loading page

Application of concept Drift Detection and Adaptive Framework for Non Linear Time Series Data from Cardiac Surgery
  • +6
  • Nitin Auluck,
  • Rajarajan Ganesan,
  • Tarunpreet Kaur,
  • Alisha Mittal,
  • Mansi Sahi,
  • Sushant Konar,
  • Tanvir Samra,
  • Goverdhan Dutt Puri,
  • Shayam Kumar Singh Thingnum
Nitin Auluck
Indian Institute of Technology Ropar

Corresponding Author:nitin@iitrpr.ac.in

Author Profile
Rajarajan Ganesan
Post Graduate Institute of Medical Education and Research
Author Profile
Tarunpreet Kaur
Indian Institute of Technology Ropar
Author Profile
Alisha Mittal
Post Graduate Institute of Medical Education and Research
Author Profile
Mansi Sahi
Indian Institute of Technology Ropar
Author Profile
Sushant Konar
Post Graduate Institute of Medical Education and Research
Author Profile
Tanvir Samra
Post Graduate Institute of Medical Education and Research
Author Profile
Goverdhan Dutt Puri
Post Graduate Institute of Medical Education and Research
Author Profile
Shayam Kumar Singh Thingnum
Post Graduate Institute of Medical Education and Research
Author Profile

Abstract

The quality of machine learning (ML) models deployed in dynamic environments tends to decline over time due to disparities between the data used for training and the upcoming data available for prediction, which is commonly known as drift. Therefore, it is important for ML models to be capable of detecting any changes or drift in the data distribution and updating the ML model accordingly. This study presents various drift detection techniques to identify drift in the survival outcomes of patients who underwent cardiac surgery. Additionally, this study proposes several drift adaptation strategies, such as adaptive learning, incremental learning, and ensemble learning. Through a detailed analysis of the results, the study confirms the superior performance of ensemble model, achieving a minimum mean absolute error (MAE) of 10.684 and 2.827 for predicting hospital stay and ICU stay, respectively. Furthermore, the models that incorporate a drift adaptive framework exhibit superior performance compared to the models that do not include such a framework.
05 Jun 2023Submitted to Computational Intelligence
04 Sep 2023Submission Checks Completed
04 Sep 2023Assigned to Editor
05 Oct 2023Review(s) Completed, Editorial Evaluation Pending
08 Nov 2023Reviewer(s) Assigned
31 Jan 2024Editorial Decision: Revise Major
05 Feb 20242nd Revision Received
05 Feb 2024Submission Checks Completed
05 Feb 2024Assigned to Editor
12 Feb 2024Reviewer(s) Assigned
25 Apr 2024Review(s) Completed, Editorial Evaluation Pending