EMEvan McNabb

Evan McNabb

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Evan McNabb has a total of 12 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Robustness of Deep Learning Segmentation Across Heterogeneous Uterine MRI Datasets: A Multi-Task Comparison of U-Net, Swin-UNETR, MedNeXt, and nnU-Net

    by Daniel A. Di Giovanni, Akiyo Takada, Evan McNabb, Jérémy Dana, Hajime Yokota, Takahiro Tsuboyama, Rita Zakarian, Martin Vallières, James Man Git Tsui, Caroline Reinhold

    Reliable medical image segmentation requires performance that is stable across tasks, acquisition settings, and evaluation domains. We evaluated U-Net, Swin-UNETR, and MedNeXt as manually configured architectures and nnU-Net as a self-configuring reference across three T2-weighted uterine MRI datasets: public multiclass uterine anatomy and fibroid segmentation (UMD, n=300), institutional endometrial cancer tumor segmentation (n=206), and institutional uterine mass lesion segmentation (n=234). A separate UMD-style cohort (n=12), relabeled to the same annotation ontology, was used to assess external domain shift. Fixed data partitions, fold ensembling, and matched evaluation were used across models. Performance was assessed with Dice, 95th-percentile Hausdorff distance, average symmetric surface distance, absolute volume difference, and paired bootstrap comparisons with within-dataset Holm correction. MedNeXt was the strongest manually configured architecture, whereas nnU-Net achieved the highest Dice on all internal datasets: 0.761 for UMD, 0.746 for endometrial cancer, and 0.814 for uterine mass segmentation, compared with 0.722, 0.726, and 0.789 for MedNeXt. The nnU-Net-MedNeXt separation was largest for multiclass UMD segmentation and smaller for the binary tasks. All models degraded substantially on external testing; Dice was 0.542 for nnU-Net, 0.490 for MedNeXt, 0.396 for U-Net, and 0.287 for Swin-UNETR. These findings show that architecture, task definition, and dataset-adaptive pipeline configuration jointly influence uterine MRI segmentation, while external domain shift remains a major source of failure.

  • Open Access

    Concordance Between MR Safety Guidance Documents on Conditions for a Safe MR Environment

    by Ives R. Levesque, Véronique Fortier, Jorge Campos Pazmiño, Evan McNabb

    Background: Environmental conditions established at facility design, construction, updates, and/or procurement of specialized equipment are the foundation of safe MR practice. Establishing an institutional MR safety program must take into consideration the conditions of the MR environment, and staff training should be tailored to its features. Purpose/Hypothesis: To identify the concordance on conditions for a safe MR environment among a variety of widely known national guidance documents. Study Type: Analysis of guidance documents. Population/Subjects/Phantom/Specimen/Animal Model: Not applicable. Field Strength/Sequence: Not specific to field strength. Assessment: Information was extracted from national MR safety guidance documents about the expected conditions in a safe MR environment, and categorized among (1) facility design, (2) MR equipment selection, and (3) peripheral equipment. Commonalities and differences were identified. Statistical Tests: Descriptive statistics. Results: Six publicly available national guidance documents for MR safety were obtained, from countries with relatively high density of MR systems (between 9 and 38 MR systems per million inhabitants): Australia, New Zealand, Sweden, Switzerland, UK, and USA. Full or partial concordance was observed among many measures related to facility design (21/25 conditions), MR equipment features (7/9 conditions), and peripheral equipment (10/13 conditions). Data Conclusion: This work identified the common and divergent recommendations on facilities and equipment from a variety of guidance documents from national sources. Highlighting common (and uncommon) features recommended by guidance can help users identify weaknesses in their MR environment. This could inform jurisdictions/countries without official guidance and serve in the development of future guidance documents or MR safety training materials. Evidence Level: 5. Technical Efficacy: 1.