Citation :

Donepudi Rohini, M. Kavitha, "Attention-Enhanced LSTM and Transformer-Based Multi-Scale Feature Extraction Optimized with LOA for Robust Cervical Cancer Detection," International Journal of Electronics and Communication Engineering, vol. 13, no. 5, pp. 13-20, 2026. Crossref, https://doi.org/10.14445/23488549/IJECE-V13I5P102

Abstract

Cervical cancer continues to be a major disease in the world, especially in low-resource countries, where early and accurate diagnosis is often difficult. It has offered promising ways for automated medical image analysis in recent years; yet, issues related to extracting features for robust cervical cancer detection with LOA optimization need to be addressed. Therefore, in this study, cervical cell images were preprocessed using attention-enhanced LSTM and extracted features. The AE-LSTM component captures sequential and contextual dependency in derived features, and preprocessing to transformer-based multi-scale feature extraction on a transformer encoder with LOA optimization. Experimental evaluation on benchmark cervical cancer datasets shows a better overall performance in terms of training accuracy curve, training loss curve, confusion matrix, ROC curve, precision-recall curve, model comparison bar graph, LOA convergence Plot, and Feature Visualization (t-SNE). It developed a transformer encoder to produce either normal or abnormal diagnosis output. Future work will include domain adaptation for low-resource environments, combination with multi-modal clinical data, and the design of lightweight models suitable for the deployment of real-time cervical cancer screening systems.

Keywords

Robust Cervical Cancer Detection, Attention Mechanisms, LSTM, Multi-Scale Feature Extraction, Lion Optimization Algorithm (LOA).

References

1. Qun Zhang, Lijun Yang, and Feng Zhou, “Attention Enhanced Long Short-Term Memory Network with Multi-Source Heterogeneous Information Fusion: An Application to BGI Genomics,” Information Sciences, vol. 553, pp. 305-330, 2021.
   [CrossRef] [Google Scholar] [Publisher Link]
2. Kai Chen, Yi Zhou, and Fangyan Dai, “A LSTM-based Method for Stock Returns Prediction: A Case Study of China Stock Market,” 2015 IEEE International Conference on Big Data (Big Data), Santa Clara, CA, USA, pp. 2823-2824, 2015.
   [CrossRef] [Google Scholar] [Publisher Link]
3. Raghav Garg et al., “Enhanced Stock Market Movements towards Economic Sustainability using LSTM Hybrid Algorithm,” 2025 5^th International Conference on Advancement in Electronics & Communication Engineering (AECE), Ghaziabad, India, pp. 792-796, 2025.
   [CrossRef] [Google Scholar] [Publisher Link]
4. Divyani Gupta, Namrata Dhanda, and Kapil Kumar Gupta, “A Comparative Study of Various LSTM Models for Stock Market Time Series Classification,” 2025 8^th International Conference on Computing Methodologies and Communication (ICCMC), Erode, India, pp. 1-6, 2025.
   [CrossRef] [Google Scholar] [Publisher Link]
5. Xinpeng Xu, and Yukai Du, “Adaptive GCRA-CNN-LSTM: An Enhanced Model for Stock Price Prediction,” 2025 International Joint Conference on Neural Networks (IJCNN), Rome, Italy, pp. 1-8, 2025.
   [CrossRef] [Google Scholar] [Publisher Link]
6. Teofilus San Prasetya et al., “LSTM-Based Analysis of BBCA: Discovering Patterns in Time-Series Data,” 2025 IEEE International Conference on Artificial Intelligence and Mechatronics Systems (AIMS), Sumedang, Indonesia, pp. 1-6, 2025.
   [CrossRef] [Google Scholar] [Publisher Link]
7. Mingjie Wang et al., “MTMD: Multi-Scale Temporal Memory Learning and Efficient Debiasing Framework for Stock Trend Forecasting,” IEEE Transactions on Emerging Topics in Computational Intelligence, vol. 9, no. 3, pp. 2151-2163, 2025.
   [CrossRef] [Google Scholar] [Publisher Link]
8. Sepp Hochreiter, and Jürgen Schmidhuber, “Long Short-Term Memory,” Neural Computation, vol. 9, no. 8, pp. 1735-1780, 1997.
   [CrossRef] [Google Scholar] [Publisher Link]
9. Kangjie Cao, Ting Zhang, and Jueqiao Huang, “Advanced Hybrid LSTM-Transformer Architecture for Real-Time Multi-Task Prediction in Engineering Systems,” Scientific Reports, vol. 14, pp. 1-24, 2024.
   [CrossRef] [Google Scholar] [Publisher Link]
10. Chuanlin Zhang et al., “A Multi-Scale Feature Extraction Fusion Model for Human Activity Recognition,” Scientific Reports, vol. 12, pp. 1-13, 2022.
    [CrossRef] [Google Scholar] [Publisher Link]
11. Muhammad Amjad Raza et al., “Advanced Feature Extraction for Cervical Cancer Image Classification: Integrating Neural Feature Extraction and AutoInt Models,” Sensors, vol. 25, no. 9, pp. 1-22, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
12. Sandeep Kumar Mathivanan et al., “Enhancing Cervical Cancer Detection and Robust Classification through a Fusion of Deep Learning Models,” Scientific Reports, vol. 14, pp. 1-14, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
13. Ye Rang Park et al., “Comparison of Machine and Deep Learning for the Classification of Cervical Cancer based on Cervicography Images,” Scientific  Reports, vol. 11, pp. 1-11, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
14. Anabik Pal et al., “Deep Metric Learning for Cervical Image Classification,” IEEE Access, vol. 9, pp. 53266-53275, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
15. Gouri Shankar Chakraborty et al., “An Ensemble-based Diagnostic Framework Integrating Hybrid Convolutional Neural Network Models for Automated Cervical Cancer Detection: A Multi-Backbone Comparative Study and Deep Performance Synthesis,” 2025 3^rd International Conference on Intelligent Cyber Physical Systems and Internet of Things (ICoICI), Coimbatore, India, pp. 27-32, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
16. Saranya Al, S. Ravil, and T. Kalaichelvi, “Enhanced Cervical Cancer Detection and Classification through Deep Learning: a Review of Methodologies and Challenges,” 2025 9^th International Conference on Inventive Systems and Control (ICISC), Coimbatore, India, pp. 132-138, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
17. Khowaja Ashfaque, Zou BeiJi, and Xiaoyan Kui, “Cervix Visionator ELM: A Novel Approach to Early Detection of Cervical Cancer,” 2023 8^th International Conference on Control, Robotics and Cybernetics (CRC), Changsha, China, pp. 254-259, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
18. N. Meenakshisundaram, and G. Ramkumar, “A Robust Boosting Model for detecting Cervical Cancer Using Histogram Boosting Gradient Classifier,” 2023 Eighth International Conference on Science Technology Engineering and Mathematics (ICONSTEM), Chennai, India, pp. 1-6, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
19. Akshitha Shetty, and Vrushika Shah, “Survey of Cervical Cancer Prediction Using Machine Learning: A Comparative Approach,” 2018 9^th International Conference on Computing, Communication and Networking Technologies (ICCCNT), Bengaluru, India, pp. 1-6, 2018.
    [CrossRef] [Google Scholar] [Publisher Link]
20. Fahima Khanam, and Md. Rubaiyat Hossain Mondal, “Ensemble Machine Learning Algorithms for the Diagnosis of Cervical Cancer,” 2021 International Conference on Science & Contemporary Technologies (ICSCT), Dhaka, Bangladesh, pp. 1-5, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
21. T.J. Nagalakshmi et al., “Detection of Cervical Cancer with Texture Analysis using Machine Learning Models,” 2022 International Conference on Advances in Computing, Communication and Applied Informatics (ACCAI), Chennai, India, pp. 1-6, 2022.
    [CrossRef] [Google Scholar] [Publisher Link]
22. Mengying Zhao et al., “Automated Classification of Cervical Image Based on Deep Neural Network,” 2022 IEEE 11^th Data Driven Control and Learning Systems Conference (DDCLS), Chengdu, China,, pp. 1-6, 2022.
    [CrossRef] [Google Scholar] [Publisher Link]
23. N. Meenakshisundaram, and G. Ramkumar “A Combined Deep CNN-LSTM Network for Chromosome Classification for Metaphase Selection,” 2022 International Conference on Inventive Computation Technologies (ICICT), Nepal, pp. 1005-1010, 2022.
    [CrossRef] [Google Scholar] [Publisher Link]