Checklist for Reproducibility of Deep Learning in Medical Imaging.

The application of deep learning (DL) in medicine introduces transformative tools with the potential to enhance prognosis, diagnosis, and treatment planning. However, ensuring transparent documentation is essential for researchers to enhance reproducibility and refine techniques. Our study addresses the unique challenges presented by DL in medical imaging by developing a comprehensive checklist using the Delphi method to enhance reproducibility and reliability in this dynamic field. We compiled a preliminary checklist based on a comprehensive review of existing checklists and relevant literature. A panel of 11 experts in medical imaging and DL assessed these items using Likert scales, with two survey rounds to refine responses and gauge consensus. We also employed the content validity ratio with a cutoff of 0.59 to determine item face and content validity. Round 1 included a 27-item questionnaire, with 12 items demonstrating high consensus for face and content validity that were then left out of round 2. Round 2 involved refining the checklist, resulting in an additional 17 items. In the last round, 3 items were deemed non-essential or infeasible, while 2 newly suggested items received unanimous agreement for inclusion, resulting in a final 26-item DL model reporting checklist derived from the Delphi process. The 26-item checklist facilitates the reproducible reporting of DL tools and enables scientists to replicate the study's results.

Reading Room Interruptions are Less Disruptive When Using Asynchronous Communication Methods.

Radiologist interruptions, though often necessary, can be disruptive. Prior literature has shown interruptions to be frequent, occurring during cases, and predominantly through synchronous communication methods such as phone or in person causing significant disengagement from the study being read. Asynchronous communication methods are now more widely available in hospital systems such as ours. Considering the increasing use of asynchronous communication methods, we conducted an observational study to understand the evolving nature of radiology interruptions. We hypothesize that compared to interruptions occurring through synchronous methods, interruptions via asynchronous methods reduce the disruptive nature of interruptions by occurring between cases, being shorter, and less severe. During standard weekday hours, 30 radiologists (14 attendings, 12 residents, and 4 fellows) were directly observed for approximately 90-min sessions across three different reading rooms (body, neuroradiology, general). The frequency of interruptions was documented including characteristics such as timing, severity, method, and length. Two hundred twenty-five interruptions (43 Teams, 47 phone, 89 in-person, 46 other) occurred, averaging 2 min and 5 s with 5.2 interruptions per hour. Microsoft Teams interruptions averaged 1 min 12 s with only 60.5% during cases. In-person interruptions averaged 2 min 12 s with 82% during cases. Phone interruptions averaged 2 min and 48 s with 97.9% during cases. A substantial portion of reading room interruptions occur via predominantly asynchronous communication tools, a new development compared to prior literature. Interruptions via predominantly asynchronous communications tools are shorter and less likely to occur during cases. In our practice, we are developing tools and mechanisms to promote asynchronous communication to harness these benefits.

The Virtual Radiology Reading Room: Initial Perceptions of Referring Providers and Radiologists.

The objective of this study is to assess the initial perception of referring providers and radiologists to a virtual consultation solution (the Virtual Radiology Reading Room, VR3). VR3 is specifically designed to replace the radiology reading room phone and enable the radiologist to continue to work as part of the clinical care team fielding radiology consults without being confined to the same physical space. Surveys of providers' and radiologists' initial experience were conducted approximately 6 months after initial deployment. Users were asked about their overall impression, and how well the solution integrates with their workflow as well as how it compares to traditional phone calls to the reading room. Forty of 71 referring providers and 27 of 44 radiologists responded to our survey. VR3 was rated 4.7 out of 5 stars by referrers and 4.1 by radiologists. Seventy percent of referrers and radiologists preferred VR3 to the phone while 4.5% of referrers and 11% of radiologists preferred the phone. Referring providers and radiologists expressed a positive initial perception of the Virtual Radiology Reading Room and prefer it to traditional phone calls to the reading room.

Approach to Cystic Lesions in the Abdomen and Pelvis, with Radiologic-Pathologic Correlation.

Cystic lesions found in and around the peritoneal cavity can often be challenging to diagnose owing to significant overlap in imaging appearance between the different entities. When the cystic lesion can be recognized to arise from one of the solid abdominal organs, the differential considerations can be more straightforward; however, many cystic lesions, particularly when large, cannot be clearly associated with one of the solid organs. Cystic lesions arising from the mesentery and peritoneum are less commonly encountered and can be caused by relatively rare entities or by a variant appearance of less-rare entities. The authors provide an overview of the classification of cystic and cystic-appearing lesions and the basic imaging principles in evaluating them, followed by a summary of the clinical, radiologic, and pathologic features of various cystic and cystic-appearing lesions found in and around the peritoneal cavity, organized by site of origin. Emphasis is given to lesions arising from the mesentery, peritoneum, or gastrointestinal tract. Cystic lesions arising from the liver, spleen, gallbladder, pancreas, urachus, adnexa, or soft tissue are briefly discussed and illustrated with cases to demonstrate the overlap in imaging appearance with mesenteric and peritoneal cystic lesions. When approaching a cystic lesion, the key imaging features to assess include cyst content, locularity, wall thickness, and presence of internal septa, solid components, calcifications, or any associated enhancement. While definitive diagnosis is not always possible with imaging, careful assessment of the imaging appearance, location, and relationship to adjacent structures can help narrow the differential diagnosis. Online supplemental material is available for this article. ©RSNA, 2021.

Errors and Misinterpretations in Imaging of Chronic Liver Diseases.

MRI is an important problem-solving tool for accurate characterization of liver lesions. Chronic liver disease alters the typical imaging characteristics and complicates liver imaging. Awareness of imaging pitfalls and technical artifacts and ways to mitigate them allows for more accurate and timely diagnosis.

Therapies for hepatocellular carcinoma: overview, clinical indications, and comparative outcome evaluation-part one: curative intention.

Hepatocellular carcinoma (HCC) offers unique management challenges as it commonly occurs in the setting of underlying chronic liver disease. The management of HCC is directed primarily by the clinical stage. The most commonly used staging system is the Barcelona-Clinic Liver Cancer system, which considers tumor burden based on imaging, liver function and the patient's performance status. Early-stage HCC can be managed with therapies of curative intent including surgical resection, liver transplantation, and ablative therapies. This manuscript reviews the various treatment options for HCC with a curative intent, such as locablative therapy types, surgical resection, and transplant. Indications, contraindications and outcomes of the various treatment options are reviewed. Multiple concepts relating to liver transplant are discussed including Milan criteria, OPTN policy, MELD exception points, downstaging to transplant and bridging to transplant.

Therapies for hepatocellular carcinoma: overview, clinical indications, and comparative outcome evaluation. Part two: noncurative intention.

Locoregional therapies can be offered to hepatocellular carcinoma patients as a bridge to transplant, to downstage disease burden for transplant eligibility, or for disease control to prolong survival. Systemic therapies also play a large role in HCC treatment, occasionally in conjunction with other methods. This manuscript reviews the various treatment options for HCC with a historically noncurative intent.

Radiology Reporting in the Era of Patient-Centered Care: How Can We Improve Readability?

Radiology reports are consumed not only by referring physicians and healthcare providers, but also by patients. We assessed report readability in our enterprise and implemented a two-part quality improvement intervention with the goal of improving report accessibility. A total of 491,813 radiology reports from ten hospitals within the enterprise from May to October, 2018 were collected. We excluded echocardiograms, rehabilitation reports, administrator reports, and reports with negative scores leaving 461,219 reports and report impressions for analysis. A grade level (GL) was calculated for each report and impression by averaging four readability metrics. Next, we conducted a readability workshop and distributed weekly emails with readability GLs over a period of 6 months to each attending radiologist at our primary institution. Following this intervention, we utilized the same exclusion criteria and analyzed 473,612 reports from May to October, 2019. The mean GL for all reports and report impressions was above 13 at every hospital in the enterprise. Following our intervention, a statistically significant drop in GL for reports and impressions was demonstrated at all locations, but a larger and significant improvement was observed in impressions at our primary site. Radiology reports across the enterprise are written at an advanced reading level making them difficult for patients and their families to understand. We observed a significantly larger drop in GL for impressions at our primary site than at all other sites following our intervention. Radiologists at our home institution improved their report readability after becoming more aware of their writing practices.

The Radiology Virtual Reading Room: During and Beyond the COVID-19 Pandemic.

The COVID-19 pandemic has disrupted the radiology reading room with a potentially lasting impact. This disruption could introduce the risk of obviating the need for the reading room, which would be detrimental to many of the roles of radiology that occur in and around the reading room. This disruption could also create the opportunity for accelerated evolution of the reading room to meet the strategic needs of radiology and health care through thoughtful re-design of the virtual reading room. In this article, we overview the impact of the COVID-19 pandemic on radiology in our institution and across the country, specifically on the dynamics of the radiology reading room. We introduce the concept of the virtual reading room, which is a redesigned alternative to the physical reading room that can serve the diverse needs of radiology and healthcare during and beyond the pandemic.

NRG Oncology/RTOG Consensus Guidelines for Delineation of Clinical Target Volume for Intensity Modulated Pelvic Radiation Therapy in Postoperative Treatment of Endometrial and Cervical Cancer: An Update.

PURPOSE: Accurate target definition is critical for the appropriate application of radiation therapy. In 2008, the Radiation Therapy Oncology Group (RTOG) published an international collaborative atlas to define the clinical target volume (CTV) for intensity modulated pelvic radiation therapy in the postoperative treatment of endometrial and cervical cancer. The current project is an updated consensus of CTV definitions, with removal of all references to bony landmarks and inclusion of the para-aortic and inferior obturator nodal regions. METHODS AND MATERIALS: An international consensus guideline working group discussed modifications of the current atlas and areas of controversy. A document was prepared to assist in contouring definitions. A sample case abdominopelvic computed tomographic image was made available, on which experts contoured targets. Targets were analyzed for consistency of delineation using an expectation-maximization algorithm for simultaneous truth and performance level estimation with kappa statistics as a measure of agreement between observers. RESULTS: Sixteen participants provided 13 sets of contours. Participants were asked to provide separate contours of the following areas: vaginal cuff, obturator, internal iliac, external iliac, presacral, common iliac, and para-aortic regions. There was substantial agreement for the common iliac region (sensitivity 0.71, specificity 0.981, kappa 0.64), moderate agreement in the external iliac, para-aortic, internal iliac and vaginal cuff regions (sensitivity 0.66, 0.74, 0.62, 0.59; specificity 0.989, 0.966, 0.986, 0.976; kappa 0.60, 0.58, 0.52, 0.47, respectively), and fair agreement in the presacral and obturator regions (sensitivity 0.55, 0.35; specificity 0.986, 0.988; kappa 0.36, 0.21, respectively). A 95% agreement contour was smoothed and a final contour atlas was produced according to consensus. CONCLUSIONS: Agreement among the participants was most consistent in the common iliac region and least in the presacral and obturator nodal regions. The consensus volumes formed the basis of the updated NRG/RTOG Oncology postoperative atlas. Continued patterns of recurrence research are encouraged to refine these volumes.

Clinicians and surgeon survey regarding current and future versions of CT/MRI LI-RADS.

PURPOSE: To determine preferences of clinicians and surgeons regarding radiology reporting of liver observations in patients at risk for hepatocellular carcinoma (HCC). METHODS: Members of the American College of Radiology Liver Imaging and Data Reporting System (LI-RADS) Outreach & Education Group (30 members) as well as Society of Abdominal Radiology Disease-Focused Panel on HCC diagnosis (27 members) created and distributed an 18-question survey to clinicians and surgeons, with focus on preferences regarding radiology reporting of liver observations in patients. The survey questions were directed to physician demographics, current use of LI-RADS by their local radiologists, their opinions about current LI-RADS and potential improvements. RESULTS: A total of 152 physicians responded, 66.4% (101/152) from North America, including 42 surgeons, 81 physicians and 29 interventional radiologists. Participants were predominantly from academic centers 83% (126/152), while 13.8% (21/152) worked in private/community centers and 3.2% (5/152) worked in a hybrid practice. Almost 90% (136/152) of participants preferred the use of LI-RADS (compared to nothing or other standardized reporting systems; OPTN and AASLD) to communicate liver-related observations. However, only 28.5% (43/152) of participants input was sought at the time of implementing LI-RADS in their institutions. Fifty-eight percent (88/152) of all participants found standardized LI-RADS management recommendations in radiology reports to be clinically helpful. However, a subgroup analysis of surgeons in academic centers showed that 61.8% (21/34) prefer not to receive standardized LI-RADS recommendations. CONCLUSIONS: Most participants preferred the use LI-RADS in reporting CT and MRI examination. When considering inclusion of management recommendations, radiologists should consult with their referring physicians, as preference may differ.

Suitability of the woodchuck HCC as a preclinical model for evaluation of intra-arterial therapies.

The most commonly used preclinical models of hepatocellular carcinoma (HCC) are limited for use in testing of intra-arterial therapies such as transarterial chemoembolization and radioembolization. Issues encountered with the more commonly used animal models include dissimilarity in their disease development compared with humans and the size of the vasculature which can make intra-arterial therapy testing difficult or impossible. Here we describe the suitability of the woodchuck HCC model for testing of intra-arterial therapies. We describe the techniques for pre-embolization imaging assessment using CT and MRI, technical tips on performing angiography and embolization, and pathological assessment of treated liver.

Integration of Peer Review in PACS Results in a Marked Increase in the Discrepancies Reported.

OBJECTIVE. The objective of this article is to assess the impact of integrating peer review in PACS on the reporting of discrepancies. Our hypothesis is that a PACS-integrated machine-randomized and semiblinded peer review tool leads to an increase in discrepancies reported. MATERIALS AND METHODS. A PACS tool was implemented to prompt radiologists to perform peer review of prior comparison studies in a randomized fashion. The reviewed radiologist's name was omitted from the prior report in PACS. Before this implementation, radiologists entered peer reviews directly on the RADPEER website. Three academic subspecialty sections comprising 24 radiologists adopted the tool (adopters group). Three sections comprising 14 radiologists did not adopt the tool (nonadopters group). Peer review submissions were analyzed for 4 months before and 4 months after the implementation. The mean rate of significant discrepancies (RADPEER score 2b or higher) reported per radiologist was calculated and the discrepancy rates of the periods before and after the implementation were compared. RESULTS. The mean significant discrepancy rate reported per radiologist in the adopters group increased from 0.19% ± 0.46% (SD) before the implementation to 0.93% ± 1.45% after implementation (p = 0.01). No significant discrepancies were reported by the nonadopters group in either period. CONCLUSION. In this single institutional retrospective analysis, integrating peer review in PACS resulted in a fivefold increase in reported significant discrepancies. These results suggest that peer review data are influenced by the design of the tool used including PACS integration, randomization, and blinding.

Pitfalls in liver MRI: Technical approach to avoiding misdiagnosis and improving image quality.

The following is an illustrative review of common pitfalls in liver MRI that may challenge interpretation. This article reviews common technical and diagnostic challenges encountered when interpreting dynamic multiphasic T1 -weighted imaging, hepatobiliary phase imaging, and diffusion-weighted imaging of the liver. Additionally, each section includes suggestions for avoiding diagnostic and technical errors. Level of Evidence: 5 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:41-58.

White paper of the Society of Abdominal Radiology hepatocellular carcinoma diagnosis disease-focused panel on LI-RADS v2018 for CT and MRI.

The Liver Imaging and Reporting Data System (LI-RADS) is a comprehensive system for standardizing the terminology, technique, interpretation, reporting, and data collection of liver imaging with the overarching goal of improving communication, clinical care, education, and research relating to patients at risk for or diagnosed with hepatocellular carcinoma (HCC). In 2018, the American Association for the Study of Liver Diseases (AASLD) integrated LI-RADS into its clinical practice guidance for the imaging-based diagnosis of HCC. The harmonization between the AASLD and LI-RADS diagnostic imaging criteria required minor modifications to the recently released LI-RADS v2017 guidelines, necessitating a LI-RADS v2018 update. This article provides an overview of the key changes included in LI-RADS v2018 as well as a look at the LI-RADS v2018 diagnostic algorithm and criteria, technical recommendations, and management suggestions. Substantive changes in LI-RADS v2018 are the removal of the requirement for visibility on antecedent surveillance ultrasound for LI-RADS 5 (LR-5) categorization of 10-19 mm observations with nonrim arterial phase hyper-enhancement and nonperipheral "washout", and adoption of the Organ Procurement and Transplantation Network definition of threshold growth (≥ 50% size increase of a mass in ≤ 6 months). Nomenclatural changes in LI-RADS v2018 are the removal of -us and -g as LR-5 qualifiers.

Providing MR Imaging for Cervical Cancer Brachytherapy: Lessons for Radiologists.

Brachytherapy (BT), the use of a locally placed or implanted radioactive source for treatment of an adjacent tumor, is an important component in the treatment of patients with both early- and advanced-stage cervical cancer and is increasingly part of the standard treatment protocol. When it is feasible, many radiation oncologists choose to include a magnetic resonance (MR) imaging examination for planning BT treatment (ie, an MR imaging examination after placement of the applicator but before radiation dosing). MR imaging provides excellent soft-tissue contrast and allows radiation oncologists to individualize the radiation dose to the target volume and minimize the dose to adjacent organs that are at risk for radiation damage. However, traditionally, the radiology department has not performed imaging studies for planning, and the requirements are different compared with those of standard diagnostic imaging. In addition, many applicators are available for use in BT treatment of cervical cancer, and each must considered separately to determine MR safety and to define the best imaging parameters. Starting and supporting a robust gynecologic BT program includes implementing imaging protocols that are helpful to both radiation oncologists and diagnostic radiologists. By becoming more familiar with this treatment modality and the logistics of imaging patients undergoing BT, radiologists can provide imaging support for colleagues in the radiation oncology department and better care for patients. ©RSNA, 2018.

MR Imaging of Prostate Zonal Anatomy.

McNeal first described the zonal anatomy of the prostate about 40 years ago, outlining 4 zones of the prostate and defining their relation to the urethra and the ejaculatory ducts. The zonal anatomy remains the accepted model for describing the prostate and the zones are well-depicted on MR imaging, including the central zone, which until recently was grouped with the transition zone in the radiology literature. An accurate understanding of the zonal anatomy and periprostatic anatomy is key for accurate interpretation of the prostate MR imaging.

Reduction of MRI signal distortion from titanium intracavitary brachytherapy applicator by optimizing pulse sequence parameters.

PURPOSE: To demonstrate that optimized pulse sequence parameters for a T2-weighted (T2w) fast spin echo acquisition reduced artifacts from a titanium brachytherapy applicator compared to conventional sequence parameters. METHODS AND MATERIALS: Following Institutional Review Board approval and informed consent, seven patients were successfully imaged with both standard sagittal T2w fast spin echo parameters (voxel size of 0.98 × 0.78 × 4.0 mm3; readout bandwidth of 200 Hz/px; repetition time of 2800 ms; echo time of 91 ms; echo train length of 15; 36 slices; and imaging time of 3:16 min) and an additional optimized T2w sequence (voxel size of 0.98 × 0.98 × 4.0 mm3; readout bandwidth of 500 Hz/px; repetition time of 3610 ms; echo time of 91 ms; echo train length of 25; 18-36 slices; and imaging time of 1:15-2:30 min), which had demonstrated artifact reduction in prior phantom work. Visualized intracavitary tandem was hand-segmented by two of the authors. Three body imaging radiologists assessed image quality and intraobserver agreement scores were analyzed. RESULTS: The average segmented volume of the intracavitary applicator significantly (p < 0.05) decreased with the experimental pulse sequence parameters as compared to the standard pulse sequence. Comparison of experimental and standard T2w sequence qualitative scores for each reviewer showed no significant differences between the two techniques. CONCLUSIONS: This study demonstrated that pulse sequence parameter optimization can significantly reduce distortion artifact from titanium applicators while maintaining image quality and reasonable imaging times.

How one institution overcame the challenges to start an MRI-based brachytherapy program for cervical cancer.

PURPOSE: Adaptive magnetic resonance imaging (MRI)-based brachytherapy results in improved local control and decreased high-grade toxicities compared to historical controls. Incorporating MRI into the workflow of a department can be a major challenge when initiating an MRI-based brachytherapy program. This project aims to describe the goals, challenges, and solutions when initiating an MRI-based cervical cancer brachytherapy program at our institution. MATERIAL AND METHODS: We describe the 6-month multi-disciplinary planning phase to initiate an MRI-based brachytherapy program. We describe the specific challenges that were encountered prior to treating our first patient. RESULTS: We describe the solutions that were realized and executed to solve the challenges that we faced to establish our MRI-based brachytherapy program. We emphasize detailed coordination of care, planning, and communication to make the workflow feasible. We detail the imaging and radiation physics solutions to safely deliver MRI-based brachytherapy. The focus of these efforts is always on the delivery of optimal, state of the art patient care and treatment delivery within the context of our available institutional resources. CONCLUSIONS: Previous publications have supported a transition to MRI-based brachytherapy, and this can be safely and efficiently accomplished as described in this manuscript.

Point A vs. HR-CTV D90 in MRI-based cervical brachytherapy of small and large lesions.

PURPOSE: To evaluate the dosimetric benefits of MRI-based brachytherapy in small and large high-risk clinical target volume (HR-CTV) in cervical cancer. METHODS AND MATERIALS: Twenty-eight fractions obtained from sixteen cervical cancer patients treated with MRI-based high-dose-rate brachytherapy with standard tandem and ovoid applicators were used; original fractions were optimized to HR-CTV D90. Fractions were separated based on the median volume into small and large (HR-CTV <25 cm3 or >25 cm3) lesion groups. Retrospective plans prescribed to Point A were created for each fraction. D0.1 cc, D2 cc, and International Commission of Radiation Unit and Measurements (ICRU) points were used to compare Point A vs. HR-CTV D90 plans for bladder, rectum, and sigmoid. RESULTS: In the small lesion group, Point A plans vs. HR-CTV D90 plans had significantly higher D0.1 cc, D2 cc, and ICRU points for bladder, rectum, and sigmoid (p < 0.05). In the large lesion group, there was no significant difference between Point A and HR-CTV D90 plans for D0.1 cc, D2 cc, and ICRU points to the organs at risk (OARs). CONCLUSIONS: The dosimetric advantages to OARs offered by MRI-based brachytherapy with prescription to HR-CTV D90 compared to Point A is most distinct for patients with smaller HR-CTV (<25 cm3). This study demonstrates sufficient tumor coverage with lower doses to OARs in HR-CTV D90 vs. Point A plans in the small lesion group. These improvements were not seen in the large lesion group, indicating a lesser dosimetric advantage of HR-CTV D90 compared to Point A planning when the cervical lesion is >25 cm3. Incorporation of interstitial needles for patients with larger HR-CTV is likely the best method to decrease dose to OARs and improve tumor coverage.