Early detection of infection and impaired healing in chronic wounds remains a fundamental challenge in clinical practice. Current assessment protocols rely on visual inspection and manual examination, both subjective and prone to delayed recognition of physiological deterioration. This delay directly impacts infection rates, amputation outcomes, and treatment duration, particularly in resource-limited healthcare settings. This work presents TARANG, a multimodal sensing framework designed to provide objective, quantifiable wound assessment at fixed standoff distance (10 cm) without direct contact. The system integrates three complementary measurement modalities: optical imaging for tissue surface charac-terization, thermal infrared imaging for detection of inflammation and perfusion asymmetry, and volatile organic com-pound (VOC) analysis for identification of metabolic biomarkers associated with bacterial activity and tissue degradation. Raw sensor data are processed through feature extraction pipelines to generate quantitative indicators of wound condition, risk classification, and healing trajectory. The analytical framework combines signal processing methods from thermal image analysis, spectral decomposition of optical reflectance patterns, and time-series VOC trend analysis. These features are fed into a multimodal ensemble model to produce a composite risk assessment. The system architecture preserves clinician oversight, functioning as an objective measurement tool rather than an autonomous diagnostic system.

Touchless Healthcare Technology, Artificial Intelligence (AI) in Healthcare, Neural-guided Healing, Biomedical Signal Processing