Development of a clinical decision support system for breast cancer detection using ensemble deep learning.
Authors
Affiliations (4)
Affiliations (4)
- Chitkara University Institute of Engineering and Technology, Chitkara University, Rajpura, Punjab, India.
- Department of Computer Engineering & Applications, GLA University, Mathura, India.
- Tech School, Computer Science Department, ICFAI University, Ranchi, Jharkhand, India.
- Department of Information Technology, Manipal Institute of Technology Bengaluru, Manipal Academy of Higher Education, Manipal, India. [email protected].
Abstract
Advancements in diagnostic technology are required to improve patient outcomes and facilitate early diagnosis, as breast cancer is a substantial global health concern. This research discusses the creation of a unique Deep Learning (DL) Ensemble Deep Learning based on a Clinical Decision Support System (EDL-CDSS) that enables the precise and expeditious diagnosis of breast cancer. Numerous DL models are combined in the proposed EDL-CDSS to create an ensemble method that optimizes the advantages and reduces the disadvantages of individual techniques. The team improves its capacity to extricate intricate patterns and features from medical imaging data by incorporating the Kelm Extreme Learning Machine (KELM), Deep Belief Network (DBN), and other DL architectures. Comprehensive testing has been conducted across various datasets to assess the efficacy of this system in comparison to individual DL models and traditional diagnostic methods. Among other objectives, the evaluation prioritizes precision, sensitivity, specificity, F1-score, accuracy, and overall accuracy to mitigate false positives and negatives. The experiment's conclusion exhibits a remarkable accuracy of 96.14% in comparison to prior advanced methodologies.