TY - GEN
T1 - An Investigative Study Exploring Machine Learning Approaches for Optimizing Deep Brain Stimulation Programming
AU - Ademola, Esther
AU - Reich, Martin
AU - Gregorova, Magda
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.
PY - 2025
Y1 - 2025
N2 - This investigative study explores machine learning models for predicting dystonia improvement scores in Deep Brain Stimulation (DBS). Leveraging data from 85 subjects across seven European DBS centers, we employ various linear and non-linear modeling approaches. In contrast to previous studies utilizing probabilistic mapping, our direct utilization of the actual dataset yields improved results. The random forest model emerges as the most accurate predictor, with a mean deviation of 9.54 ± 6.08%. This implies that for a patient with an improvement score of 78%, the model predicts an improvement between 68% and 87%. This advancement in predictive accuracy holds potential implications for refining DBS programming, ultimately enhancing therapeutic outcomes for individuals with dystonia. In addition, regularization techniques play a pivotal role in determining feature importance thereby contributing to a nuanced understanding of factors influencing DBS therapy outcomes.
AB - This investigative study explores machine learning models for predicting dystonia improvement scores in Deep Brain Stimulation (DBS). Leveraging data from 85 subjects across seven European DBS centers, we employ various linear and non-linear modeling approaches. In contrast to previous studies utilizing probabilistic mapping, our direct utilization of the actual dataset yields improved results. The random forest model emerges as the most accurate predictor, with a mean deviation of 9.54 ± 6.08%. This implies that for a patient with an improvement score of 78%, the model predicts an improvement between 68% and 87%. This advancement in predictive accuracy holds potential implications for refining DBS programming, ultimately enhancing therapeutic outcomes for individuals with dystonia. In addition, regularization techniques play a pivotal role in determining feature importance thereby contributing to a nuanced understanding of factors influencing DBS therapy outcomes.
KW - Deep brain stimulation
KW - dystonia
KW - machine learning
KW - regularization
UR - https://www.scopus.com/pages/publications/105005934958
U2 - 10.1007/978-3-031-87386-7_6
DO - 10.1007/978-3-031-87386-7_6
M3 - Conference contribution
AN - SCOPUS:105005934958
SN - 9783031873850
T3 - Communications in Computer and Information Science
SP - 75
EP - 89
BT - Modelling and Development of Intelligent Systems - 9th International Conference, MDIS 2024, Revised Selected Papers
A2 - Simian, Dana
A2 - Stoica, Laura Florentina
PB - Springer Science and Business Media Deutschland GmbH
T2 - 9th International Conference on Modelling and Development of Intelligent Systems, MDIS 2024
Y2 - 17 October 2024 through 19 October 2024
ER -