Mark Hoogendoorn1, Peter Szolovits2, Leon M G Moons3, Mattijs E Numans4. 1. Department of Computer Science, VU University Amsterdam, De Boelelaan 1081, 1081 HV Amsterdam, The Netherlands; Computer Science and Artificial Intelligence Lab, Massachusetts Institute of Technology, 32 Vassar Street, Cambridge, MA 02139, USA. Electronic address: m.hoogendoorn@vu.nl. 2. Computer Science and Artificial Intelligence Lab, Massachusetts Institute of Technology, 32 Vassar Street, Cambridge, MA 02139, USA. Electronic address: psz@mit.edu. 3. Department of Gastroenterology and Hepatology, Utrecht University Medical Center, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands. Electronic address: l.m.g.moons@umcutrecht.nl. 4. Department of Public Health and Primary Care, Leiden University Medical Center, Hippocratespad 21, 2333 ZD Leiden, The Netherlands. Electronic address: m.e.numans@lumc.nl.
Abstract
OBJECTIVE: Machine learning techniques can be used to extract predictive models for diseases from electronic medical records (EMRs). However, the nature of EMRs makes it difficult to apply off-the-shelf machine learning techniques while still exploiting the rich content of the EMRs. In this paper, we explore the usage of a range of natural language processing (NLP) techniques to extract valuable predictors from uncoded consultation notes and study whether they can help to improve predictive performance. METHODS: We study a number of existing techniques for the extraction of predictors from the consultation notes, namely a bag of words based approach and topic modeling. In addition, we develop a dedicated technique to match the uncoded consultation notes with a medical ontology. We apply these techniques as an extension to an existing pipeline to extract predictors from EMRs. We evaluate them in the context of predictive modeling for colorectal cancer (CRC), a disease known to be difficult to diagnose before performing an endoscopy. RESULTS: Our results show that we are able to extract useful information from the consultation notes. The predictive performance of the ontology-based extraction method moves significantly beyond the benchmark of age and gender alone (area under the receiver operating characteristic curve (AUC) of 0.870 versus 0.831). We also observe more accurate predictive models by adding features derived from processing the consultation notes compared to solely using coded data (AUC of 0.896 versus 0.882) although the difference is not significant. The extracted features from the notes are shown be equally predictive (i.e. there is no significant difference in performance) compared to the coded data of the consultations. CONCLUSION: It is possible to extract useful predictors from uncoded consultation notes that improve predictive performance. Techniques linking text to concepts in medical ontologies to derive these predictors are shown to perform best for predicting CRC in our EMR dataset.
OBJECTIVE: Machine learning techniques can be used to extract predictive models for diseases from electronic medical records (EMRs). However, the nature of EMRs makes it difficult to apply off-the-shelf machine learning techniques while still exploiting the rich content of the EMRs. In this paper, we explore the usage of a range of natural language processing (NLP) techniques to extract valuable predictors from uncoded consultation notes and study whether they can help to improve predictive performance. METHODS: We study a number of existing techniques for the extraction of predictors from the consultation notes, namely a bag of words based approach and topic modeling. In addition, we develop a dedicated technique to match the uncoded consultation notes with a medical ontology. We apply these techniques as an extension to an existing pipeline to extract predictors from EMRs. We evaluate them in the context of predictive modeling for colorectal cancer (CRC), a disease known to be difficult to diagnose before performing an endoscopy. RESULTS: Our results show that we are able to extract useful information from the consultation notes. The predictive performance of the ontology-based extraction method moves significantly beyond the benchmark of age and gender alone (area under the receiver operating characteristic curve (AUC) of 0.870 versus 0.831). We also observe more accurate predictive models by adding features derived from processing the consultation notes compared to solely using coded data (AUC of 0.896 versus 0.882) although the difference is not significant. The extracted features from the notes are shown be equally predictive (i.e. there is no significant difference in performance) compared to the coded data of the consultations. CONCLUSION: It is possible to extract useful predictors from uncoded consultation notes that improve predictive performance. Techniques linking text to concepts in medical ontologies to derive these predictors are shown to perform best for predicting CRC in our EMR dataset.
Authors: Katherine P Liao; Tianxi Cai; Vivian Gainer; Sergey Goryachev; Qing Zeng-treitler; Soumya Raychaudhuri; Peter Szolovits; Susanne Churchill; Shawn Murphy; Isaac Kohane; Elizabeth W Karlson; Robert M Plenge Journal: Arthritis Care Res (Hoboken) Date: 2010-08 Impact factor: 4.794
Authors: Antonieta Medina-Lara; Bogdan Grigore; Ruth Lewis; Jaime Peters; Sarah Price; Paolo Landa; Sophie Robinson; Richard Neal; William Hamilton; Anne E Spencer Journal: Health Technol Assess Date: 2020-11 Impact factor: 4.014
Authors: Laura H van Dongen; Peter P Harms; Mark Hoogendoorn; Dominic S Zimmerman; Elisabeth M Lodder; Leen M 't Hart; Ron Herings; Henk C P M van Weert; Giel Nijpels; Karin M A Swart; Amber A van der Heijden; Marieke T Blom; Petra J Elders; Hanno L Tan Journal: Open Heart Date: 2021-02
Authors: Hailun Liang; Lei Yang; Lei Tao; Leiyu Shi; Wuyang Yang; Jiawei Bai; Da Zheng; Ning Wang; Jiafu Ji Journal: Chin J Cancer Res Date: 2020-04 Impact factor: 5.087