Complementary Detection for Hardware Efficient On-Site Monitoring of Parkinsonian Progress

The progress of Parkinson's disease (PD) in patients is conventionally monitored through follow-up visits. These may be insufficient for clinicians to obtain a good understanding of the occurrence and severity of symptoms in order to adjust therapy to the patients' needs. Portable platforms for PD diagnostics can provide in-depth information, thus reducing the frequency of face-to-face visits. This paper describes the first known on-site PD detection and monitoring processor. This is achieved by employing complementary detection which uses a combination of weak k-NN classifiers to produce a classifier with a higher consistency and confidence level than the individual classifiers. Various implementations of the classifier are investigated for tradeoffs in terms of area, power, and detection performance. Detection performances are validated on an field programmable gate array platform. Achieved accuracy measures were: Matthews correlation coefficient of 0.6162, mean F1-score of 91.38%, and mean classification accuracy of 91.91%. By mapping the implemented designs on a 45-nm CMOS process, the optimal configuration achieved a dynamic power per channel of 2.26 \mu \text{W} and an area per channel of 0.24 mm 2 .