Tomer Czyzewski1, Nati Daniel1, Mark Rochman2, Julie M Caldwell2, Garrett A Osswald2, Margaret H Collins3, Marc E Rothenberg2, Yonatan Savir1. 1. Department of Physiology, Biophysics, and System Biology, Faculty of Medicine, Technion Israel Institute of Technology, Haifa 35254, Israel. 2. Division of Allergy, and Immunology, Cincinnati Children's Hospital Medical Center, Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH 45229 USA. 3. Division of Pathology, Cincinnati Children's Hospital Medical Center, Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH 45229 USA.
Abstract
GOAL: Eosinophilic esophagitis (EoE) is an allergic inflammatory condition characterized by eosinophil accumulation in the esophageal mucosa. EoE diagnosis includes a manual assessment of eosinophil levels in mucosal biopsies-a time-consuming, laborious task that is difficult to standardize. One of the main challenges in automating this process, like many other biopsy-based diagnostics, is detecting features that are small relative to the size of the biopsy. RESULTS: In this work, we utilized hematoxylin- and eosin-stained slides from esophageal biopsies from patients with active EoE and control subjects to develop a platform based on a deep convolutional neural network (DCNN) that can classify esophageal biopsies with an accuracy of 85%, sensitivity of 82.5%, and specificity of 87%. Moreover, by combining several downscaling and cropping strategies, we show that some of the features contributing to the correct classification are global rather than specific, local features. CONCLUSIONS: We report the ability of artificial intelligence to identify EoE using computer vision analysis of esophageal biopsy slides. Further, the DCNN features associated with EoE are based on not only local eosinophils but also global histologic changes. Our approach can be used for other conditions that rely on biopsy-based histologic diagnostics.
GOAL: Eosinophilic esophagitis (EoE) is an allergic inflammatory condition characterized by eosinophil accumulation in the esophageal mucosa. EoE diagnosis includes a manual assessment of eosinophil levels in mucosal biopsies-a time-consuming, laborious task that is difficult to standardize. One of the main challenges in automating this process, like many other biopsy-based diagnostics, is detecting features that are small relative to the size of the biopsy. RESULTS: In this work, we utilized hematoxylin- and eosin-stained slides from esophageal biopsies from patients with active EoE and control subjects to develop a platform based on a deep convolutional neural network (DCNN) that can classify esophageal biopsies with an accuracy of 85%, sensitivity of 82.5%, and specificity of 87%. Moreover, by combining several downscaling and cropping strategies, we show that some of the features contributing to the correct classification are global rather than specific, local features. CONCLUSIONS: We report the ability of artificial intelligence to identify EoE using computer vision analysis of esophageal biopsy slides. Further, the DCNN features associated with EoE are based on not only local eosinophils but also global histologic changes. Our approach can be used for other conditions that rely on biopsy-based histologic diagnostics.
Entities:
Keywords:
Decision support system; deep convolutional network; digital pathology; eosinophilic esophagitis; small features detection.
Authors: Evan S Dellon; Chris A Liacouras; Javier Molina-Infante; Glenn T Furuta; Jonathan M Spergel; Noam Zevit; Stuart J Spechler; Stephen E Attwood; Alex Straumann; Seema S Aceves; Jeffrey A Alexander; Dan Atkins; Nicoleta C Arva; Carine Blanchard; Peter A Bonis; Wendy M Book; Kelley E Capocelli; Mirna Chehade; Edaire Cheng; Margaret H Collins; Carla M Davis; Jorge A Dias; Carlo Di Lorenzo; Ranjan Dohil; Christophe Dupont; Gary W Falk; Cristina T Ferreira; Adam Fox; Nirmala P Gonsalves; Sandeep K Gupta; David A Katzka; Yoshikazu Kinoshita; Calies Menard-Katcher; Ellyn Kodroff; David C Metz; Stephan Miehlke; Amanda B Muir; Vincent A Mukkada; Simon Murch; Samuel Nurko; Yoshikazu Ohtsuka; Rok Orel; Alexandra Papadopoulou; Kathryn A Peterson; Hamish Philpott; Philip E Putnam; Joel E Richter; Rachel Rosen; Marc E Rothenberg; Alain Schoepfer; Melissa M Scott; Neil Shah; Javed Sheikh; Rhonda F Souza; Mary J Strobel; Nicholas J Talley; Michael F Vaezi; Yvan Vandenplas; Mario C Vieira; Marjorie M Walker; Joshua B Wechsler; Barry K Wershil; Ting Wen; Guang-Yu Yang; Ikuo Hirano; Albert J Bredenoord Journal: Gastroenterology Date: 2018-09-06 Impact factor: 22.682
Authors: Emily M Stucke; Katherine E Clarridge; Margaret H Collins; Carol J Henderson; Lisa J Martin; Marc E Rothenberg Journal: J Pediatr Gastroenterol Nutr Date: 2015-07 Impact factor: 2.839
Authors: Kelly M O'Shea; Seema S Aceves; Evan S Dellon; Sandeep K Gupta; Jonathan M Spergel; Glenn T Furuta; Marc E Rothenberg Journal: Gastroenterology Date: 2017-07-27 Impact factor: 22.682
Authors: M H Collins; L J Martin; E S Alexander; J Todd Boyd; R Sheridan; H He; S Pentiuk; P E Putnam; J P Abonia; V A Mukkada; J P Franciosi; M E Rothenberg Journal: Dis Esophagus Date: 2017-02-01 Impact factor: 3.429
Authors: Evan S Dellon; Karen J Fritchie; Tara C Rubinas; John T Woosley; Nicholas J Shaheen Journal: Dig Dis Sci Date: 2009-10-15 Impact factor: 3.199
Authors: Nati Daniel; Ariel Larey; Eliel Aknin; Garrett A Osswald; Julie M Caldwell; Mark Rochman; Margaret H Collins; Guang-Yu Yang; Nicoleta C Arva; Kelley E Capocelli; Marc E Rothenberg; Yonatan Savir Journal: Annu Int Conf IEEE Eng Med Biol Soc Date: 2022-07
Authors: Marc E Rothenberg; Shawna K B Hottinger; Nirmala Gonsalves; Glenn T Furuta; Margaret H Collins; Nicholas J Talley; Kathryn Peterson; Calies Menard-Katcher; Macie Smith; Ikuo Hirano; Robert M Genta; Mirna Chehade; Sandeep K Gupta; Jonathan M Spergel; Seema S Aceves; Evan S Dellon Journal: J Allergy Clin Immunol Date: 2021-12-22 Impact factor: 10.793