Author:
Javaid Aamir,Fernandes Philip,Adorno William,Catalano Alexis,Ehsan Lubaina,von Eckstaedt Hans Vizthum,Barnes Barrett,Khan Marium,Raghavan Shyam S.,McGowan Emily,Collins Margaret H.,Rothenberg Marc E.,Moskaluk Christopher A,Brown Donald E.,Syed Sana
Abstract
ABSTRACTBackgroundEosinophilic Esophagitis (EoE) is a chronic inflammatory condition diagnosed by ≥15 eosinophils (Eos) per high-power field (HPF). There is no gold standard for clinical remission and Eo-associated metrics are poorly correlated with symptoms. Deep learning can be used to explore the relationships of tissue features with clinical response.ObjectivesTo determine if deep learning can elucidate tissue patterns in EoE that predict treatments or symptoms at remission.MethodsWe created two deep learning models using esophageal biopsies from histologically normal and EoE patients: one to identify Eos in esophageal biopsies and a second to broadly classify esophageal tissue as EoE vs. normal. We used these models to analyze biopsies at diagnosis and first remission timepoint, as defined by <15 Eos/HPF, in a subset of 19 treatment-naïve patients. Differences in deep learning metrics between patient groups were assessed using Wilcoxon Rank-Sum tests.ResultsAll initial patients were symptomatic at diagnosis and a majority were still suffering from dysphagia at remission. The Eo identification model had a low mean (SD) error of −0.3 (11.5) Eos/HPF. Higher peak and average Eo counts at diagnosis were associated with higher likelihood of being on a food-elimination diet at remission than steroids or proton-pump inhibitor (p<0.05). The EoE classification model had an F1-score of 0.97 for distinguishing normal tissue from EoE. There was a significant decrease from diagnosis in the percentage of EoE-classified tissue among asymptomatic remission patients (p<0.05).ConclusionsDeep learning may have utility in diagnosing EoE and predicting future treatment response at diagnosis and resolution of symptoms at follow-up.Clinical Implications or Key Messages (for mechanistic article)We developed two deep learning approaches for tissue analysis in eosinophilic esophagitis, which may improve histologic assessment of patients at diagnosis and predict treatment response and symptoms at remission.Capsule summaryTwo deep learning approaches for eosinophilic esophagitis (EoE): (1) Quantification of eosinophils throughout an entire biopsy, which predicted treatment at remission (2) Classifying esophageal tissue as EoE or normal, which predicted symptoms at remission.
Publisher
Cold Spring Harbor Laboratory
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献