Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{199927,
        author = {Mohammad Fayazuddin and Zeba Tanwar and Deepak and Maharajan. S and Suruthi Suga. A},
        title = {AI-Powered Predictive Modeling for Early Diagnosis of PCOS and Thyroid Disorders Using Electronic Health Record Data},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {833-856},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=199927},
        abstract = {Polycystic Ovary Syndrome (PCOS) and thyroid disorders are among the most prevalent endocrine conditions worldwide, yet both continue to be significantly underdiagnosed and often identified only after prolonged clinical progression. This delay in diagnosis is largely attributed to the non-specific and overlapping nature of symptoms, including fatigue, weight fluctuations, menstrual irregularities, and metabolic disturbances, which frequently lead to fragmented clinical evaluation and missed opportunities for early intervention. In recent years, the widespread adoption of Electronic Health Records (EHRs) has resulted in the accumulation of extensive longitudinal patient data, encompassing laboratory results, clinical observations, and treatment histories. Despite this, much of the data remains passively stored and underutilised for proactive disease detection. Advancements in artificial intelligence (AI), particularly in machine learning and deep learning, offer a promising pathway to transform this paradigm by enabling the identification of complex patterns and predictive signals embedded within EHR data. This paper explores the application of AI-powered predictive modelling for the early diagnosis of PCOS and thyroid disorders, focusing on the integration of multi-dimensional clinical data and temporal health trajectories. By leveraging structured and semi-structured EHR datasets, predictive models can detect subtle biochemical and physiological changes that precede overt clinical manifestation, thereby facilitating earlier diagnosis and timely clinical intervention. The study discusses various modelling approaches, including traditional statistical methods, ensemble learning techniques, and advanced deep learning architectures such as hybrid convolutional neural network–long short-term memory (CNN-LSTM) models, which are particularly effective in capturing both cross-sectional feature interactions and longitudinal trends. In addition, the importance of explainable artificial intelligence (XAI) is emphasised to ensure that model predictions are interpretable, clinically relevant, and aligned with established medical knowledge, thereby enhancing clinician trust and supporting informed decision-making. Furthermore, the paper critically examines the challenges associated with implementing AI-driven predictive systems in healthcare settings, including issues related to data quality, interoperability, algorithmic bias, and ethical considerations such as patient privacy and regulatory compliance. It also highlights the potential of such systems to reduce diagnostic delays, improve disease management, and promote equitable access to healthcare, particularly in resource-constrained environments. 
In conclusion, AI-powered predictive modelling using EHR data represents a significant advancement in the early detection of endocrine disorders. By shifting the focus from reactive diagnosis to proactive risk identification, such approaches have the potential to improve clinical outcomes, reduce long-term complications, and contribute to the broader goal of precision and preventive medicine.},
        keywords = {Polycystic Ovary Syndrome, Thyroid Disorder, Electronic Health Records, Machine Learning, Deep Learning, Predictive Modelling, Clinical Decision Support, SHAP Explainability, Endocrine Disorders, Early Diagnosis},
        month = {May},
        }