Predicting cardiovascular health trajectories in time-series electronic health records with LSTM models

Medicine & Life Sciences

• Long-Term Memory 99%
• Electronic Health Records 88%
• Short-Term Memory 81%
• Health 42%
• Health Status 33%
• Cardiovascular Diseases 20%
• LDL Lipoproteins 19%
• Population Health Management 15%
• Smoking 13%
• Body Mass Index 13%
• Blood Pressure 12%
• Ambulatory Care Facilities 9%
• Health Care Costs 8%
• Hemoglobins 7%
• Cause of Death 7%
• Datasets 7%
• Quality of Life 6%
• Guidelines 6%
• Neoplasms 3%

Engineering & Materials Science

• Long short-term memory 100%
• Time series 74%
• Health 63%
• Trajectories 59%
• Lipoproteins 17%
• Blood pressure 11%
• Hemoglobin 7%
• Costs 1%

Social Sciences

• time series 91%
• electronics 60%
• health 39%
• cardiovascular disease 20%
• blood pressure 13%
• smoking 10%
• recipient 9%
• outpatient clinic 7%
• cause of death 6%
• health promotion 5%
• performance 4%
• cancer 4%
• quality of life 4%
• costs 3%