Iolanda Valentina Popa1, Mircea Diculescu2, Catalina Mihai1, Cristina Cijevschi Prelipcean1, Alexandru Burlacu3. 1. University of Medicine and Pharmacy "Gr. T. Popa", Iași, Romania;Institute of Gastroenterology and Hepatology, Iași, Romania. 2. Department of Gastroenterology, Fundeni Clinical Institute, Bucharest, Romania;Carol Davila' University of Medicine and Pharmacy, Bucharest, Romania. 3. University of Medicine and Pharmacy "Gr. T. Popa", Iași, Romania; Department of Interventional Cardiology, Cardiovascular Diseases Institute, Iași, Romania;Romanian Academy of Medical Sciences, Bucharest, Romania.
Abstract
BACKGROUND: Colonoscopy with biopsy is the "gold" standard for evaluating disease activity in inflammatory bowel diseases (IBD). Current research is geared toward finding non-invasive, cost-efficient methods that estimate disease activity. We aimed to develop a neural network (NN) model for the non-invasive prediction of histologic activity in IBD using routinely available clinical-biological parameters. METHODS: Standard clinical-biological parameters and histologic activity from 371 ulcerative colitis (UC) and 115 Crohn's disease (CD) patient records were collected. A training set, a test set, and a validation set were used for building/validating 2 models for each disease. All models had binary output predicting the active/inactive histologic disease status. For both diseases, the first model used both clinical and biological inputs, while the second used only biological data. RESULTS: First UC model obtained an accuracy of 95.59% on the test set and 96.67% on the validation set. The second UC model achieved accuracies of 88.24% and 86.67% on the test and validation sets, respectively. The First CD classifier resulted in 90.48% accuracy on the test set and 91.67% on the validation set. Finally, the second CD classifier obtained an accuracy of 85.71% on the test set and 91.67% on the validation set. CONCLUSIONS: An accurate and non-invasive artificial intelligence system to predict histologic disease activity in IBD is designed. Our models achieved similar or better results compared to the documented performance of fecal calprotectin (the best non-invasive IBD biomarker to date). Given these favorable results, we anticipate the future utility in the clinical setting of a non-invasive disease activity prediction.
BACKGROUND: Colonoscopy with biopsy is the "gold" standard for evaluating disease activity in inflammatory bowel diseases (IBD). Current research is geared toward finding non-invasive, cost-efficient methods that estimate disease activity. We aimed to develop a neural network (NN) model for the non-invasive prediction of histologic activity in IBD using routinely available clinical-biological parameters. METHODS: Standard clinical-biological parameters and histologic activity from 371 ulcerative colitis (UC) and 115 Crohn's disease (CD) patient records were collected. A training set, a test set, and a validation set were used for building/validating 2 models for each disease. All models had binary output predicting the active/inactive histologic disease status. For both diseases, the first model used both clinical and biological inputs, while the second used only biological data. RESULTS: First UC model obtained an accuracy of 95.59% on the test set and 96.67% on the validation set. The second UC model achieved accuracies of 88.24% and 86.67% on the test and validation sets, respectively. The First CD classifier resulted in 90.48% accuracy on the test set and 91.67% on the validation set. Finally, the second CD classifier obtained an accuracy of 85.71% on the test set and 91.67% on the validation set. CONCLUSIONS: An accurate and non-invasive artificial intelligence system to predict histologic disease activity in IBD is designed. Our models achieved similar or better results compared to the documented performance of fecal calprotectin (the best non-invasive IBD biomarker to date). Given these favorable results, we anticipate the future utility in the clinical setting of a non-invasive disease activity prediction.
Authors: Parambir S Dulai; Laurent Peyrin-Biroulet; Silvio Danese; Bruce E Sands; Axel Dignass; Dan Turner; Gerassimos Mantzaris; Juergen Schölmerich; Jean-Yves Mary; Walter Reinisch; William J Sandborn Journal: Gastroenterology Date: 2019-06-19 Impact factor: 22.682
Authors: Christian Maaser; Andreas Sturm; Stephan R Vavricka; Torsten Kucharzik; Gionata Fiorino; Vito Annese; Emma Calabrese; Daniel C Baumgart; Dominik Bettenworth; Paula Borralho Nunes; Johan Burisch; Fabiana Castiglione; Rami Eliakim; Pierre Ellul; Yago González-Lama; Hannah Gordon; Steve Halligan; Konstantinos Katsanos; Uri Kopylov; Paulo G Kotze; Eduards Krustinš; Andrea Laghi; Jimmy K Limdi; Florian Rieder; Jordi Rimola; Stuart A Taylor; Damian Tolan; Patrick van Rheenen; Bram Verstockt; Jaap Stoker Journal: J Crohns Colitis Date: 2019-02-01 Impact factor: 9.071
Authors: T Sipponen; P Kärkkäinen; E Savilahti; K-L Kolho; H Nuutinen; U Turunen; M Färkkilä Journal: Aliment Pharmacol Ther Date: 2008-08-26 Impact factor: 8.171