Endocrine disorder affecting women in their reproductive age is called Polycystic Ovary Syndrome (PCOS), which leads to health consequences. Diagnostic methods fail to detect early signs of the disorder in the monitoring of PCOS biomarkers. Due to advancements in the healthcare domain, monitoring tools can now provide personalized treatment for patients. In continuous monitoring for PCOS, there is underutilization of wearable technologies, as existing devices fail in tracking biomarkers like hormones along with fluctuations in glucose. As current methods do not offer non-invasive, accessible techniques for detection at the early stages in management, the integration of wearable biosensors along with mobile technology presents aresearch gap. The aim is to develop an AI-driven wearable system that integrates both a mobile device and a biosensor for monitoring PCOS-related biomarkers including hormone levels and glucose variability. Using silicon photonics, the developed system helps in the detection of imbalances in hormone levels non-invasively by saliva analysis. Data are analyzed through the usage of YOLO algorithms, which are used for sending alerts based on the health of users. The Ava Bracelet, a wearable with a mobile-enabled biosensor system, integrates AI algorithms to monitor biomarkers related to PCOS through saliva. By employing nanomaterials and thin-film silicon photonic sensors, it provides precise and non-invasive detection of hormonal fluctuations. YOLOv8 is one of the significant models in AI that helps in the identification of anomalies; the engagement of users is also improved, and interventions are promptly facilitated. In contrast to existing works, the proposed system achieves an accuracy of 92%, a sensitivity of 89%, and a specificity of 94% while detecting abnormalities related to PCOS. The AI-driven system improves upon existing PCOS diagnostics through continuous, real-time monitoring and early detection of hormonal imbalances. Utilizing YOLOv8 enables rapid and accurate data processing, providing personalized insights and facilitating proactive health management. Biomarkers like luteinizing hormone (LH) and follicle-stimulating hormone (FSH) are targeted in detection by utilizing aptamer-based and antibody-based biorecognition elements.