Direct Gene Expression Profile Prediction for Uveal Melanoma from Digital Cytopathology Images via Deep Learning and Salient Image Region Identification

• Published in: Ophthalmology science (Online) Vol. 3; no. 1; p. 100240
• Main Authors: Liu, T. Y. Alvin, Chen, Haomin, Gomez, Catalina, Correa, Zelia M., Unberath, Mathias
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Inc 01.03.2023; Elsevier
• Subjects: Artificial intelligence; Cytopathology; Deep learning; Gene expression profile; Ophthalmology; Original; Uveal melanoma; ANN; Uveal melanoma; DL; FNAB; ROI; CAM; Gene expression profile; Deep learning; GEP; UM; Cytopathology; FA; Artificial intelligence; DA; ML; deep learning; gene expression profile; machine learning; cytopathology; artificial intelligence; FA, feature aggregation; UM, uveal melanoma; CAM, class activation map; ML, machine learning; DL, deep learning; FNAB, fine-needle aspiration biopsy; ANN, artificial neural network; DA, data augmentation; ROI, region-of-interest; GEP, gene expression profile
• ISSN: 2666-9145; 2666-9145
• DOI: 10.1016/j.xops.2022.100240

To demonstrate that deep learning (DL) methods can produce robust prediction of gene expression profile (GEP) in uveal melanoma (UM) based on digital cytopathology images. Evaluation of a diagnostic test or technology. Deidentified smeared cytology slides stained with hematoxylin and eosin obtained from a fine needle aspirated from UM. Digital whole-slide images were generated by fine-needle aspiration biopsies of UM tumors that underwent GEP testing. A multistage DL system was developed with automatic region-of-interest (ROI) extraction from digital cytopathology images, an attention-based neural network, ROI feature aggregation, and slide-level data augmentation. The ability of our DL system in predicting GEP on a slide (patient) level. Data were partitioned at the patient level (73% training; 27% testing). In total, our study included 89 whole-slide images from 82 patients and 121 388 unique ROIs. The testing set included 24 slides from 24 patients (12 class 1 tumors; 12 class 2 tumors; 1 slide per patient). Our DL system for GEP prediction achieved an area under the receiver operating characteristic curve of 0.944, an accuracy of 91.7%, a sensitivity of 91.7%, and a specificity of 91.7% on a slide-level analysis. The incorporation of slide-level feature aggregation and data augmentation produced a more predictive DL model (P = 0.0031). Our current work established a complete pipeline for GEP prediction in UM tumors: from automatic ROI extraction from digital cytopathology whole-slide images to slide-level predictions. Our DL system demonstrated robust performance and, if validated prospectively, could serve as an image-based alternative to GEP testing.