Noninvasive diabetes mellitus classification based on HbA1c using the mel-frequency cepstral coefficients from the photoplethysmographic signal

Diabetes mellitus (DM) is a metabolic condition with improper regulation of blood sugar&#xD;levels and is one of the leading global causes of death. Glycated hemoglobin (HbA1c)&#xD;serves as a crucial indicator for managing diabetes. This study proposes a noninvasive&#xD;approach to classify glycemic status based on HbA1c levels. This work uses novel&#xD;Mel-frequency cepstral coefficient features of finger photoplethysmographic (PPG) signals&#xD;and physiological parameters, enabling straightforward detection of DM. A finger PPG&#xD;dataset (in reflective mode) comprising 180 subjects with diabetes, prediabetes, and&#xD;normal HbA1c levels is curated and used to validate the proposed method. The dataset&#xD;comprises 93 normal (HbA1c < 5.7%), 57 prediabetic (HbA1c 5.7% - 6.4%), and 30&#xD;diabetic (HbA1c ≥ 6.5%) individuals. Furthermore, a hybrid feature (HyF) selection&#xD;method is employed for feature reduction. The HyF Selection-based gradient boosting&#xD;model achieved effective accuracies of 93.89% for binary classification and 91.67% for&#xD;multiclass classification. The results are compared with the gold standard HbA1c test.&#xD;Both the binary and multiclass classifications show improved overall performance. These&#xD;findings indicate that PPG signals are a feasible substitute for noninvasive HbA1c&#xD;detection and have potential for wearable HbA1c monitoring.