Publications by authors named "Huy Q Vo"
Background: The advent of digital nephropathology offers the potential to integrate deep learning algorithms into the diagnostic workflow. We introduce PICASO, a novel permutation-invariant set operator to dynamically aggregate histopathologic features from instances. We applied PICASO to two nephropathology scenarios: detecting active crescent lesions in sets of glomerular crops with IgA nephropathy and case-level classification for antibody-mediated rejection (AMR) in kidney transplant.
View Article and Find Full Text PDFArticle Synopsis
- The study aimed to replicate the Oxford Classification for IgA nephropathy using a deep learning pipeline called MESCnn, which integrates automatic glomerular segmentation and classification for key glomerular components.
- A dataset of 1056 whole slide images from kidney biopsies was annotated, and models were trained and tested to achieve accurate detection and classification of mesangial hypercellularity, endocapillary hypercellularity, segmental sclerosis, and active crescents.
- Results showed that the segmentation models performed well, demonstrated by high accuracy metrics, with EfficientNetV2-L and MobileNetV2 providing the best results for different classification categories, indicating the potential for effective computer-aided diagnosis in nephropathology.