Publications by authors named "E P Ambinder"
Introduction: Focal non-mass enhancement (NME) is a common breast MRI finding with limited data to guide management. This study aimed to assess clinical and imaging features of malignant BI-RADS 4 focal NME.
Methods: This IRB-approved, retrospective study included breast MRI exams between August 1, 2013 and September 1, 2022 yielding BI-RADS 4 focal NME lesions that underwent core biopsy or excision with available pathology result or demonstrated decrease or resolution during follow-up MRI or at least 2 years of MRI stability.
Cross-site Validation of AI Segmentation and Harmonization in Breast MRI.
J Imaging Inform Med
September 2024
Article Synopsis
- This study evaluated an automated system for segmenting breast cancers in MRI scans and compared its effectiveness to that of radiologists across multiple clinical sites.
- A 3D U-Net model was trained on a substantial dataset and validated against test data from different sites, showing similar performance between the AI and radiologists.
- The findings indicate that the AI can match radiologists' segmentation accuracy and the code and model weights are shared publicly to encourage reproducibility in radiology AI research.
MixTURE: L1-Norm-Based Mixed Second-Order Continuity in Strain Tensor Ultrasound Elastography.
IEEE Trans Ultrason Ferroelectr Freq Control
August 2024
Energy-based displacement tracking of ultrasound images can be implemented by optimizing a cost function consisting of a data term, a mechanical congruency term, and first- and second-order continuity terms. This approach recently provided a promising solution to two-dimensional axial and lateral displacement tracking in ultrasound strain elastography. However, the associated second-order regularizer only considers the unmixed second derivatives and disregards the mixed derivatives, thereby providing suboptimal noise suppression and limiting possibilities for total strain tensor imaging.
View Article and Find Full Text PDFIntroduction: MRI-guided biopsy is the standard of care for breast imaging findings seen only by MRI. Although a non-zero false-negative rate of MRI-guided breast biopsy has been reported by multiple studies, there are varied practice patterns for imaging follow-up after a benign concordant MRI guided biopsy. This study assessed the outcomes of benign concordant MRI-guided biopsies at a single institution.
View Article and Find Full Text PDF