Mixed Supervision of Histopathology Improves Prostate Cancer Classification from MRI.

Non-invasive prostate cancer classification from MRI has the potential to revolutionize patient care by providing early detection of clinically significant disease, but has thus far shown limited positive predictive value. To address this, we present a image-based deep learning method to predict clinically significant prostate cancer from screening MRI in patients that subsequently underwent biopsy with results ranging from benign pathology to the highest grade tumors. Specifically, we demonstrate that mixed supervision via diverse histopathological ground truth improves classification performance despite the cost of reduced concordance with image-based segmentation. Where prior approaches have utilized pathology results as ground truth derived from targeted biopsies and whole-mount prostatectomy to strongly supervise the localization of clinically significant cancer, our approach also utilizes weak supervision signals extracted from nontargeted systematic biopsies with regional localization to improve overall performance. Our key innovation is performing regression by distribution rather than simply by value, enabling use of additional pathology findings traditionally ignored by deep learning strategies. We evaluated our model on a dataset of 973 (testing n = 198) multi-parametric prostate MRI exams collected at UCSF from 2016-2019 followed by MRI/ultrasound fusion (targeted) biopsy and systematic (nontargeted) biopsy of the prostate gland, demonstrating that deep networks trained with mixed supervision of histopathology can feasibly exceed the performance of the Prostate Imaging-Reporting and Data System (PI-RADS) clinical standard for prostate MRI interpretation (71.6% vs 66.7% balanced accuracy and 0.724 vs 0.716 AUC).

A Practical Guide for Paid Family and Medical Leave in Radiology, From the AJR Special Series on DEI.

Paid family and medical leave (FML) has significant benefits to organizations, including improvements in employee recruitment and retention, workplace culture, and employee morale and productivity, and is supported by evidence for overall cost savings. Furthermore, paid FML related to childbirth has significant benefits to individuals and families, including but not limited to improved maternal and infant health outcomes and improved breastfeeding initiation and duration. In the case of nonchildbearing parental leave, paid FML is associated with more equitable long-term division of household labor and childcare. Paid FML is increasingly being recognized as an important issue in medicine, as evidenced by the recent passage of policies by national societies and governing bodies, including the American Board of Medical Specialties, American Board of Radiology, Accreditation Council for Graduate Medical Education (ACGME), American College of Radiology, and American Medical Association. Implementation of paid FML requires adherence to federal, state, and local laws as well as institutional requirements. Specific requirements pertain to trainees from national governing bodies, such as the ACGME and medical specialty boards. Flexibility, work coverage, culture, and finances are additional considerations for ensuring an optimal paid FML policy that accounts for concerns of all impacted individuals.

Career Trajectory Factors Affecting Gender Diversity in Academic Radiology Department Chairs: Results of a Survey of SCARD Members.

In this survey of academic radiology department chairs, pathways to first chair appointment were similar between men and women in terms of prior professional accomplishments and chair position preparedness. However, women more commonly perceived that their gender negatively affected their career trajectory, and they more frequently reported experiencing overt discrimination and unconscious bias.

Federated Learning with Research Prototypes: Application to Multi-Center MRI-based Detection of Prostate Cancer with Diverse Histopathology.

RATIONALE AND OBJECTIVES: Early prostate cancer detection and staging from MRI is extremely challenging for both radiologists and deep learning algorithms, but the potential to learn from large and diverse datasets remains a promising avenue to increase their performance within and across institutions. To enable this for prototype-stage algorithms, where the majority of existing research remains, we introduce a flexible federated learning framework for cross-site training, validation, and evaluation of custom deep learning prostate cancer detection algorithms. MATERIALS AND METHODS: We introduce an abstraction of prostate cancer groundtruth that represents diverse annotation and histopathology data. We maximize use of this groundtruth if and when they are available using UCNet, a custom 3D UNet that enables simultaneous supervision of pixel-wise, region-wise, and gland-wise classification. We leverage these modules to perform cross-site federated training using 1400+ heterogeneous multi-parameteric prostate MRI exams from two University hospitals. RESULTS: We observe a positive result, with significant improvements in cross-site generalization performance with negligible intra-site performance degradation for both lesion segmentation and per-lesion binary classification of clinically-significant prostate cancer. Cross-site lesion segmentation performance intersection-over-union (IoU) improved by 100%, while cross-site lesion classification performance overall accuracy improved by 9.5-14.8%, depending on the optimal checkpoint selected by each site. CONCLUSION: Federated learning can improve the generalization performance of prostate cancer detection models across institutions while protecting patient health information and institution-specific code and data. However, even more data and participating institutions are likely required to improve the absolute performance of prostate cancer classification models. To enable adoption of federated learning with limited re-engineering of federated components, we open-source our FLtools system at https://federated.ucsf.edu, including examples that can be easily adapted to other medical imaging deep learning projects.

American College of Radiology Paid Family/Medical Leave Policy: A Call to Action for the House of Medicine.

The American College of Radiology (ACR) passed a historic paid family/medical leave (PFML) resolution at its April 2022 meeting, resolving that "diagnostic radiology, interventional radiology, radiation oncology, medical physics, and nuclear medicine practices, departments and training programs strive to provide 12 weeks of paid family/medical leave in a 12-month period for its attending physicians, medical physicists, and members in training as needed." The purpose of this article is to share this policy beyond radiology so that it may serve as a call to action for other medical specialties. Such a PFML policy (1) supports physician well-being, which in turn supports patient care; (2) is widely needed across American medical specialties; and (3) should not take nearly a decade to achieve, as it did in radiology, especially given increasing physician burnout and the ongoing COVID-19 pandemic. Supported by information on the step-by-step approach used to achieve radiology-specific leave policies and considering current and normative policies at the national level, this article concludes by reviewing specific strategies that could be applied toward achieving a 12-week PFML policy for all medical specialties.

Utility of Normalized Body Composition Areas, Derived From Outpatient Abdominal CT Using a Fully Automated Deep Learning Method, for Predicting Subsequent Cardiovascular Events.

BACKGROUND. CT-based body composition (BC) measurements have historically been too resource intensive to analyze for widespread use and have lacked robust comparison with traditional weight metrics for predicting cardiovascular risk. OBJECTIVE. The aim of this study was to determine whether BC measurements obtained from routine CT scans by use of a fully automated deep learning algorithm could predict subsequent cardiovascular events independently from weight, BMI, and additional cardiovascular risk factors. METHODS. This retrospective study included 9752 outpatients (5519 women and 4233 men; mean age, 53.2 years; 890 patients self-reported their race as Black and 8862 self-reported their race as White) who underwent routine abdominal CT at a single health system from January 2012 through December 2012 and who were given no major cardiovascular or oncologic diagnosis within 3 months of undergoing CT. Using publicly available code, fully automated deep learning BC analysis was performed at the L3 vertebral body level to determine three BC areas (skeletal muscle area [SMA], visceral fat area [VFA], and subcutaneous fat area [SFA]). Age-, sex-, and race-normalized reference curves were used to generate z scores for the three BC areas. Subsequent myocardial infarction (MI) or stroke was determined from the electronic medical record. Multivariable-adjusted Cox proportional hazards models were used to determine hazard ratios (HRs) for MI or stroke within 5 years after CT for the three BC area z scores, with adjustment for normalized weight, normalized BMI, and additional cardiovascular risk factors (smoking status, diabetes diagnosis, and systolic blood pressure). RESULTS. In multivariable models, age-, race-, and sex-normalized VFA was associated with subsequent MI risk (HR of highest quartile compared with lowest quartile, 1.31 [95% CI, 1.03-1.67], p = .04 for overall effect) and stroke risk (HR of highest compared with lowest quartile, 1.46 [95% CI, 1.07-2.00], p = .04 for overall effect). In multivariable models, normalized SMA, SFA, weight, and BMI were not associated with subsequent MI or stroke risk. CONCLUSION. VFA derived from fully automated and normalized analysis of abdominal CT examinations predicts subsequent MI or stroke in Black and White patients, independent of traditional weight metrics, and should be considered an adjunct to BMI in risk models. CLINICAL IMPACT. Fully automated and normalized BC analysis of abdominal CT has promise to augment traditional cardiovascular risk prediction models.

How to implement paid family and medical leave: A toolkit for practices.

Paid family and medical leave policies are increasingly popular in today's competitive labor market and provide well-documented advantages to all stakeholders. Implementing paid leave for radiologists can seem daunting due to overlapping legal and institutional policies, logistical challenges and call coverage, as well as industry-specific special considerations such as resident education and historical workplace attitudes. This toolkit can empower radiology leaders to implement written paid leave policies in their home institutions and demonstrate that equitable, compassionate institutional policies for paid leave are financially favorable, widely desirable, and increasingly achievable with the right tools in hand.

A Fully Automated Deep Learning Pipeline for Multi-Vertebral Level Quantification and Characterization of Muscle and Adipose Tissue on Chest CT Scans.

Body composition on chest CT scans encompasses a set of important imaging biomarkers. This study developed and validated a fully automated analysis pipeline for multi-vertebral level assessment of muscle and adipose tissue on routine chest CT scans. This study retrospectively trained two convolutional neural networks on 629 chest CT scans from 629 patients (55% women; mean age, 67 years ± 10 [standard deviation]) obtained between 2014 and 2017 prior to lobectomy for primary lung cancer at three institutions. A slice-selection network was developed to identify an axial image at the level of the fifth, eighth, and 10th thoracic vertebral bodies. A segmentation network (U-Net) was trained to segment muscle and adipose tissue on an axial image. Radiologist-guided manual-level selection and segmentation generated ground truth. The authors then assessed the predictive performance of their approach for cross-sectional area (CSA) (in centimeters squared) and attenuation (in Hounsfield units) on an independent test set. For the pipeline, median absolute error and intraclass correlation coefficients for both tissues were 3.6% (interquartile range, 1.3%-7.0%) and 0.959-0.998 for the CSA and 1.0 HU (interquartile range, 0.0-2.0 HU) and 0.95-0.99 for median attenuation. This study demonstrates accurate and reliable fully automated multi-vertebral level quantification and characterization of muscle and adipose tissue on routine chest CT scans. Keywords: Skeletal Muscle, Adipose Tissue, CT, Chest, Body Composition Analysis, Convolutional Neural Network (CNN), Supervised Learning Supplemental material is available for this article. © RSNA, 2022.

US lesion visibility predicts clinically significant upgrade of prostate cancer by systematic biopsy.

PURPOSE: To identify predictors of when systematic biopsy leads to a higher overall prostate cancer grade compared to targeted biopsy. METHODS AND MATERIALS: 918 consecutive patients who underwent prostate MRI followed by MRI/US fusion biopsy and systematic biopsies from January 2015 to November 2019 at a single academic medical center were retrospectively identified. The outcome was upgrade of PCa by systematic biopsy, defined as cases when systematic biopsy led to a Gleason Grade (GG) ≥ 2 and greater than the maximum GG detected by targeted biopsy. Generalized linear regression and conditional logistic regression were used to analyze predictors of upgrade. RESULTS: At the gland level, the presence of an US-visible lesion was associated with decreased upgrade (OR 0.64, 95% CI 0.44-0.93, p = 0.02). At the sextant level, upgrade was more likely to occur through the biopsy of sextants with MRI-visible lesions (OR 2.58, 95% CI 1.87-3.63, p < 0.001), US-visible lesions (OR 1.83, 95% CI 1.14-2.93, p = 0.01), and ipsilateral lesions (OR 3.89, 95% CI 2.36-6.42, p < 0.001). CONCLUSION: Systematic biopsy is less valuable in patients with an US-visible lesion, and more likely to detect upgrades in sextants with imaging abnormalities. An approach that takes additional samples from regions with imaging abnormalities may provide analogous information to systematic biopsy.

The SCARD Fellowship Policy and the Abdominal Imaging Fellowship: A Follow-up Survey After the First Year.

RATIONALE AND OBJECTIVES: To assess resident and fellowship program director (PD) perceptions of the abdominal radiology fellowship application process following the first cycle in which an embargo on interviews until December 1, 2019 was set according to the Society of Chairs of Academic Radiology Departments (SCARD) timeline for the 2021-2022 abdominal imaging fellowship year. MATERIALS AND METHODS: Eligible study participants included fellowship PDs of all abdominal imaging programs in the United States and residents that attended the Society of Abdominal Radiology (SAR) 2020 Annual Meeting. A questionnaire was developed by content and survey experts, pilot tested, and administered from May to June 2020. RESULTS: A total of 39% (36/92) of all PDs and 30% (46/152) of all individuals identified as residents with valid email addresses that attended the SAR 2020 Annual Meeting responded to the survey with an overall response rate of 34%. Only 42% of PDs and 33% of residents supported moving to a match, while 62% of PDs and 70% of residents thought that a match would limit the autonomy of applicants. While most PDs and residents also agreed that the first iteration of the SCARD timeline allowed residents to make a more informed choice, the majority of PDs were dissatisfied with their experience. Most PDs and residents additionally want applications to be accepted no earlier than July and/or August of the R3 year (initial SCARD guidelines did not restrict timing), interviews to begin on November 1st or earlier of the R3 year (compared to December 1st set in the first iteration of the guidelines), and a gap of 2-4 weeks between the date of first interviews and notification of first offers (initial SCARD guidelines did not restrict timing). Lastly, an overwhelming majority of PDs and residents agreed that SAR should enforce the abdominal imaging fellowship application process. CONCLUSION: Following the first cycle of abdominal imaging fellowship applications conducted according to the SCARD guidelines, a majority of trainees and PDs felt the changes were favorable and were opposed to a formal match. Specific suggestions for improvement were elicited from stakeholders and will be incorporated for the next cycle.

Imaging AI in Practice: A Demonstration of Future Workflow Using Integration Standards.

Artificial intelligence (AI) tools are rapidly being developed for radiology and other clinical areas. These tools have the potential to dramatically change clinical practice; however, for these tools to be usable and function as intended, they must be integrated into existing radiology systems. In a collaborative effort between the Radiological Society of North America, radiologists, and imaging-focused vendors, the Imaging AI in Practice (IAIP) demonstrations were developed to show how AI tools can generate, consume, and present results throughout the radiology workflow in a simulated clinical environment. The IAIP demonstrations highlight the critical importance of semantic and interoperability standards, as well as orchestration profiles for successful clinical integration of radiology AI tools. Keywords: Computer Applications-General (Informatics), Technology Assessment © RSNA, 2021.

The Trials and Tribulations of Assembling Large Medical Imaging Datasets for Machine Learning Applications.

With vast interest in machine learning applications, more investigators are proposing to assemble large datasets for machine learning applications. We aim to delineate multiple possible roadblocks to exam retrieval that may present themselves and lead to significant time delays. This HIPAA-compliant, institutional review board-approved, retrospective clinical study required identification and retrieval of all outpatient and emergency patients undergoing abdominal and pelvic computed tomography (CT) at three affiliated hospitals in the year 2012. If a patient had multiple abdominal CT exams, the first exam was selected for retrieval (n=23,186). Our experience in attempting to retrieve 23,186 abdominal CT exams yielded 22,852 valid CT abdomen/pelvis exams and identified four major categories of challenges when retrieving large datasets: cohort selection and processing, retrieving DICOM exam files from PACS, data storage, and non-recoverable failures. The retrieval took 3 months of project time and at minimum 300 person-hours of time between the primary investigator (a radiologist), a data scientist, and a software engineer. Exam selection and retrieval may take significantly longer than planned. We share our experience so that other investigators can anticipate and plan for these challenges. We also hope to help institutions better understand the demands that may be placed on their infrastructure by large-scale medical imaging machine learning projects.

Past, present, and future of abdominal radiology fellowship recruitment.

The authors provide a commentary on the current status of the Abdominal Radiology Fellowship recruitment process, which is not presently governed by a formal Match. Abdominal Radiology is the largest radiology subspecialty fellowship that remains outside of the Match. The Society of Abdominal Radiology convened a task force in 2019 to assess stakeholder viewpoints on a Match and found that the community was divided. Radiology departments and Abdominal Radiology fellowship program directors have voluntarily complied with a series of guidelines laid out by the Society of Chairs in Academic Radiology Departments during the two most recent recruiting cycles, but challenges in the process persist. Stakeholders report improved organization and fairness as a result of these procedural changes, and the authors suggest that Abdominal Radiology may continue to consider a formal fellowship Match in coming years.

A Conference-Friendly, Hands-on Introduction to Deep Learning for Radiology Trainees.

Artificial or augmented intelligence, machine learning, and deep learning will be an increasingly important part of clinical practice for the next generation of radiologists. It is therefore critical that radiology residents develop a practical understanding of deep learning in medical imaging. Certain aspects of deep learning are not intuitive and may be better understood through hands-on experience; however, the technical requirements for setting up a programming and computing environment for deep learning can pose a high barrier to entry for individuals with limited experience in computer programming and limited access to GPU-accelerated computing. To address these concerns, we implemented an introductory module for deep learning in medical imaging within a self-contained, web-hosted development environment. Our initial experience established the feasibility of guiding radiology trainees through the module within a 45-min period typical of educational conferences.

New blood: trainee and early career engagement augmenting the society of abdominal radiology experience.

Many national radiology societies are recognizing the need for early career and trainee engagement as crucial to keeping their societies relevant, active, and invigorated with new ideas. In this descriptive paper, we review the benefits of establishing the Society of Abdominal Radiology's Resident and Fellow Section and Early Career Committee-including our activities and experience, advice for committee structure, and opportunities for growth.