This study evaluates the potential for the use of low-cost discrete optical semiconductors, specifically light-emitting diodes (LEDs) and a photodiode, for non-invasive measurement of microvascular tissue oxygen saturation (StO). StO is a crucial biomarker in monitoring microvascular function and tissue viability. Spectrometer-based methods typically use complex and expensive equipment, with the cost per patient potentially amounting to hundreds of dollars. This study aims to provide understanding of tissue–light interaction with broader implications extending to applications such as photoplethysmography (PPG). Our approach involves a system that includes three specifically selected LEDs coupled with a photodiode, focusing on assessing microvascular StO. The methodology includes several phases: in vitro calibration using a controlled deoxygenation process in a liquid tissue phantom, computational simulations to estimate the penetration depths of selected LED wavelengths, an analysis of the effects of variability in LED output on measurement accuracy, and a preliminary human study. Results from the in vitro experiments demonstrated a root mean square error of 3.9 StO-% between a spectrometer reference and our technique. The human study including baseline, occlusion and post-occlusion StO measurements in six volunteers resulted in 76.0, 52.6 and 77.5 StO-%, respectively. Computational simulations confirmed the effective penetration of selected wavelengths into targeted microvascular layers. The intrinsic and external factors affecting the measurement accuracy were analyzed. The findings support the feasibility of a cost-effective, simplified, and effective system for continuous monitoring of microvascular tissue oxygenation.