Start Date
10-12-2017 12:00 AM
Description
Although researchers have uncovered potential positive impacts of digital technologies in healthcare and medical centers have been increasingly making use of technology to digitally store their data, the use of healthcare analytics in clinical practice remains limited. In particular, the application of machine learning (ML) approaches, although holding the potential of providing valuable insights, is mainly restricted to descriptive ML, due to the approximate nature of ML, the impact of inaccuracies, and the perceived potential additional efforts in clinical workflows. Taking into account these barriers to healthcare analytics adoption, in this multidisciplinary study, we obtained and jointly analyzed cancer data on 799 cases of cranio-maxillofacial and oral-maxillofacial surgery. We developed a real-time decision support system that predicts optimal treatments and communicates its prediction confidence along with patient attributes that are significant to decision making, thereby providing potentials simultaneously for improving quality of care and for increasing process efficiency for physicians.
Recommended Citation
Tofangchi, Schahin; Hanelt, Andre; and Böhrnsen, Florian, "Distributed Cognitive Expert Systems in Cancer Data Analytics: A Decision Support System for Oral and Maxillofacial Surgery" (2017). ICIS 2017 Proceedings. 19.
https://aisel.aisnet.org/icis2017/IT-and-Healthcare/Presentations/19
Distributed Cognitive Expert Systems in Cancer Data Analytics: A Decision Support System for Oral and Maxillofacial Surgery
Although researchers have uncovered potential positive impacts of digital technologies in healthcare and medical centers have been increasingly making use of technology to digitally store their data, the use of healthcare analytics in clinical practice remains limited. In particular, the application of machine learning (ML) approaches, although holding the potential of providing valuable insights, is mainly restricted to descriptive ML, due to the approximate nature of ML, the impact of inaccuracies, and the perceived potential additional efforts in clinical workflows. Taking into account these barriers to healthcare analytics adoption, in this multidisciplinary study, we obtained and jointly analyzed cancer data on 799 cases of cranio-maxillofacial and oral-maxillofacial surgery. We developed a real-time decision support system that predicts optimal treatments and communicates its prediction confidence along with patient attributes that are significant to decision making, thereby providing potentials simultaneously for improving quality of care and for increasing process efficiency for physicians.