Figure 2 | Scientific Reports

Figure 2

From: Non-invasively accuracy enhanced blood glucose sensor using shallow dense neural networks with NIR monitoring and medical features

Figure 2

Our real-time glucose monitoring (a) the overall framework (b) experiment on real usage in a person (low-pass filtered) Shallow Dense Neural Networks using mbNIR and PMF as input data for DM classification and Real-time Blood Glucose Monitoring (a) mbNIR non-invasive measurement, (b) glucose absorbance spectrum (c) brief alignment between finger and sensor and (d) Schematic of the mbNIR/PMF + SDNN illustrating that light scattered from finger was filtered before entering readout channel of the Controller, for band#1 (800–900 nm) centered at 850 nm, band#2 (900–1000 nm) centered at 950 nm, and band#3 (1100–1200 nm) centered at 1150 nm respectively.

Back to article page