BriefCase-Quantification

BriefCase-Quantification is an AI-powered medical image processing software designed to assist clinicians by automatically measuring the maximum abdominal aortic diameter from CT scans using an orthogonal measurement technique. Its methodology relies on a deep-learning algorithm trained on a large dataset to analyze DICOM images, providing quantitative data intended to support, but not replace, radiologist assessment in evaluating normal and aneurysmal aortas.

BriefCase-Quantification is a cloud-based software device utilizing an artificial intelligence, deep-learning algorithm for radiological medical image management and processing. Its primary purpose is to assist trained medical specialists by automatically measuring the maximum abdominal aortic diameter from CT exams with contrast in adult and transitional adolescent patients (18 years and older). The software processes filtered DICOM images, with its algorithm identifying relevant series to perform the measurement. A key technological advancement is its method of measuring orthogonally to the blood flow, presenting the key slice with long and short axes of the aorta in a reformatted slice, an improvement over previous axial measurement techniques. This AI-driven analysis is intended for normal and aneurysmal abdominal aortas, excluding post-operative cases, and the results are provided as an aid, not a standalone diagnostic tool. Performance evaluation of BriefCase-Quantification was conducted through a retrospective, blinded, multicenter study involving 162 cases from diverse US clinical sites where the software's measurements were compared against a ground truth established by three board-certified radiologists. The study successfully met its primary endpoint, demonstrating a mean absolute error of 1.52 mm (95% CI: 1.20 mm, 1.83 mm) between the algorithm's measurement and the ground truth, indicating improved accuracy over its predicate. Additionally, a Bland-Altman analysis revealed a mean difference of 0.58 mm, suggesting minimal bias, and these results, obtained from a dataset distinct from the training data and encompassing various clinical subgroups and scanner manufacturers, affirm the device's capability to provide reliable adjunctive measurements for abdominal aortic diameter.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the algorithm's performance in measuring abdominal aortic diameter vary across different CT scanner manufacturers and imaging protocols beyond those included in the pivotal study?; What is the specific impact of training the algorithm on a "larger data set" on its ability to accurately identify and measure aortas with complex pathologies (e.g., significant tortuosity, mural thrombus, or subtle aneurysms)?; Could the orthogonal measurement technique employed by BriefCase-Quantification offer advantages in specific patient subgroups or aortic morphologies compared to traditional axial measurements, and how might this be quantified?; What are the limitations or failure modes of the AI algorithm, particularly in cases with poor image quality, atypical anatomy, or the presence of artifacts not extensively covered in the clinical subgroups?; How does the presentation of the reformatted slice with orthogonal measurements affect radiologist workflow, interpretation time, and confidence in the provided measurements compared to reviewing axial slices alone or with axial AI measurements?