Phenotype Inference with Semi-Supervised Mixed Membership Models

2018/12/07 by Rodriguez, Victor, Perotte, Adler
#FOS: Biological sciences #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Quantitative Methods (q-bio.QM)

paper · doi:10.48550/arxiv.1812.03222

Abstract

Disease phenotyping algorithms process observational clinical data to identify patients with specific diseases. Supervised phenotyping methods require significant quantities of expert-labeled data, while unsupervised methods may learn non-disease phenotypes. To address these limitations, we propose the Semi-Supervised Mixed Membership Model (SS3M) -- a probabilistic graphical model for learning disease phenotypes from clinical data with relatively few labels. We show SS3M can learn interpretable, disease-specific phenotypes which capture the clinical characteristics of the diseases specified by the labels provided.

Related

• Phenotyping with Positive Unlabelled Learning for Genome-Wide Association Studies
• Computational Phenotype Discovery via Probabilistic Independence
• Phenotype Detection in Real World Data via Online MixEHR Algorithm
• PheME: A deep ensemble framework for improving phenotype prediction from multi-modal data
• Unsupervised Learning for Computational Phenotyping