Peer-reviewed article
Automated Eosinophil Quantification Using Deep Learning to Predict Therapy Escalation in Pediatric Ulcerative Colitis.
James Reigle, Xiaoxuan Liu, Oscar Lopez-Nunez, Erik Drysdale, Dua Qtaish et al. ·
Summary
Written by AI, not yet AICS's view · claude-haiku-4-5-20251001This is a medical research paper about using deep learning to automatically count eosinophils (a type of white blood cell) in tissue samples from children with ulcerative colitis, a serious digestive disease. The researchers developed an artificial intelligence model to analyze microscope images and tested whether eosinophil counts could predict which children would need stronger medications. While the technical aspects worked well, this is primarily a clinical diagnostic tool study rather than an intervention aimed at improving child well-being outcomes.
Key points
- A deep learning model successfully automated eosinophil detection with high accuracy (AUROC 0.94, F1 score 0.86)
- Lower eosinophil counts predicted which children would need escalation to anti-TNFα therapy
- Higher eosinophil density correlated with corticosteroid-free remission at 1 year
- The model achieved strong agreement with expert pathologist counts (ρ = 0.87-0.89)
- Eosinophil metrics could be incorporated into clinical prediction models for treatment planning
Limitations
This study develops and validates a diagnostic tool rather than testing an intervention to improve child health outcomes. The document does not provide evidence that using this tool actually changes treatment decisions or improves children's health compared to current practice.
Abstract
INTRODUCTION: Emerging evidence implicates eosinophils as important modulators of disease activity and therapeutic response in ulcerative colitis (UC). Automated image analysis provides a scalable and reproducible approach to their evaluation, overcoming the limitations of manual quantification. METHODS: We developed a U-Net-based deep learning model to automate eosinophil detection, using whole-slide images from 2 pediatric sites. Pathologist-annotated 500 × 500-pixel patches served as training data; model performance was evaluated using area under the receiver operator curve (AUROC), precision, recall, and F1 score and benchmarked against SAU-Net, YOLO-v8, and HoVer-Next. The validated algorithm was applied to rectal biopsies from treatment-naïve children with moderate-to-severe UC in the PROTECT inception cohort (n = 221). Eosinophil metrics (density in cells/mm 2 and peak high-power field counts) were analyzed for associations with baseline clinical, histologic, and endoscopic features and 1-year clinical outcomes including anti-tumor necrosis factor α (TNFα) escalation and corticosteroid-free clinical remission. Predictive models were evaluated with logistic regression and machine learning approaches using 5-fold cross-validation. RESULTS: The segmentation model achieved high accuracy (test AUROC = 0.94, F1 score = 0.86) and strong concordance with expert counts (ρ = 0.87-0.89). Higher eosinophil density and counts correlated with basal lymphoid aggregates ( P < 0.001) and predicted week 4 and corticosteroid-free remission ( P = 0.03 and P = 0.01, respectively). Lower eosinophil counts predicted escalation to anti-TNFα therapy ( P = 0.04). Incorporation of eosinophil metrics into Predicting Response to Standardized Pediatric Colitis Therapy (PROTECT) clinical models maintained predictive accuracy for anti-TNFα escalation (AUROC = 0.79; accuracy = 0.76). DISCUSSION: Deep learning-based eosinophil quantification has the potential to transform pediatric UC care by serving as a reproducible biomarker that informs risk stratification and to guide therapy.