Machine learning streamlines the morphometric characterization and multi-class segmentation of nuclei in different follicular thyroid lesions: everything in a NUTSHELL.

The diagnostic assessment of thyroid nodules is hampered by the persistence of uncertainty in borderline cases, and further complicated by the inclusion of non-invasive follicular tumor with papillary-like nuclear features (NIFTP) as a less aggressive alternative to papillary thyroid carcinoma (PTC). In this setting, computational methods might facilitate the diagnostic process by unmasking key nuclear characteristics of NIFTPs. The main aims of this work were to (1) identify morphometric features of NIFTP and PTC that are interpretable for the human eye, and (2) develop a deep learning model for multi-class segmentation as a support tool to reduce diagnostic variability. Our findings confirmed that nuclei in NIFTP and PTC share multiple characteristics, setting them apart from hyperplastic nodules (HP). The morphometric analysis identified 15 features that can be translated into nuclear alterations readily understandable by pathologists, such as a remarkable inter-nuclear homogeneity for HP in contrast to a major complexity in the chromatin texture of NIFTP, and to the peculiar pattern of nuclear texture variability of PTC. A few NIFTP cases with available NGS data were also analyzed to initially explore the impact of RAS-related mutations on nuclear morphometry. Finally, a pixel-based deep learning model was trained and tested on whole slide images (WSIs) of NIFTP, PTC, and HP cases. The model, named NUTSHELL (NUclei from Thyroid tumors Segmentation to Highlight Encapsulated Low-malignant Lesions), successfully detected and classified the majority of nuclei in all WSIs' tiles, showing comparable results with already well-established pathology nuclear scores. NUTSHELL provides an immediate overview of NIFTP areas and can be used to detect microfoci of PTC within extensive glandular samples or identify lymph node metastases. NUTSHELL can be run inside WSInfer with an easy rendering in QuPath, thus facilitating the democratization of digital pathology.

Informations de copyright

Copyright © 2024. Published by Elsevier Inc.