Classification of salivary gland biopsies in Sjögren's syndrome by a convolutional neural network using an auto-machine learning platform

BackgroundThe histopathological analysis of minor salivary gland biopsies, particularly through the quantification of the Focus Score (FS), is pivotal in the diagnostic workflow for Sj & ouml;gren's Syndrome (SS). AI-based image recognition using deep learning models has demonstrated potential in enhancing diagnostic accuracy and efficiency in preclinical research.ObjectivesThe primary aim of this investigation was to utilize an auto-machine learning (autoML) platform for the automated segmentation and quantification of FS on histopathological slides, aiming to augment diagnostic precision and speed in SS.MethodsA cohort comprising 86 patients with sicca syndrome (37 diagnosed with SS based on the 2016 ACR/EULAR Classification Criteria and 49 non-SS) was selected for an in-depth histological examination. A repository of 172 slides (two per patient) was assembled, encompassing 74 slides meeting the classificatory thresholds for SS (FS >= 1, indicative of lymphocytic infiltration) and 98 slides showcasing normal salivary gland histology. The autoML platform utilized (Giotto, L2F, Lausanne Switzerland) employed a Convolutional Neural Network (CNN) architecture (ResNet-152) for the training and validation phases, using a dataset of 172 slides.ResultsThe developed model exhibited a reliability score of 0.88, proficiently distinguishing SS cases, with a sensitivity of 89.47% (95% CI: 66.86% to 98.70%) and a specificity of 88.24% (95% CI: 63.56% to 98.54%). The model found histological slides of suboptimal quality (e.g., those compromised during fixation or staining processes) to be the most challenging for accurate classification.ConclusionAutoML platforms offer a rapid and flexible approach to developing machine learning models, even with smaller datasets, as demonstrated in this study for SS. These platforms hold significant potential for enhancing diagnostic precision and efficiency in both clinical and research settings. Multicentric studies with larger patient cohorts are essential for thorough evaluation and validation of this innovative diagnostic approach. Histopathological analysis of minor salivary gland biopsies, especially the quantification of the Focus Score (FS), is crucial for the diagnostic workflow in Sj & ouml;gren's Syndrome (SS).- AI-based image recognition and deep learning models have shown promise in enhancing diagnostic workflows and have potential applications in preclinical research- This study utilizes an auto-machine learning (autoML) platform for automated delineation and quantification of FS on histopathological slides, aiming to improve diagnostic precision and speed in SS.- Offers a potential solution to mitigate the challenge of interobserver variability in FS assessment, thus strengthening the diagnostic criteria and potentially improving diagnostic accuracy in SS.- The flexibility and agility of autoML platforms like Giotto could expedite the diagnostic process, presenting a viable tool for clinicians and researchers, even with smaller datasets.