Related Experiment Video
Updated: Jul 19, 2026
08:05
Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence
Published on: June 10, 2025
Novel round-robin tabu search algorithm for prostate cancer classification and diagnosis using multispectral imagery
Muhammad Atif Tahir1, Ahmed Bouridane
1Faculty of Computing, Engineering, and Mathematical Sciences, University of the West of England, Bristol, UK. muhammad.tahir@uwe.ac.uk
Summary
A new round-robin tabu search (RR-TS) algorithm effectively addresses the curse of dimensionality in multispectral prostate cancer image analysis. This method achieved 98%-100% classification accuracy, outperforming existing techniques.
Area of Science:
- Digital Pathology
- Computational Imaging
- Machine Learning in Oncology
Background:
- Quantitative cell imagery is crucial for cancer diagnosis via biopsy analysis.
- Prostate cancer diagnosis involves distinguishing between stroma, benign hyperplasia, neoplasia, and carcinoma.
- Multispectral imagery offers rich data but faces the curse of dimensionality.
Purpose of the Study:
- To propose a novel algorithm to overcome the curse of dimensionality in multispectral prostate cancer image classification.
- To develop an effective pattern recognition technique for high-dimensional medical image data.
Main Methods:
- Development of a novel round-robin tabu search (RR-TS) algorithm.
- Application of RR-TS to classify multispectral images of prostate cancer.
- Experimental validation on prostate cancer textured multispectral image datasets.
Main Results:
- The RR-TS algorithm achieved 98%-100% classification accuracy on two datasets.
- The proposed system demonstrated superior performance compared to PCA-LDA, TS-1NN, and C4.5 classifiers.
- Effective mitigation of the curse of dimensionality for multiclass image analysis.
Conclusions:
- The RR-TS algorithm is a highly effective solution for high-dimensional multispectral image classification in prostate cancer.
- This approach significantly improves diagnostic accuracy in digital pathology.
- The findings suggest a promising direction for automated cancer diagnosis using advanced computational methods.
