A data science approach for reliable classification of neuro-degenerative diseases using gait patterns

Neuro-degenerative diseases (NDD) continue to increase globally and have significant impact on health, developmental and financial fronts. Recent studies have shown that gait impairment as one of the earliest signs of the disease. However, classification of multiple types of NDD becomes more challenging because of the high overlapping symptoms specifically at early stages. This paper entails a composite of signal processing and machine intelligence algorithms to process the gait data captured through multi-sensors for a reliable classification different types of NDD. The captured dataset used in this research consisted of 60 patients’ records representing three different types of NDD. Our simulation results indicated that the proposed approach outperformed existing works in this domain. The proposed work might help the mitigation plans for NDD, reliable monitoring of the disease progression and can assist the evaluation of possible therapy and treatments that would benefit the individuals, associated families, society and healthcare services.