Related Experiment Video For Adaptive causal decision transformers
Updated: Mar 29, 2026

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
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.
Abstract:
The main cause of cervical cancer emerges from long-term Human Papillomavirus (HPV) infections damaging the cervix. The detection methods produced by deep learning algorithms include many available options yet operating systems typically demonstrate poor performance levels and produce mistakes frequently. The research introduces a cervical cancer identification and segmentation within histopathology images through Self-guided Quantum Generative Adversarial Network with Musical Chairs Optimization (SQGAN-PKO). APMF is the initial processing step of the method because it improves image clarity without compromising vital picture elements. Graph Enhanced Fuzzy Clustering (GEFC) is used for segmentation in order to guarantee accurate region identification. The feature extraction is performed with Adaptive Causal Decision Transformers (ACDT), the Classification with the help of a Self-guided Quantum Generative Adversarial Network (SQGAN). The categorization accuracy is further increased using the Musical Chairs Optimization Algorithm (MCOA). The suggested approach works better than the current ones evaluated on the SIPaKMeD dataset. The results showed considerable improvements in key metrics as Accuracy, precision, recall, F1-score, and AUC across all classes (Normal, LSIL, HSIL, and SCC), experimental assessments show that the proposed SQGAN-MCO framework works noticeably better than current techniques, obtaining greater Accuracy 98.6%, precision98.4%, recall 98.8%, and F1-score 98.7%. This study advances the creation of extremely precise and computationally effective diagnostic instruments that can help pathologists identify cervical cancer in its early stages and lower diagnostic mistakes. The results highlight how well the suggested framework works in clinical settings, opening the door for more automated medical image analysis and better cancer decision-making.
