Mental Health Correlates of PCOD in Indian Women: A Machine Learning-Based Predictive Approach

¹ Brainware University, Kolkata, India
² Department of Computer Science and Engineering (IoT, CS, BT), University of Engineering and Management, Kolkata, India

Abstract

Polycystic Ovarian Disease (PCOD) is a multifactorial endocrine disorder that significantly affects reproductive-aged women, manifesting through a combination of physiological and psychological symptoms. In South India, cultural expectations, limited awareness in rural areas, and the stigma surrounding mental health often compound the emotional burden of the disorder. This chapter explores the intersection between PCOD and mental health, analyzing patterns of fatigue, mood swings, stress, cognitive challenges, and lifestyle changes in diagnosed women. Polycystic Ovarian Disease (PCOD) is a multifactorial endocrine disorder affecting reproductive-aged women, often leading to infertility, obesity, and psychological distress. The growing application of machine learning in healthcare provides promising avenues for accurate, early detection of PCOD. In this chapter, we investigate the efficacy of gradient boosting algorithms for PCOD prediction using a dataset containing physiological and psychological attributes. Employing train-test splits (80-20, 50-50, 66-34) and 10-fold cross-validation for robustness, our model achieves a commendable accuracy of 97.65%, highlighting its potential utility for clinical decision support systems. By combining traditional survey-based analysis with advanced predictive models, the study emphasizes the critical link between PCOD symptoms and mental health outcomes. The findings support the integration of mental wellness indicators in diagnostic frameworks and advocate for culturally sensitive healthcare interventions across southern Indian populations.