Eric D Adler1, Adriaan A Voors2, Liviu Klein3, Fima Macheret4, Oscar O Braun5, Marcus A Urey1, Wenhong Zhu4, Iziah Sama2, Matevz Tadel6, Claudio Campagnari7, Barry Greenberg1, Avi Yagil1,6. 1. Division of Cardiology, Department of Medicine, UC San Diego, La Jolla, CA, USA. 2. University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. 3. Division of Cardiology, Department of Medicine, UC San Francisco, San Francisco, CA, USA. 4. Altman Clinical and Translational Research Institute (ACTRI), UC San Diego, La Jolla, CA, USA. 5. Cardiology, Department of Clinical Sciences, Lund University and Skåne University Hospital, Lund, Sweden. 6. Physics Department, UC San Diego, La Jolla, CA, USA. 7. Physics Department, UC Santa Barbara, Santa Barbara, CA, USA.
Abstract
BACKGROUND: Predicting mortality is important in patients with heart failure (HF). However, current strategies for predicting risk are only modestly successful, likely because they are derived from statistical analysis methods that fail to capture prognostic information in large data sets containing multi-dimensional interactions. METHODS AND RESULTS: We used a machine learning algorithm to capture correlations between patient characteristics and mortality. A model was built by training a boosted decision tree algorithm to relate a subset of the patient data with a very high or very low mortality risk in a cohort of 5822 hospitalized and ambulatory patients with HF. From this model we derived a risk score that accurately discriminated between low and high-risk of death by identifying eight variables (diastolic blood pressure, creatinine, blood urea nitrogen, haemoglobin, white blood cell count, platelets, albumin, and red blood cell distribution width). This risk score had an area under the curve (AUC) of 0.88 and was predictive across the full spectrum of risk. External validation in two separate HF populations gave AUCs of 0.84 and 0.81, which were superior to those obtained with two available risk scores in these same populations. CONCLUSIONS: Using machine learning and readily available variables, we generated and validated a mortality risk score in patients with HF that was more accurate than other risk scores to which it was compared. These results support the use of this machine learning approach for the evaluation of patients with HF and in other settings where predicting risk has been challenging.
BACKGROUND: Predicting mortality is important in patients with heart failure (HF). However, current strategies for predicting risk are only modestly successful, likely because they are derived from statistical analysis methods that fail to capture prognostic information in large data sets containing multi-dimensional interactions. METHODS AND RESULTS: We used a machine learning algorithm to capture correlations between patient characteristics and mortality. A model was built by training a boosted decision tree algorithm to relate a subset of the patient data with a very high or very low mortality risk in a cohort of 5822 hospitalized and ambulatory patients with HF. From this model we derived a risk score that accurately discriminated between low and high-risk of death by identifying eight variables (diastolic blood pressure, creatinine, blood ureanitrogen, haemoglobin, white blood cell count, platelets, albumin, and red blood cell distribution width). This risk score had an area under the curve (AUC) of 0.88 and was predictive across the full spectrum of risk. External validation in two separate HF populations gave AUCs of 0.84 and 0.81, which were superior to those obtained with two available risk scores in these same populations. CONCLUSIONS: Using machine learning and readily available variables, we generated and validated a mortality risk score in patients with HF that was more accurate than other risk scores to which it was compared. These results support the use of this machine learning approach for the evaluation of patients with HF and in other settings where predicting risk has been challenging.
Authors: Geoffrey H Tison; Robert Avram; Gregory Nah; Liviu Klein; Barbara V Howard; Matthew A Allison; Ramon Casanova; Rachael H Blair; Khadijah Breathett; Randi E Foraker; Jeffrey E Olgin; Nisha I Parikh Journal: Can J Cardiol Date: 2021-08-13 Impact factor: 5.223
Authors: Michael F Gensheimer; Balasubramanian Narasimhan; A Solomon Henry; Douglas J Wood; Daniel L Rubin Journal: JCO Clin Cancer Inform Date: 2022-06
Authors: Karola S Jering; Claudio Campagnari; Brian Claggett; Eric Adler; Liviu Klein; Faraz S Ahmad; Adriaan A Voors; Scott Solomon; Avi Yagil; Barry Greenberg Journal: Eur J Heart Fail Date: 2022-05-22 Impact factor: 17.349
Authors: Arjun Sinha; Deepak K Gupta; Clyde W Yancy; Sanjiv J Shah; Laura J Rasmussen-Torvik; Elizabeth M McNally; Philip Greenland; Donald M Lloyd-Jones; Sadiya S Khan Journal: Circ Heart Fail Date: 2021-02-04 Impact factor: 8.790
Authors: Veli K Topkara; Pierre Elias; Rashmi Jain; Gabriel Sayer; Daniel Burkhoff; Nir Uriel Journal: Circ Heart Fail Date: 2021-12-24 Impact factor: 8.790