Table 2 Non-invasive blood glucose approaches in literature.

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

Non-invasive methods

Part of the body

Advantages

Disadvantages

References

Photoacoustic spectroscopy

Intersitial fluid

Minimally invasive

Need recalibration and require human’s secretion

20

Hybrid NIR and GHz wavelength

In vitro (aquas solution)

Improved accuracy compared to barely NIR

Not satisfied clinical accuracy; Further in-invo experiments needed

6

Ring resonators

Solution of glucose

Broad glucose range can be measured

Further validation on human required

21

photoplethysmogram (PPG) signal + Galvanic skin response

Skin

50 human subject samples reported

Temperature and skin moisture interference Tested only in non-diabetic subjects

22

Microwave + Particle Swarm Optimization

Earlobe

 

High complexity but low accuracy

23

Back to article page