{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,6]],"date-time":"2026-03-06T12:26:45Z","timestamp":1772800005010,"version":"3.50.1"},"reference-count":34,"publisher":"Oxford University Press (OUP)","issue":"9","license":[{"start":{"date-parts":[[2020,9,1]],"date-time":"2020-09-01T00:00:00Z","timestamp":1598918400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"funder":[{"name":"King Abdullah University of Science and Technology Center Partnership Fund"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2020,9,1]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Objective<\/jats:title><jats:p>We sought to predict if patients with type 2 diabetes mellitus (DM2) would develop 10 selected complications. Accurate prediction of complications could help with more targeted measures that would prevent or slow down their development.<\/jats:p><\/jats:sec><jats:sec><jats:title>Materials and Methods<\/jats:title><jats:p>Experiments were conducted on the Healthcare Cost and Utilization Project State Inpatient Databases of California for the period of 2003 to 2011. Recurrent neural network (RNN) long short-term memory (LSTM) and RNN gated recurrent unit (GRU) deep learning methods were designed and compared with random forest and multilayer perceptron traditional models. Prediction accuracy of selected complications were compared on 3 settings corresponding to minimum number of hospitalizations between diabetes diagnosis and the diagnosis of complications.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The diagnosis domain was used for experiments. The best results were achieved with RNN GRU model, followed by RNN LSTM model. The prediction accuracy achieved with RNN GRU model was between 73% (myocardial infarction) and 83% (chronic ischemic heart disease), while accuracy of traditional models was between 66% \u2013 76%.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>The number of hospitalizations was an important factor for the prediction accuracy. Experiments with 4 hospitalizations achieved significantly better accuracy than with 2 hospitalizations. To achieve improved accuracy deep learning models required training on at least 1000 patients and accuracy significantly dropped if training datasets contained 500 patients. The prediction accuracy of complications decreases over time period. Considering individual complications, the best accuracy was achieved on depressive disorder and chronic ischemic heart disease.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>The RNN GRU model was the best choice for electronic medical record type of data, based on the achieved results.<\/jats:p><\/jats:sec>","DOI":"10.1093\/jamia\/ocaa120","type":"journal-article","created":{"date-parts":[[2020,5,28]],"date-time":"2020-05-28T11:09:18Z","timestamp":1590664158000},"page":"1343-1351","source":"Crossref","is-referenced-by-count":65,"title":["Predicting complications of diabetes mellitus using advanced machine learning algorithms"],"prefix":"10.1093","volume":"27","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3287-3741","authenticated-orcid":false,"given":"Branimir","family":"Ljubic","sequence":"first","affiliation":[{"name":"Center for Data Analytics and Biomedical Informatics, Temple University, Philadelphia, Pennsylvania, USA"}]},{"given":"Ameen Abdel","family":"Hai","sequence":"additional","affiliation":[{"name":"Center for Data Analytics and Biomedical Informatics, Temple University, Philadelphia, Pennsylvania, USA"}]},{"given":"Marija","family":"Stanojevic","sequence":"additional","affiliation":[{"name":"Center for Data Analytics and Biomedical Informatics, Temple University, Philadelphia, Pennsylvania, USA"}]},{"given":"Wilson","family":"Diaz","sequence":"additional","affiliation":[{"name":"Center for Data Analytics and Biomedical Informatics, Temple University, Philadelphia, Pennsylvania, USA"}]},{"given":"Daniel","family":"Polimac","sequence":"additional","affiliation":[{"name":"Center for Data Analytics and Biomedical Informatics, Temple University, Philadelphia, Pennsylvania, USA"}]},{"given":"Martin","family":"Pavlovski","sequence":"additional","affiliation":[{"name":"Center for Data Analytics and Biomedical Informatics, Temple University, Philadelphia, Pennsylvania, USA"}]},{"given":"Zoran","family":"Obradovic","sequence":"additional","affiliation":[{"name":"Center for Data Analytics and Biomedical Informatics, Temple University, Philadelphia, Pennsylvania, USA"}]}],"member":"286","published-online":{"date-parts":[[2020,9,1]]},"reference":[{"issue":"5","key":"2020110613083428200_ocaa120-B1","doi-asserted-by":"crossref","first-page":"S35","DOI":"10.1016\/j.jacc.2008.09.055","article-title":"The pathologic continuum of diabetic vascular disease","volume":"53","author":"Orasanu","year":"2009","journal-title":"J Am Coll Cardiol"},{"issue":"2","key":"2020110613083428200_ocaa120-B2","doi-asserted-by":"crossref","first-page":"105","DOI":"10.3909\/ricm0742","article-title":"Management of patients with stable angina and type 2 diabetes","volume":"16","author":"Deedwania","year":"2015","journal-title":"Rev Cardiovasc Med"},{"issue":"14","key":"2020110613083428200_ocaa120-B3","doi-asserted-by":"crossref","first-page":"e93751","DOI":"10.1172\/jci.insight.93751","article-title":"Diabetic retinopathy: current understanding, mechanisms, and treatment strategies","volume":"2","author":"Duh","year":"2017","journal-title":"JCI Insight"},{"key":"2020110613083428200_ocaa120-B4","author":"Centers for Disease Control and Prevention"},{"key":"2020110613083428200_ocaa120-B5","author":"World Health Organzation. Diabetes"},{"key":"2020110613083428200_ocaa120-B6","author":"National Institute of Diabetes and Digestive and Kidney Diseases"},{"key":"2020110613083428200_ocaa120-B7","doi-asserted-by":"crossref","first-page":"l5887","DOI":"10.1136\/bmj.l5887","article-title":"Benefits and harms of intensive glycemic control in patients with type 2 diabetes","volume":"367","author":"Rodriguez-Gutierrez","year":"2019","journal-title":"BMJ"},{"issue":"1","key":"2020110613083428200_ocaa120-B8","doi-asserted-by":"crossref","first-page":"91","DOI":"10.4158\/CS-2017-0153","article-title":"Consensus statement by the American Association of Clinical Endocrinologists and American College of Endocrinology on the comprehensive type 2 Diabetes management Algorithm - 2018 Executive summary","volume":"24","author":"Garber","year":"2018","journal-title":"Endocr Pract"},{"issue":"4","key":"2020110613083428200_ocaa120-B9","doi-asserted-by":"crossref","first-page":"273","DOI":"10.14797\/mdcj-14-4-273","article-title":"Update on management of type 2 diabetes for cardiologists","volume":"14","author":"Qureshi","year":"2018","journal-title":"Methodist Debakey Cardiovasc J"},{"issue":"1","key":"2020110613083428200_ocaa120-B10","doi-asserted-by":"crossref","first-page":"123","DOI":"10.1093\/brimed\/ldy013","article-title":"Treatment of type 2 diabetes: future approaches","volume":"126","author":"Bailey","year":"2018","journal-title":"Br Med Bull"},{"key":"2020110613083428200_ocaa120-B11","article-title":"Use of a Machine Learning Algorithm Improves Prediction of Progression to Diabetes","volume":"67 (suppl 1","author":"Cahn","year":"2018","journal-title":"Diabetes"},{"issue":"5","key":"2020110613083428200_ocaa120-B12","doi-asserted-by":"crossref","first-page":"e10775","DOI":"10.2196\/10775","article-title":"Artificial intelligence for diabetes management and decision support: literature review","volume":"20","author":"Contreras","year":"2018","journal-title":"J Med Internet Res"},{"key":"2020110613083428200_ocaa120-B13","volume-title":"Diabetic Neuropathy","author":"Bodman","year":"2020"},{"issue":"6","key":"2020110613083428200_ocaa120-B14","doi-asserted-by":"crossref","first-page":"521","DOI":"10.1016\/S1474-4422(12)70065-0","article-title":"Diabetic neuropathy: clinical manifestations and current treatments","volume":"11","author":"Callaghan","year":"2012","journal-title":"Lancet Neurol"},{"issue":"5","key":"2020110613083428200_ocaa120-B15","first-page":"390","article-title":"Diabetic kidney disease","volume":"113","author":"Bouhairie","year":"2016","journal-title":"Mo Med"},{"issue":"6","key":"2020110613083428200_ocaa120-B16","doi-asserted-by":"crossref","first-page":"491","DOI":"10.1007\/s11892-014-0491-3","article-title":"Diabetes and depression","volume":"14","author":"Holt","year":"2014","journal-title":"Curr Diab Rep"},{"issue":"4","key":"2020110613083428200_ocaa120-B17","doi-asserted-by":"crossref","first-page":"381","DOI":"10.1097\/AUD.0000000000000137","article-title":"Hearing impairment in relation to severity of diabetes in a Veteran cohort","volume":"36","author":"Konrad-Martin","year":"2015","journal-title":"Ear Hear"},{"issue":"13","key":"2020110613083428200_ocaa120-B18","doi-asserted-by":"crossref","first-page":"1246","DOI":"10.4239\/wjd.v6.i13.1246","article-title":"Diabetes and cardiovascular disease: epidemiology, biological mechanisms, treatment recommendations and future research","volume":"6","author":"Leon","year":"2015","journal-title":"World J Diabetes"},{"issue":"2","key":"2020110613083428200_ocaa120-B19","doi-asserted-by":"crossref","first-page":"295","DOI":"10.1177\/1932296817706375","article-title":"Machine learning methods to predict diabetes complications","volume":"12","author":"Dagliati","year":"2018","journal-title":"J Diabetes Sci Technol"},{"key":"2020110613083428200_ocaa120-B20","article-title":"Predicting future glucose fluctuations using machine learning and wearable sensor data","volume":"67 (suppl 1","author":"Hayeri","year":"2018","journal-title":"Diabetes"},{"issue":"17","key":"2020110613083428200_ocaa120-B21","doi-asserted-by":"crossref","first-page":"3532","DOI":"10.3390\/app9173532","article-title":"LSTM DSS automatism and dataset optimization for diabetes prediction","volume":"9","author":"Massaro","year":"2019","journal-title":"Appl Sci"},{"key":"2020110613083428200_ocaa120-B22","author":"Perveen"},{"key":"2020110613083428200_ocaa120-B23","first-page":"364","volume-title":"Computational Vision and Bio-Inspired Computing. ICCVBIC 2019. Advances in Intelligent Systems and Computing","author":"Apoorva","year":"2020"},{"key":"2020110613083428200_ocaa120-B24","author":"Sarwar"},{"key":"2020110613083428200_ocaa120-B25","first-page":"2487","author":"Zhang"},{"issue":"3","key":"2020110613083428200_ocaa120-B26","doi-asserted-by":"crossref","first-page":"293","DOI":"10.1001\/jamainternmed.2018.7117","article-title":"Deep learning in medicine\u2014promise, progress, and challenges","volume":"179","author":"Wang","year":"2019","journal-title":"JAMA Intern Med"},{"key":"2020110613083428200_ocaa120-B27","doi-asserted-by":"crossref","first-page":"54","DOI":"10.1136\/emj.20.1.54","article-title":"Observational research methods. Research design II: cohort, cross sectional, and case-control studies","volume":"20","author":"Man","year":"2003","journal-title":"Emerg Med J"},{"issue":"2","key":"2020110613083428200_ocaa120-B28","doi-asserted-by":"crossref","first-page":"164","DOI":"10.1109\/TAC.1980.1102314","article-title":"The singular value decomposition: its computation and some applications","volume":"25","author":"Klema","year":"1980","journal-title":"IEEE Trans Automat Contr"},{"issue":"2","key":"2020110613083428200_ocaa120-B29","doi-asserted-by":"crossref","first-page":"270","DOI":"10.1162\/neco.1989.1.2.270","article-title":"A learning algorithm for continually running fully recurrent neural networks","volume":"1","author":"Williams","year":"1989","journal-title":"Neural Comput"},{"key":"2020110613083428200_ocaa120-B30","first-page":"1724","author":"Cho","year":"2014"},{"issue":"1","key":"2020110613083428200_ocaa120-B31","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Machine Learning"},{"key":"2020110613083428200_ocaa120-B32","doi-asserted-by":"crossref","first-page":"318","DOI":"10.7551\/mitpress\/5236.001.0001","volume-title":"Parallel Distributed Processing","author":"Rumelhart","year":"1986"},{"key":"2020110613083428200_ocaa120-B33","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1016\/j.jbi.2018.09.001","article-title":"Mixed Effect Machine Learning: a framework for predicting longitudinal change in hemoglobin A1c","volume":"89","author":"Ngufor","year":"2019","journal-title":"J Biomed Inform"},{"issue":"2","key":"2020110613083428200_ocaa120-B34","doi-asserted-by":"crossref","first-page":"361","DOI":"10.1093\/jamia\/ocw112","article-title":"Using recurrent neural network models for early detection of heart failure onset","volume":"24","author":"Choi","year":"2017","journal-title":"J Am Med Inform Assoc"}],"container-title":["Journal of the American Medical Informatics Association"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/academic.oup.com\/jamia\/article-pdf\/27\/9\/1343\/34153184\/ocaa120.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"http:\/\/academic.oup.com\/jamia\/article-pdf\/27\/9\/1343\/34153184\/ocaa120.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,10,25]],"date-time":"2022-10-25T10:06:10Z","timestamp":1666692370000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jamia\/article\/27\/9\/1343\/5899846"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,9,1]]},"references-count":34,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2020,9,1]]},"published-print":{"date-parts":[[2020,9,1]]}},"URL":"https:\/\/doi.org\/10.1093\/jamia\/ocaa120","relation":{},"ISSN":["1067-5027","1527-974X"],"issn-type":[{"value":"1067-5027","type":"print"},{"value":"1527-974X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2020,9]]},"published":{"date-parts":[[2020,9,1]]}}}