Meghana K B, Abia P, Priyanka R, Arpitha, P Divya, Shilpa S P

DOI: https://doi.org/10.5281/zenodo.23539754

Polycystic Ovary Syndrome (PCOS) is a common hormonal disorder affecting women of reproductive age and is associated with infertility, metabolic problems and psychological stress. Many existing studies focus on its causes, lifestyle factors, mental health effects and prediction using machine learning models such as neural networks, Convolutional neural network (CNN) and hybrid algorithms. The proposed system combines medical data, lifestyle factors and psychological indicators to improve PCOS prediction. It integrates hybrid deep learning models with explainable AI to provide accurate, transparent and early diagnosis. Additional features such as stress levels, sleep patterns, diet are included to enhance prediction and support personalized healthcare recommendations. This system also provides risk assessment, early warning alerts and personalized lifestyle suggestions to help users manage symptoms and reduce long term complications. By integrating multiple data sources and advanced analytics, the model aims to assist doctors in decision making and promote early detection and better management of PCOS.

PCOS, Hybrid deep learning, Machine learning, Explainable AI (XAI), Early diagnosis, Risk Prediction

The proposed Data-Driven Predictive Model for PCOS Risk Assessment and Personalized Lifestyle Intervention offers an integrated approach to supporting early PCOS risk identification and preventive healthcare. The system brings together clinical, lifestyle, psychological, and optional hormonal data with machine learning-based prediction. When an ultrasound image is available, the CNN module can analyze image-based ovarian patterns, while the SVM module handles tabular health information. The proposed hybrid framework combines these prediction outputs to generate a final PCOS prediction along with a confidence score. Beyond prediction, the system evaluates the user’s overall risk as Low, Moderate, or High by taking into account factors such as BMI, symptoms, lifestyle, stress, age and any available hormone data. Personalized recommendations covering diet, exercise, sleep, and stress management are then provided based on the user’s risk level and health profile. The system also supports report generation, data storage, history tracking and a user-friendly dashboard, making it more comprehensive than systems that offer only a PCOS classification. By bringing prediction, risk assessment, and personalized lifestyle guidance together within a single framework, the proposed system aims to support early awareness, preventive care and informed healthcare decision-making.

[1] Jayati Dixit, Anshuman Rai, Pulak Tandon, and Dhiraj Pandey,“ASHA:Machine Learning-Based Polycystic Ovary Syndrome (PCOS) Detection and Prediction System, ”Pune, India, Jan. 16–17, 2025, pp. 1–5, IEEE, 2025.
[2] Xianqing Hu, Amit Yadav, Asif Khan, Abhishek Pratap Sah, and Sami Azam, “Construction of PCOS Prediction Model Based on BP NeuralNetwork,”pp.885–889, 2025.
[3] S. Singh, N. Pal, S. Shubham, D. K. Sarma, V. Verma, F. Marotta, and M. Kumar, “Polycystic Ovary Syndrome: Etiology, Current Management, and Future Therapeutics,” Journal of Clinical Medicine, vol. 12, no. 4, p. 1454, 2023.
[4] J. J. Benjamin, K. MaheshKumar, V. Radha, K. Rajamani, N. Puttaswamy, T. Koshy, K. N. Maruthy, and R. Padmavathi, “Stress and polycystic ovarian syndrome – a case control study among Indian women,” Clinical Epidemiology and Global Health, vol. 22, p. 101326, 2023.
[5] A.Sawant, N. Sayyad, S. Ugale, M. Gaikwad, and T. Kande, “PCOS:An Overview of Lifestyle Disorder Found in 90% Females Worldwide,” International Journal of Creative Research Thoughts (IJCRT), vol. 10, no. 12, pp. C233–C239, Dec. 2022.
[6] G. Konduri, A. Bollam, and R. Balaprabha, “Assessment of Anxiety, Stress and Depression in Women With and Without PCOS: A Comparative Study,” International Journal of Creative Research Thoughts, vol. 10, no. 5, pp. d807–d811, May 2022.
[7] I. Waghmare, “Effect of Polycystic Ovary Syndrome (PCOS) in Indian Women of Reproductive Age", International Journal of Creative Research Thoughts, vol. 10, no. 1, Jan. 2022.
[8] A. G. Wale, S. S. Walvekar, and S. R. Zine,“Female Infertility: A Review on Definition, Causes and Its Treatment,”International Journal of Creative Research Thoughts, vol. 8, no. 9, Sep. 2020.
[9] Priyanka Nalawade, Vishal Borate, Rajlakshmi N. Shindalkar, Snehal S. Bansode, Shreya D. More, and Cynthia Pachal,“Explainable CNN & Hybrid Deep Learning Model for Health Risk PCOS Prediction & Diagnosis,”International Conference on Intelligent Embedded, Microelectronics, Communication and Optical Networks, 2025.

Meghana K B, Abia P, Priyanka R, Arpitha, P Divya, Shilpa S P (2026). A Data-Driven Predictive Model for PCOS Risk Assessment and Personalized Lifestyle Intervention. International Journal of Computer Techniques, 13(5), 530–535. ISSN: 2394-2231. DOI: https://doi.org/10.5281/zenodo.23539754