Related Experiment Video
Updated: Mar 29, 2026
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
43.8K
Cervical cancer image analysis: Detection and segmentation using self-guided quantum GANs and musical chairs
A Naresh Kumar1, S Sageengran2, Parikshit Narendra Mahalle3
1Department of Computer Science and Business Systems, Sri Sairam Engineering College, Sai Leo Nagar, West Tambaram Poonthandalam, Village, Chennai, Tamil Nadu, India.
Computers in Biology and Medicine
|March 27, 2026
Summary
This study introduces a novel deep learning framework, SQGAN-MCO, for accurate cervical cancer detection from histopathology images. The method significantly improves diagnostic accuracy, aiding early detection and reducing errors.
Area of Science:
- Oncology
- Medical Imaging
- Artificial Intelligence
Background:
- Cervical cancer is primarily caused by long-term Human Papillomavirus (HPV) infections.
- Current deep learning methods for cervical cancer detection often suffer from poor performance and frequent errors.
- Accurate and early detection is crucial for effective cervical cancer treatment and patient outcomes.
Purpose of the Study:
- To develop an advanced deep learning framework for precise identification and segmentation of cervical cancer in histopathology images.
- To enhance the accuracy and reliability of diagnostic tools for cervical cancer detection.
- To improve upon existing methods in terms of accuracy, precision, recall, and F1-score.
Main Methods:
- The study proposes a Self-guided Quantum Generative Adversarial Network with Musical Chairs Optimization (SQGAN-MCO) framework.
- Image preprocessing involved Adaptive Preprocessing Module Fusion (APMF) for clarity and Graph Enhanced Fuzzy Clustering (GEFC) for segmentation.
- Feature extraction utilized Adaptive Causal Decision Transformers (ACDT), with classification performed by Self-guided Quantum Generative Adversarial Network (SQGAN) and optimization via Musical Chairs Optimization Algorithm (MCOA).
Main Results:
- The SQGAN-MCO framework achieved high performance on the SIPaKMeD dataset, outperforming current techniques.
- Key metrics demonstrated significant improvements: Accuracy (98.6%), Precision (98.4%), Recall (98.8%), and F1-score (98.7%).
- The framework showed consistent high performance across all classes: Normal, LSIL, HSIL, and SCC.
Conclusions:
- The proposed SQGAN-MCO framework offers a computationally effective and highly precise diagnostic tool for cervical cancer.
- This advancement can assist pathologists in early detection and reduction of diagnostic errors.
- The study highlights the potential for automated medical image analysis and improved cancer decision-making in clinical settings.
