Related Experiment Video
Updated: Jun 10, 2025
10:52
Microelectrode Guided Implantation of Electrodes into the Subthalamic Nucleus of Rats for Long-term Deep Brain Stimulation
Published on: October 2, 2015
19.6K
Machine learning for the localization of Subthalamic Nucleus during deep brain stimulation surgery: a systematic
Made Agus Mahendra Inggas1, Terry Coyne2, Takaomi Taira3
1Department of Neurosurgery, Universitas Pelita Harapan, Tangerang, Banten, Indonesia. made.inggas@lecturer.uph.edu.
Neurosurgical Review
|October 10, 2024
Summary
Artificial intelligence, specifically Hidden Markov Models (HMM), shows promise in accurately localizing the subthalamic nucleus (STN) for deep brain stimulation (DBS) surgery. This machine learning approach significantly improves diagnostic accuracy compared to other methods.
Area of Science:
- Neurosurgery
- Artificial Intelligence
- Biomedical Engineering
Background:
- Accurate subthalamic nucleus (STN) delineation is crucial for effective deep brain stimulation (DBS) electrode placement.
- Microelectrode recordings (MER) and trajectory history are key resources for neurosurgeons.
Purpose of the Study:
- To evaluate the application of artificial intelligence, specifically Hidden Markov Models (HMM), for STN localization.
- To assess the diagnostic performance of HMM in STN localization for DBS.
Main Methods:
- A systematic review of studies utilizing HMM for predicting DBS outcomes based on patient data.
- Searched PubMed, EuroPMC, and MEDLINE using keywords related to AI, machine learning, and DBS.
- Included English-language studies meeting specific inclusion criteria for HMM application.
Main Results:
- The Hidden Markov Model (HMM) achieved a high diagnostic odds ratio (DOR) of 838.677.
- K-Nearest Neighbors (KNN) and Support Vector Machine (SVM) models showed lower DORs of 25.151 and 13.959, respectively.
- The review incorporated 14 studies comparing various machine learning approaches for STN diagnosis.
Conclusions:
- Machine learning, particularly HMM, significantly aids in subthalamic nucleus diagnosis.
- While MER data shows variability, AI approaches offer improved accuracy for STN localization in DBS.
- The accuracy of machine learning methods for STN diagnosis varies, highlighting the potential of HMM.
