† People involved in the organization of the challenge. ‡ People contributing data from their institutions.§ Equal senior authors.

No abstract

Fractures of the cervical spine are a medical emergency and may lead to permanent paralysis and even death. Accurate diagnosis in patients with suspected fractures by computed tomography (CT) is critical to patient management. In this paper, we propose a deep convolutional neural network (DCNN) with a bidirectional long-short term memory (BLSTM) layer for the automated detection of cervical spine fractures in CT axial images. We used an annotated dataset of 3,666 CT scans (729 positive and 2,937 negative cases) to train and validate the model. The validation results show a classification accuracy of 70.92% and 79.18% on the balanced (104 positive and 104 negative cases) and imbalanced (104 positive and 419 negative cases) test datasets, respectively.

Introduction: This needs assessment evaluated residents’ and medical students’ knowledge of Competence by Design (CBD), perceived benefits of and challenges or barriers to the transition to CBD for residents, and perceived overall preparedness for the transition to CBD in diagnostic radiology. Materials and Methods: All diagnostic radiology residents and medical students in Canada were eligible to participate in this national cross-sectional, questionnaire-based needs assessment. Knowledge of CBD was evaluated through participants’ self-reported rating of their knowledge of CBD on a 5-point Likert scale. Perceived benefits of and challenges or barriers to the transition to CBD for residents were rank ordered. Participants’ overall self-reported preparedness for the transition to CBD was assessed on a 5-point Likert scale. Data were summarized by descriptive statistics and bivariate analyses were conducted as appropriate. Results: Ninety-four residents ( n = 77) and medical students ( n = 17) participated in this needs assessment. Participants’ mean ± standard deviation self-reported rating of their overall knowledge of CBD was 2.86 ± .94. Provision of meaningful feedback to learners and learners’ ability to identify their own educational needs were among the highest ranked perceived benefits of the transition to CBD, while demands on time and increased frequency of evaluation were among the highest ranked perceived challenges or barriers to the transition to CBD. Few participants reported being either “prepared” (4.7%) or “somewhat prepared” (14.0%) for the transition to CBD. Conclusion: Preparedness for the transition to CBD in diagnostic radiology may be improved. Targeted interventions to augment the preparedness of residents and medical students should be considered.

Objective During the COVID-19 pandemic, there was a temporary cessation of mammography screening. However, in some facilities, diagnostic breast imaging services continued for patients with a high clinical suspicion of breast cancer. The objective of this study was to evaluate changes in the diagnostic interval (DI) of non-screening patients presenting for diagnostic mammography during the first wave of the COVID-19 pandemic. Methods Retrospective chart review was performed on patients presenting for non-screening diagnostic mammography from April 1 to June 30, 2020 ( pandemic group ) and April 1 to June 30, 2019 ( pre-pandemic group ). Age, reason for referral, number and type of imaging studies/biopsies necessary for a final diagnosis were recorded. Diagnostic interval (DI) was defined as the number of days from the date of the diagnostic mammogram to the date of the final diagnosis. Results Compared to the pre-pandemic group (n = 64), the pandemic group (n = 77) showed a reduction in DI of the entire cohort (pandemic: 1 day; pre-pandemic: 15 days, p < 0.0001) for patients not requiring tissue sampling (pandemic: 1 day; pre-pandemic: 11 days, . p < 0.0001) and those requiring tissue sampling with benign pathology (pandemic 9 days; pre-pandemic, 33 days, p = 0.0002). A higher percentage of patients in the pandemic group had their assessment completed during the initial visit (pandemic: 50.6%; pre-pandemic: 23.4%, p = 0.0009). Conclusion During the first wave of the COVID-19 pandemic, the DI for patients with non-screening-related diagnostic mammography was significantly shorter, with a higher percentage of patients completing their assessments on the initial visit, compared to one year prior. Key Points • Despite reductions in manpower and clinical services, during pandemic times, it is possible to maintain a diagnostic breast imaging service for women at high clinical suspicion for breast cancer. • During pandemic times, breast imaging departments should consider restructuring to a Rapid Diagnostic Unit model with a navigation team that follows patients through the assessment process to a final diagnosis. • Departmental restructuring and patient navigation during pandemic times could either maintain or shorten the diagnostic interval for patients presenting for diagnostic mammography.