Infertility Prediction using Long Short-Term Memory Algorithm with PCOS Electronic Healthcare Data

• Published in: 2025 4th OPJU International Technology Conference (OTCON) on Smart Computing for Innovation and Advancement in Industry 5.0 pp. 1 - 7
• Main Authors: Sarath, T., Brindha, K.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 09.04.2025
• Subjects: Analytical models; Data models; Electronic Health Record; Electronic medical records; Infertility; Long Short-Term Memory; Monitoring; Obesity; Patient Care; Polycystic Ovarian Syndrome; Prediction algorithms; Predictive models; Technological innovation
• DOI: 10.1109/OTCON65728.2025.11071074

Among women of Menstrual age, Polycystic Ovarian Syndrome (PCOS) ranks highest among Infertility disorders. The dissimilar diagnostic conditions for PCOS make it a diverse illness. When other potential causes of infertility have been eliminated, the diagnosis can make on presence of two or three cardinal features: hyperandrogenism, polycystic ovary morphology, and irregular menstrual cycles. Consequently, women's infertility might not provide enough evidence for PCOS diagnosis on a large scale. Obesity increases is one of the cause that risks metabolic diseases like metabolic syndrome, type 2 diabetes, decreased glucose tolerance and hyperandrogenism among other characteristics. In this way, PCOS patients could be more precisely identified using the data and text included in Electronic Health Records (EHR). So, the goal of this paper is to analyze the infertility in women based on PCOS patient record data with Long Short-Term Memory (LSTM) model to increase the monitoring, patient care, diagnosis and treatment outcomes.