Abstract
Enormous amounts of healthcare data are becoming increasingly accessible through the large-scale adoption of electronic health records. In this work, structured and unstructured (textual) data are combined to assign clinical diagnostic and procedural codes (specifically ICD-9-CM) to patient stays. We investigate whether integrating these heterogeneous data types improves prediction strength compared to using the data types in isolation. Two separate data integration approaches were evaluated. Early data integration combines features of several sources within a single model, and late data integration learns a separate model per data source and combines these predictions with a meta-learner. This is evaluated on data sources and clinical codes from a broad set of medical specialties. When compared with the best individual prediction source, late data integration leads to improvements in predictive power (eg, overall F-measure increased from 30.6% to 38.3% for International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnostic codes), while early data integration is less consistent. The predictive strength strongly differs between medical specialties, both for ICD-9-CM diagnostic and procedural co...Continue Reading
References
Aug 10, 2000·Medical Care·A G LawthersL I Iezzoni
Oct 24, 2002·Medical Care·Patrick S RomanoJulie A Rainwater
Aug 27, 2005·IEEE Transactions on Pattern Analysis and Machine Intelligence·Hanchuan PengChris Ding
Jun 27, 2006·Journal of the American Medical Informatics Association : JAMIA·Serguei V S PakhomovChristopher G Chute
May 9, 2008·BMC Bioinformatics·Richárd Farkas, György Szarvas
Sep 20, 2008·BMC Bioinformatics·Leming ShiWeida Tong
Mar 6, 2009·Clinical Otolaryngology : Official Journal of ENT-UK ; Official Journal of Netherlands Society for Oto-Rhino-Laryngology & Cervico-Facial Surgery·S A R NouraeiG S Sandhu
Apr 10, 2009·Genome Medicine·Anneleen DaemenBart De Moor
Jan 1, 2009·Human Genomics and Proteomics : HGP·Jemila S HamidJoseph Beyene
Oct 22, 2010·Journal of the American Medical Informatics Association : JAMIA·Mary H StanfillWilliam R Hersh
Feb 2, 2011·Critical Care Medicine·Mohammed SaeedRoger G Mark
Mar 14, 2013·JAMA : the Journal of the American Medical Association·James J Cimino
Nov 6, 2013·Journal of the American Medical Informatics Association : JAMIA·Jyotishman PathakChristopher G Chute
Dec 4, 2013·Journal of the American Medical Informatics Association : JAMIA·Adler PerotteNoémie Elhadad
Jan 5, 2014·Journal of the American Medical Informatics Association : JAMIA·Swapna AbhyankarClement J McDonald
Apr 16, 2014·Bioinformatics·Robert J CarrollJoshua C Denny
May 30, 2014·Surgical Infections·Rajesh RamanathanTherese M Duane
Dec 9, 2014·Methods : a Companion to Methods in Enzymology·Sune Pletscher-FrankildLars Juhl Jensen
Citations
Jan 24, 2018·Journal of the American Medical Informatics Association : JAMIA·Honghan WuRichard J B Dobson
May 13, 2017·Clinical Journal of the American Society of Nephrology : CJASN·Anne-Sophie JannotNicolas Pallet
Jun 13, 2017·Journal of Biomedical Informatics·Elyne ScheurwegsKim Luyckx
May 26, 2021·Health Care Management Science·José Carlos FerrãoHenrique M G Martins
Mar 1, 2020·Health Systems·José Carlos FerrãoDaniel Gartner
Jul 27, 2021·Journal of Healthcare Engineering·Ayoub BagheriDaniel L Oberski
Apr 28, 2017·Mayo Clinic Proceedings. Innovations, Quality & Outcomes·Sudhi G UpadhyayaDaryl J Kor
Nov 4, 2019··Yun XiongYangyong Zhu