Nakeya Dewaswala1, David Chen2, Huzefa Bhopalwala1, Vinod C Kaggal3, Sean P Murphy4, J Martijn Bos1, Jeffrey B Geske1, Bernard J Gersh1, Steve R Ommen1, Philip A Araoz5, Michael J Ackerman1,6,7, Adelaide M Arruda-Olson8. 1. Department of Cardiovascular Medicine, Mayo Clinic Rochester, Rochester, MN, USA. 2. Department of Cardiovascular Surgery, Cleveland Clinic, OH, Cleveland, USA. 3. Enterprise Technology Services, Shared Service Offices, Mayo Clinic, MN, Rochester, USA. 4. Advanced Analytics Services, Mayo Clinic Rochester, Rochester, MN, USA. 5. Department of Radiology, Mayo Clinic Rochester, Rochester, MN, USA. 6. Department of Pediatric and Adolescent Medicine, Mayo Clinic Rochester, Rochester, MN, USA. 7. Department of Molecular Pharmacology and Experimental Therapeutics, Mayo Clinic Rochester, Rochester, MN, USA. 8. Department of Cardiovascular Medicine, Mayo Clinic Rochester, Rochester, MN, USA. ArrudaOlson.Adelaide@mayo.edu.
Abstract
BACKGROUND: Cardiac magnetic resonance (CMR) imaging is important for diagnosis and risk stratification of hypertrophic cardiomyopathy (HCM) patients. However, collection of information from large numbers of CMR reports by manual review is time-consuming, error-prone and costly. Natural language processing (NLP) is an artificial intelligence method for automated extraction of information from narrative text including text in CMR reports in electronic health records (EHR). Our objective was to assess whether NLP can accurately extract diagnosis of HCM from CMR reports. METHODS: An NLP system with two tiers was developed for information extraction from narrative text in CMR reports; the first tier extracted information regarding HCM diagnosis while the second extracted categorical and numeric concepts for HCM classification. We randomly allocated 200 HCM patients with CMR reports from 2004 to 2018 into training (100 patients with 185 CMR reports) and testing sets (100 patients with 206 reports). RESULTS: NLP algorithms demonstrated very high performance compared to manual annotation. The algorithm to extract HCM diagnosis had accuracy of 0.99. The accuracy for categorical concepts included HCM morphologic subtype 0.99, systolic anterior motion of the mitral valve 0.96, mitral regurgitation 0.93, left ventricular (LV) obstruction 0.94, location of obstruction 0.92, apical pouch 0.98, LV delayed enhancement 0.93, left atrial enlargement 0.99 and right atrial enlargement 0.98. Accuracy for numeric concepts included maximal LV wall thickness 0.96, LV mass 0.99, LV mass index 0.98, LV ejection fraction 0.98 and right ventricular ejection fraction 0.99. CONCLUSIONS: NLP identified and classified HCM from CMR narrative text reports with very high performance.
BACKGROUND: Cardiac magnetic resonance (CMR) imaging is important for diagnosis and risk stratification of hypertrophic cardiomyopathy (HCM) patients. However, collection of information from large numbers of CMR reports by manual review is time-consuming, error-prone and costly. Natural language processing (NLP) is an artificial intelligence method for automated extraction of information from narrative text including text in CMR reports in electronic health records (EHR). Our objective was to assess whether NLP can accurately extract diagnosis of HCM from CMR reports. METHODS: An NLP system with two tiers was developed for information extraction from narrative text in CMR reports; the first tier extracted information regarding HCM diagnosis while the second extracted categorical and numeric concepts for HCM classification. We randomly allocated 200 HCM patients with CMR reports from 2004 to 2018 into training (100 patients with 185 CMR reports) and testing sets (100 patients with 206 reports). RESULTS: NLP algorithms demonstrated very high performance compared to manual annotation. The algorithm to extract HCM diagnosis had accuracy of 0.99. The accuracy for categorical concepts included HCM morphologic subtype 0.99, systolic anterior motion of the mitral valve 0.96, mitral regurgitation 0.93, left ventricular (LV) obstruction 0.94, location of obstruction 0.92, apical pouch 0.98, LV delayed enhancement 0.93, left atrial enlargement 0.99 and right atrial enlargement 0.98. Accuracy for numeric concepts included maximal LV wall thickness 0.96, LV mass 0.99, LV mass index 0.98, LV ejection fraction 0.98 and right ventricular ejection fraction 0.99. CONCLUSIONS: NLP identified and classified HCM from CMR narrative text reports with very high performance.
Authors: Chengyi Zheng; Benjamin C Sun; Yi-Lin Wu; Ming-Sum Lee; Ernest Shen; Rita F Redberg; Maros Ferencik; Shaw Natsui; Aniket A Kawatkar; Visanee V Musigdilok; Adam L Sharp Journal: J Am Heart Assoc Date: 2020-02-21 Impact factor: 5.501
Authors: Vinod C Kaggal; Ravikumar Komandur Elayavilli; Saeed Mehrabi; Joshua J Pankratz; Sunghwan Sohn; Yanshan Wang; Dingcheng Li; Majid Mojarad Rastegar; Sean P Murphy; Jason L Ross; Rajeev Chaudhry; James D Buntrock; Hongfang Liu Journal: Biomed Inform Insights Date: 2016-06-23