Electronic Health Record (EHR) phenotypes & genotypes extraction

   March 01, 2019 - June 30, 2019

   Taipei, Taiwan

   College of Medicine AI team, National Taiwan University

   Website / Database


     Yun-Cheng Tsai  


Wang Zhiming and I developed this text-mining website to help the NTU Medical Genie AI team to get the genotypes information. There are two main steps. In the first step, we used MetaMap and Clinphen to extract phenotype information from electronic health record (EHR); in the second step, we used Phenolyzer and Variant Prioritizer to convert selected phenotypes to genotype information. The website was developed by Python Django framework.

We cooperated with Doctor Wuh-Liang ​​Hwu who evalutated the extracted results to help us improve the phenotypes and genotypes selection accuracy.

» Website Entry «

  Demo video

The video below shows the basic functions of this website. We input the EHR inside the text box and press the search button. The web server will run MetaMap and do natural-language processing (NLP) to extract phenotype infromation from the EHR. Six categories of phenotypes will be highlighted in different colors.