Toward On-Demand Deep Brain Stimulation Using Online Parkinson’s Disease Prediction Driven by Dynamic Detection

Abstract

In Parkinson's disease (PD), on-demand deep brain stimulation is required so that stimulation is regulated to reduce side effects resulting from continuous stimulation and PD exacerbation due to untimely stimulation. Also, the progressive nature of PD necessitates the use of dynamic detection schemes that can track the nonlinearities in PD. This paper proposes the use of dynamic feature extraction and dynamic pattern classification to achieve dynamic PD detection taking into account the demand for high accuracy, low computation, and real-time detection. The dynamic feature extraction and dynamic pattern classification are selected by evaluating a subset of feature extraction, dimensionality reduction, and classification algorithms that have been used in brain-machine interfaces. A novel dimensionality reduction technique, the maximum ratio method (MRM) is proposed, which provides the most efficient performance. In terms of accuracy and complexity for hardware implementation, a combination having discrete wavelet transform for feature extraction, MRM for dimensionality reduction, and dynamic k-nearest neighbor for classification was chosen as the most efficient. It achieves a classification accuracy of 99.29%, an F1-score of 97.90%, and a choice probability of 99.86%.

An Adaptive Neural Spike Processor With Embedded Active Learning for Improved Unsupervised Sorting Accuracy

“…The chip core area 1 occupies 6 mm 2 . The processor uses four different clock rates (30 kHz, 1 It comprises 2.7 mm 2 non-training area and 3.3 mm 2 training area. The logic cells occupy 55% of the core area and the rest is for routing (only 4 metal layers are available in the 180-nm CMOS technology used).…”

Toward Adaptive Deep Brain Stimulation in Parkinson’s Disease: A Review

“…Classification algorithms can be computationally intensive and directly mapping states to therapy may not provide therapeutic benefit due to the non-linearity of the disease/disorder. [74,76,77,80,81] Internal Model Driven Input-output model may assist in adapting to changes in dynamics of disease or disorder.…”

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

“…Fitting an ARMA model provides the flexibility to manipulate the signal characteristics so that all underlying conditions can be represented. The complete LFP data synthesis process and a detailed description of the LFP recordings are provided in [8].…”

“…There are preliminary processing stages before classification; feature extraction and feature selection. The maximum ratio method (MRM) is used for feature and channel selection [8]. It has two operating clocks: clk1 and clk3.…”

“…This can be computationally intensive. Alternatively, the MRM estimates the most informative channel and uses the features from this channel (as against using all channels [8]). The complete procedure for feature selection using the MRM is described in [8].…”