International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 01 | Jan 2025
p-ISSN: 2395-0072
www.irjet.net
Integrative Approach to PCOS Detection Using Machine Learning and Convolutional Neural Networks Aiswarya Rani1, Sagar1, Shivangi Mishra1 ¹Student, Delhi Pharmaceutical Sciences and Research University, New Delhi, India ¹Student, Delhi Pharmaceutical Sciences and Research University, New Delhi, India ¹Student, Delhi Pharmaceutical Sciences and Research University, New Delhi, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - This study developed two models for Polycystic
adolescent females, with higher rates in urban areas compared to rural regions [10].
Ovary Syndrome (PCOS) detection using clinical and imaging data. A stacking ensemble model achieved 93% accuracy (Area Under the Curve [AUC] 0.99), identifying key predictors such as follicle counts, Body Mass Index (BMI), and hormonal markers. A Convolutional Neural Network (CNN) achieved 92.72% accuracy (AUC 0.89) with zero misclassifications, ensuring reliable ultrasound image classification. Advanced feature selection methods and a Streamlit-based interface enable real-time diagnostics, supporting early detection and improved clinical outcomes.
Conversely, research in Lucknow observed a prevalence of only 3.7% using the NIH criteria among women aged 18–25 with menstrual irregularities and hirsutism [11]. Similarly, a study in Andhra Pradesh reported a 9.13% prevalence based on the Rotterdam criteria [12]. These variations are often influenced by the choice of diagnostic framework, lifestyle factors, and healthcare access [13,14].
Key Words: Polycystic Ovary Syndrome (PCOS), Diagnostic
The pathophysiology of PCOS involves a complex interplay of hormonal, genetic, and environmental factors. Hyperandrogenism, characterized by elevated androgen levels, disrupts normal ovarian function and follicular development, leading to anovulation and polycystic ovarian morphology [15,16]. Insulin resistance is another critical factor, as hyperinsulinemia exacerbates androgen production while reducing sex hormone-binding globulin (SHBG) levels, amplifying symptoms like hirsutism and acne [17,18].
Models, Machine Learning, Prevalence, Reproductive Health
1.INTRODUCTION Polycystic Ovary Syndrome (PCOS) is a multifaceted endocrine disorder that affects 4% to 20% of women of reproductive age globally, with prevalence varying based on diagnostic criteria and geographic location [1, 2]. First described in 1935 by Stein and Leventhal, it is characterized by a wide spectrum of reproductive, metabolic, and psychological symptoms, including infertility, hyperandrogenism, menstrual irregularities, insulin resistance, obesity, and mood disorders such as anxiety and depression [3,4].
Genetic predispositions, such as mutations in FSHR, LHCGR, INSR, and THADA genes, further contribute to PCOS, along with epigenetic changes resulting from prenatal androgen exposure and elevated maternal anti-Müllerian hormone (AMH) levels [19][20][21].
Long-term complications of PCOS include type 2 diabetes, cardiovascular disease, endometrial cancer, and obstructive sleep apnea, underscoring its public health significance [4,5]. Despite these health concerns, up to 70% of cases remain undiagnosed globally due to the disorder’s heterogeneous presentation and overlapping symptoms with other endocrine conditions [6].
Environmental and lifestyle factors play a significant role in the onset and progression of PCOS. High-calorie diets, sedentary behaviour, and exposure to endocrine-disrupting chemicals (EDCs) have been linked to increased prevalence, particularly in urban populations [22,12]. Geographical and socioeconomic disparities further influence symptom severity and healthcare access. Indian women, for instance, frequently present with unique clinical features, such as higher incidences of insulin resistance, acanthosis nigricans, and thyroid dysfunction, necessitating tailored diagnostic and therapeutic approaches [14,9] [23].
Globally, PCOS affects approximately 8–13% of women of reproductive age, with significant variations in prevalence across populations due to differences in diagnostic criteria and study methodologies.[7]. In India, the prevalence of PCOS ranges from 3.7% to 22%, highlighting regional and demographic disparities [8]. For example, a community-based study in Mumbai reported a prevalence of 22.5% using the Rotterdam criteria [9], while a pilot study in Tamil Nadu found an 18% prevalence among
© 2025, IRJET
|
Impact Factor value: 8.315
The diagnostic framework for PCOS has evolved significantly over the decades. The Rotterdam criteria, established in 2003, are the most widely used today and require the presence of at least two of the following: oligo-anovulation,
|
ISO 9001:2008 Certified Journal
|
Page 203