the most likely to contribute to the discrimination between groups in the 59 separate support vector machine models,” wrote Emerson. With data collection resulting in 26,335 usable ROI pairs exemplifying each infant’s whole-brain functional constitution by training MATLAB’s Statistics and Machine Learning Toolbox (Mathworks, Inc.) to ascertain the causal patterns of individual separation, the probability that infants with a positive classification truly had ASD (positive predictive value) at 24 months was 100% (95% CI, 62.9–100). Negative predictive value at 24 months was 96% (95% CI, 85.1–99.3).
A first-of-its-kind study from November 2018 that leveraged the imaging archive of Geisinger Health System in Danville, Pennsylvania, takes us back to the future, examining early brain markers in ASD to further the promise of artificial intelligence for earlier detection. Renewing his dissertation research, Gajendra J. Katuwal and colleagues applied random forest ensemble learning to models trained on 687 brain features of Freesurfer v 5.3.0 (Martinos Center for Biomedical Imaging) to compare cortical and sub-cortical morphometric features for ASD vs. non-ASD classification. Their query of head MR images from Geisinger’s institutional tranche, after removing those with artifacts, motion, lesions, abnormally large ventricles, and neurodevelopmental disorders as identified by International Classification of Diseases code, yielded 112 non-ASD and 115 ASD subjects. Eschewing gender confounds, 20 non-ASD and 34 ASD scans of female subjects were excluded. Although total intracranial volume (TIV) of ASD measured 5.5% larger than the control, brain volumes of other ROI, when calculated as TIV percentage, measured smaller in ASD—partially due to larger (> 10%) ventricles in ASD. ASD’s larger TIV exhibited correlates with greater surface area and aggregate cortical folding, yet not with cortical thickness. ASD frontal and temporal white-matter tracts evidenced less image intensity, seemingly suggesting myelination deficit. Ultimately, Katuwal’s methodology was able to achieve 95% AUC for ASD vs. non-ASD classification using all brain features. When stochastic discrimination was discrete for each feature type, image intensity yielded the highest predictive power (95% AUC), followed by cortical folding index (69%), cortical and subcortical volume (69%), and surface area (68%).
According to Katuwal, “the most important classification feature was white matter intensity surrounding the rostral middle frontal gyrus,” which measured lower (d = 0.77, p = 0.04) in ASD.
Because medical technology also rises, medical imaging, itself, is sure to manifest a more prominent role over time among allied sciences with regards to forthcoming ASD diagnoses and concomitant, personalized care. To that end, in order to fully apprehend the neuroanatomical foundations of ASD, a comprehensive, multimodal surveillance of early brain alterations would seem to light the best forward path. Progress isn’t always a straight line, of course, so radiology has places yet to go, indeed.
The opinions expressed in InPractice magazine are those of the author(s); they do not necessarily reflect the viewpoint or position of the editors, reviewers, or publisher.
When Thomas H. Berquist logs off his iPad this summer, his 12-year tenure as the 12th editor in chief of the American Journal of Roentgenology (AJR) will capstone a period of unprecedented growth for the 113-year-old publication. Truly the end of an era, only two other men will have occupied the AJR’s chief chair longer than Berquist: Lawrence Reynolds, who picked up the mantle in 1930 and died in office 31 years later, and his immediate successor, Traian Leucutia, whose editorship (1961–75) lasted just two years longer than this lauded, yet humble Mayo Clinic radiologist.
Articulating an expansive vision for radiology’s beloved “yellow journal” from his first days at the desk in late 2008, the ARRS Publications Committee, to whom Berquist has reported for some 150 issues of AJR, agrees that he’s fostered a unique editorial climate ever since—one of both exacting rigor and earnest diversity.
“Dr. Berquist is one of the most inclusive leaders that I have ever had the pleasure to work with,” says Deborah Baumgarten, ARRS Publications Committee chair. “He solicits opinions and really listens to and considers what others have said. You feel like you matter to him.”
Acknowledging achievement only if it’s data-borne, the internationally recognized author of 38 books on medical imaging tells InPractice he remains very much “a numbers guy.”
Incidentally, despite ARRS’ announcement of Berquist’s retirement more than 15 months ago, submissions to the journal continue to pour in unabated. And although AJR’s acceptance rate “still hovers right around 20%,” Berquist does admit that the overall quality of the articles being submitted is likely as good as it has ever been.
For once, there’s a causation implied by the correlation.
Two years ago, Berquist himself took to these same pages to reassert AJR’s raison d’être in two words: “evidence-based articles”. Codifying the journal’s reporting guidelines for Original Research and Review articles “to assist authors in providing optimal consistent content,” he also detailed significant revisions to the Standards for Reporting Diagnostic Accuracy Studies (STARD) and Standards for Reporting Observation Studies in Epidemiology (STROBE) checklists, “in an effort to provide more imaging-friendly guidelines.”
With humility and precision, Berquist now notes, “to date, there have been 128 STARD submissions and 33 STROBE submissions to AJR,” casually mentioning the “significant recruitment initiative” he’s spearheading to further improve these types of content enhancement, which he’s keen to note will soldier on even without him aboard.
Circulation is up, too. With AJR enjoying record readership worldwide—especially unique views and clicks, online and mobile, at AJRonline.org—as the outgoing editor in chief wrote in his “Things We Learned Along the Way” editorial from November, “the online version is the journal of record.”
It was Berquist’s predecessor, Robert Stanley, who introduced the notion of electronic article submission. Not surprisingly, his institution of web-based submissions yielded a marked increase in international authors submitting to AJR, sending Stanley and staff scrambling to enlist foreign-language reviewers. Sixteen years post-Stanley, Berquist recalls yet another telling audit.
“Currently, there are 2,321 total AJR reviewers,” he says. “Eighty percent hail from the United States, and 20% are based internationally.”
Asked how a more international distribution due to ever-increasing scholar globalization might alter the scope of AJR content to come, thus far, Berquist says he’s seen only one year, 2014, where foreign submissions outpaced submissions from the U.S. As always, Berquist’s bias tends toward scientific scrutiny, not identity politics, “particularly where a benign cultural difference could escalate to the level of significant medicolegal dilemma.”
Ultimately, he’d much rather talk residents and reviewers than matters foreign and domestic. Regarding residents, Berquist is quick to credit Howard P. Forman for initiating the journal’s Trainee Reviewer program, pointing out the present group of “61 trainee reviewers and additional new reviewers who have mentors as they begin their reviewer role.”
Naturally, Berquist has streamlined the onboarding process; it began two years ago at the Marriott Wardman Park Hotel in Washington, D.C.
In 2018, alongside Cheryl S. Merrill, ARRS’ director of publications, Berquist debuted what would eventually become an essential component of the ARRS Annual Meeting that speaks volumes about the significance of scientific integrity—a two hour course the duo affectionately dubbed “Rock the Review: How to Get a Perfect Score.” A perfect score equals 4.0, sure, but what does said “rocking it” actually look like on the page?
For that inquiry, Berquist settles in: “The review must include sophisticated, detailed comments to authors with line and page referencing to enhance the content and relevance of the work; concise, confidential comments to the editor; and the reviews must be completed in the allotted 14 days or earlier.”
As for his active reviewers, again, Berquist knows their numbers by heart.
“Ten percent of AJR reviewers have scores less than 3.0, and 40% have scores between 3.0–3.5,” he says. “Half of the reviewers for AJR, nearly 1,200, have a perfect 4.0,” Berquist half-beams, adding that each and every reviewer is evaluated at least yearly, “more frequently should their approach warrant it.” He reveals “any reviewer may request a review of themselves at any time,” too
Resigned to the inherent difficulty of “consistent communication” with more than 2,300 reviewers across the globe, Berquist’s not bereft of procedure here either.
“There are multiple data points available each day,” he says of his quality control, “including how many invitations a reviewer has received, accepted, and declined; how many times a reviewer has been uninvited for not responding; the last review accepted and completed and the last review declined; reviews in progress; and the mean reviewer score.”
Lest you think he’s all scores and no play, there are prizes— albeit hard-won ones.
Known for penning personal letters to stellar reviewers, Berquist also established “a Distinguished Reviewers category for individuals performing 10 or more reviews in a given year with score of 3.0 or higher.” Reviewers’ names are featured on the AJR masthead for the entirety of the following year, and their departmental chairs are notified of the distinction.
For “above and beyond assistance,” states Berquist, “we initiated the Gold and Silver Distinguished Reviewer Achievement Awards in 2018 for reviewers with 100 or more reviews and 50–99 reviews, respectively.” These reviews must be scored at least 3.5, and Berquist remarks that 91 AJR reviewers have received 14 gold and 77 silver trophies during the Reviewer’s Luncheon at the ARRS Annual Meeting.
“We now have Platinum and Diamond Distinguished Reviewer Achievement Awards for scores 3.5 or higher for 150– 199 reviews and 200 or more reviews, respectively,” he says with that muted smile returning.
Acknowledging that the whole notion of peer review itself is in flux, Berquist’s not averse to the creep of new ideas. His interest in zeitgeist systems thinking, like Just Culture, has been abiding, and he confesses to “a certain anticipation” for an updated model of shared accountability.
Never not teaching, the diagnostic radiologist is wont for a metaphor.
“Peer review is a lot like the Supreme Court,” Berquist claims. “It’s by no means perfect, but it’s the best we have now.”
For all reviewers, authors, and journal staff, time is always of the essence. Recalling an AJR authors’ survey “where 85% of respondents considered speed to publication extremely important,” once more, the numbers stay on Berquist’s side.
“In 2013, the time to first decision was 37.6 days,” he says. Streamlined protocols in 2017, implemented by the journal’s inhouse staff, shrunk that time down to 25 days. For 2019, Berquist tallies “the average time to first decision is 18.8 days,” compared to 20 days at the same time the previous year.
Prior to recent concerted efforts, he laments that the elapsed time from first decision to AJR publication measured a “protracted 147 days.” Understandably, Berquist happily reports that number has been cut almost in half.
Asked who or what is most responsible for this optimized circle to publication, AJR’s chief editorial officer neither hesitates nor equivocates: “The journal staff deserve the credit—all of it.”
Berquist balks at the term officer. His self-effacing streak matched only by his work ethic, this self-described “policeman with no gun” has doggedly pursued often competing commitments at AJR—everything from article enhancements, reviewer recognition, and production improvements to that ever-important journal impact factor.
And at least until June, armed or not, the buck still stops with Berquist.
“Right now, the impact score of AJR is 3.161,” he rattles off to the third decimal. “That’s the highest the score has ever been. But it’s not yet a 4, so in that regard,” he doesn’t pause, “I’ve failed.”
Impact factor is a scientometric index, yes, but Berquist concedes there are more popular proxies, especially in pixels. It was under his administration that ARRS joined Twitter, posted a DOI on Instagram, and published an AJR video to YouTube. It was during Berquist’s watch that democratized platforms such as these breathed new, sometimes second lives to articles about magnetic eyelashes as MRI artifacts and his personal favorite, radiologic detection of inadvertently ingested wire bristles from a grill-cleaning brush.
“It’s exciting, it’s all visibility, and we’re getting much better at it,” Berquist says of social media exposure. “We’re also exploring more and more things adjacent to it,” his subtle reference to AJR Podcasts, available cross-platform via iTunes and Google Play.
At the end of the day, it’s not impact factor or Just Culture or homepage views that’s kept Berquist up at night these last 12 years.
“For me, the key thing has always been scientific integrity,” he says. “In fact, it’s really only this. But in reality, there are times when I feel alone on this white horse, swinging at windmills.”
Of course, the man of La Mancha never had to traverse this modern landscape of so many open-access journals—an “exponential proliferation” our Jacksonville, Florida physician cites as his chief concern for medical publishing moving forward.
“There were about 80 radiology journals when I began at AJR,” Berquist remembers. Today, he inventories more than 800 open-access journals publishing medical imaging content on a regular basis. How could Berquist alone possibly guard against every duplicative breach?
“I’m not sure we can anymore,” Berquist answers, invoking both pronouns.
According to the editor in chief, AJR uses a huge database, Similarity Check (CrossRef), for manuscript evaluation, as well as “a 10% or greater duplication and singles source greater than or equal to 3% to help assess duplication more thoroughly.” Given that a typical year will end in more than 1,800 submissions to the cue, the journal’s two-factor safeguard invariably yields “a staggering number” of replicate queries.
Noting that medical schools, residency programs, and fellowships lack a proper course in publishing ethics—Berquist’s biggest regret is not advocating harder for a nationwide curriculum—“only 1.8% of the submissions AJR receives contain less than 10% duplication,” he sighs. “We can’t keep kicking this can down the road.”
With nuclear medicine becoming de rigueur in the 1960s, ARRS members of a certain vintage will recall how Traian Leucutia relented to rechristening AJR the American Journal of Roentgenology, Radium Therapy, and Nuclear Medicine. The late Melvin M. Figley not only changed it back in 1976, he kept the word “Roentgenology” in the title, maintaining its myriad historical associations. As radiology enters the third decade of the new millennium, does Berquist see an emerging modality or pressing topic that could necessitate another retitling of the journal?
“Believe it or not, changing the name of AJR has come up,” he says. “Imaging, as a whole, is a lot of different things, and imaging is always evolving, so if the ABR’s requirements change accordingly, I could foresee the ARRS Executive Council maybe taking a vote.”
Berquist has but one simple request for the next editor: “Whatever happens, I just hope we’ll keep on calling AJR the yellow journal.” Having championed the leading resource for practicing physicians and allied health professionals engaged in patient-centered medical imaging, let no one dare call what he’s done “yellow journalism,” though.
The opinions expressed in InPractice magazine are those of the author(s); they do not necessarily reflect the viewpoint or position of the editors, reviewers, or publisher.
