2026-05-16T07:15:43.671Z http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai

ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai:3:10799 2025-06-13T13:58:39Z 3

Parkinson’s Detection From Gait Time Series Classification Using LSTM Tuned by Modified RSA Algorithm, Chapter in LNNS Lecture Notes in Networks and Systems: ICCCT 2023: International Conference on Communication and Computational Technologies, Springer, volume 1121 2024 http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai/record/3/10799 https://link.springer.com/chapter/10.1007/978-981-97-7423-4_10 M. Zivkovic N. Bacanin T. Zivkovic L. Jovanovic J. Kaljevic M. Antonijevic Parkinson’s disease is an intricate neurological disorder characterized by the deterioration of neuronal function in the basal ganglia region of the brain and the loss of nerve endings. Although the precise cause of this condition is still debated, professionals believe that a mix of genetics and environment contributes to its development. Diagnosing Parkinson’s disease poses a significant challenge due to its gradual progression. The majority of patients only seek treatment when advanced symptoms emerge, causing uncontrollable tremors and involuntary movements that affect their quality of life. Although there is no effective treatment capable of reversing the neurological damage associated with the condition, there are treatments available that can slow its progression, consequently significantly alleviating the worst of symptoms for patients. This research delves into the use of long short-term memory neural networks to diagnose Parkinson’s disease by monitoring accelerometer sensors attached to shoes for early detection and diagnostics. Moreover, to achieve optimal performance, network hyperparameters are fine-tuned using an altered variant of the relatively recent reptile search algorithm. The effectiveness of this approach is assessed using real-world data, and the results appear promising when compared to other contemporary optimization methods. bookPart Springer, Singapore 119 134 10.1007/978-981-97-7423-4_10 LNNS Lecture Notes in Networks and Systems: ICCCT 2023: Proceedings of International Conference on Communication and Computational Technologies, volume 1121