Cristina Soguero-Ruiz1, Kristian Hindberg2, Inmaculada Mora-Jiménez3, José Luis Rojo-Álvarez3, Stein Olav Skrøvseth4, Fred Godtliebsen2, Kim Mortensen5, Arthur Revhaug6, Rolv-Ole Lindsetmo5, Knut Magne Augestad7, Robert Jenssen8. 1. Dept. of Signal Theory and Communications, Telematics and Computing, Universidad Rey Juan Carlos, Fuenlabrada, Spain. Electronic address: cristina.soguero@urjc.es. 2. Dept. Mathematics and Statistics, University of Tromsø (UiT), Tromsø, Norway. 3. Dept. of Signal Theory and Communications, Telematics and Computing, Universidad Rey Juan Carlos, Fuenlabrada, Spain. 4. Norwegian Centre for Integrated Care and Telemedicine, Norway; University Hospital of North Norway (UNN), Norway; IBM T.J. Watson Research Center, Yorktown Heights, NY, USA. 5. Dept. of Gastrointestinal Surgery, UNN, Tromsø, Norway; Institute of Clinical Medicine, UiT, Tromsø, Norway. 6. Dept. of Gastrointestinal Surgery, UNN, Tromsø, Norway; Clinic for Surgery, Cancer and Women's Health, UNN, Tromsø, Norway. 7. Norwegian Centre for Integrated Care and Telemedicine, Norway; Dept. of Surgery, Hammerfest Hospital, Norway; Dept. of Colorectal Surgery, University Hospitals Case Medical Center, Cleveland, USA; Institute of Clinical Medicine, UiT, Tromsø, Norway. 8. Norwegian Centre for Integrated Care and Telemedicine, Norway; Dept. of Physics and Technology, UiT, Tromsø, Norway.
Abstract
OBJECTIVE: In this work, we have developed a learning system capable of exploiting information conveyed by longitudinal Electronic Health Records (EHRs) for the prediction of a common postoperative complication, Anastomosis Leakage (AL), in a data-driven way and by fusing temporal population data from different and heterogeneous sources in the EHRs. MATERIAL AND METHODS: We used linear and non-linear kernel methods individually for each data source, and leveraging the powerful multiple kernels for their effective combination. To validate the system, we used data from the EHR of the gastrointestinal department at a university hospital. RESULTS: We first investigated the early prediction performance from each data source separately, by computing Area Under the Curve values for processed free text (0.83), blood tests (0.74), and vital signs (0.65), respectively. When exploiting the heterogeneous data sources combined using the composite kernel framework, the prediction capabilities increased considerably (0.92). Finally, posterior probabilities were evaluated for risk assessment of patients as an aid for clinicians to raise alertness at an early stage, in order to act promptly for avoiding AL complications. DISCUSSION: Machine-learning statistical model from EHR data can be useful to predict surgical complications. The combination of EHR extracted free text, blood samples values, and patient vital signs, improves the model performance. These results can be used as a framework for preoperative clinical decision support.
OBJECTIVE: In this work, we have developed a learning system capable of exploiting information conveyed by longitudinal Electronic Health Records (EHRs) for the prediction of a common postoperative complication, Anastomosis Leakage (AL), in a data-driven way and by fusing temporal population data from different and heterogeneous sources in the EHRs. MATERIAL AND METHODS: We used linear and non-linear kernel methods individually for each data source, and leveraging the powerful multiple kernels for their effective combination. To validate the system, we used data from the EHR of the gastrointestinal department at a university hospital. RESULTS: We first investigated the early prediction performance from each data source separately, by computing Area Under the Curve values for processed free text (0.83), blood tests (0.74), and vital signs (0.65), respectively. When exploiting the heterogeneous data sources combined using the composite kernel framework, the prediction capabilities increased considerably (0.92). Finally, posterior probabilities were evaluated for risk assessment of patients as an aid for clinicians to raise alertness at an early stage, in order to act promptly for avoiding AL complications. DISCUSSION: Machine-learning statistical model from EHR data can be useful to predict surgical complications. The combination of EHR extracted free text, blood samples values, and patient vital signs, improves the model performance. These results can be used as a framework for preoperative clinical decision support.
Authors: Feichen Shen; David W Larson; James M Naessens; Elizabeth B Habermann; Hongfang Liu; Sunghwan Sohn Journal: J Healthc Inform Res Date: 2018-11-06
Authors: Tom van den Bosch; Anne-Loes K Warps; Michael P M de Nerée Tot Babberich; Christina Stamm; Bart F Geerts; Louis Vermeulen; Michel W J M Wouters; Jan Willem T Dekker; Rob A E M Tollenaar; Pieter J Tanis; Daniël M Miedema Journal: JAMA Netw Open Date: 2021-04-01
Authors: Kristin M Corey; Sehj Kashyap; Elizabeth Lorenzi; Sandhya A Lagoo-Deenadayalan; Katherine Heller; Krista Whalen; Suresh Balu; Mitchell T Heflin; Shelley R McDonald; Madhav Swaminathan; Mark Sendak Journal: PLoS Med Date: 2018-11-27 Impact factor: 11.069