ABSTRACT Diabetes, a chronic disorder affecting nearly 422 million people worldwide, is measured through an invasive process. Although this process is highly accurate it is painful and uncomfortable especially when multiple readings are needed. The invention focuses on developing a Hybrid Quartz Crystal Micrpbalance (QCM) sensor with Optical Waveguides to measure the acetone concentration (biomarker for diabetes). Present studies have found that diabetes exists in five different types and this classification - can help patients to treat this chronic disorder more efficiently. The invention focuses on used deep learning algorithms to classify the raw signals obtained from the hybrid sensor into these five different classes namely Severe Autoimmune Diabetes, Severe Insulin- Deficient Diabetes, Severe Insulin-Resistant Diabetes, Mild Obesity-Related Diabetes and Mild Age-Related Diabetes. This system shall help diabetic patients continuously mpnitor their glucose levels npn-invasively and thus help them improve their quality of life by controlling this disorder effectively.
