Diagnosis of Celiac Disease and Environmental Enteropathy on Biopsy Images Using Color Balancing on Convolutional Neural Networks

摘要: Celiac Disease (CD) and Environmental Enteropathy (EE) are common causes of malnutrition adversely impact normal childhood development. CD is an autoimmune disorder that prevalent worldwide caused by increased sensitivity to gluten. Gluten exposure destructs the small intestinal epithelial barrier, resulting in nutrient mal-absorption under-nutrition. EE also results barrier dysfunction but thought be vulnerability infections. has been implicated as predominant cause under-nutrition, oral vaccine failure, impaired cognitive development low-and-middle-income countries. Both conditions require a tissue biopsy for diagnosis, major challenge interpreting clinical images differentiate between these gastrointestinal diseases striking histopathologic overlap them. In current study, we propose convolutional neural network (CNN) classify duodenal from subjects with CD, EE, healthy controls. We evaluated performance our proposed model using large cohort containing 1000 images. Our evaluations show achieves area under ROC 0.99, 1.00, 0.97 controls, respectively. These demonstrate discriminative power biopsies classification.