Machine learning in deep brain stimulation: A systematic review

STN localization using local field potentials based on wavelet packet features and stacking ensemble learning

BACKGROUND

DBS entails the insertion of an electrode into the patient brain, enabling Subthalamic nucleus (STN) stimulation. Accurate delineation of STN borders is a critical but time-consuming task, traditionally reliant on the neurosurgeon experience in deciphering the intricacies of microelectrode recording (MER). While clinical outcomes of MER have been satisfactory, they involve certain risks to patient safety. Recently, there has been a growing interest in exploring the potential of local field potentials (LFP) due to their correlation with the STN motor territory.

METHOD

A novel STN detection system, integrating LFP and wavelet packet transform (WPT) with stacking ensemble learning, is developed. Initial steps involve the inclusion of soft thresholding to increase robustness to LFP variability. Subsequently, non-linear WPT features are extracted. Finally, a unique ensemble model, comprising a dual-layer structure, is developed for STN localization. We harnessed the capabilities of support vector machine, Decision tree and k-Nearest Neighbour in conjunction with long short-term memory (LSTM) network. LSTM is pivotal for assigning adequate weights to every base model.

RESULTS

Results reveal that the proposed model achieved a remarkable accuracy and F1-score of 89.49% and 91.63%.

COMPARISON WITH EXISTING METHODS

Ensemble model demonstrated superior performance when compared to standalone base models and existing meta techniques.

CONCLUSION

This framework is envisioned to enhance the efficiency of DBS surgery and reduce the reliance on clinician experience for precise STN detection. This achievement is strategically significant to serve as an invaluable tool for refining the electrode trajectory, potentially replacing the current methodology based on MER.

Systematic reviews of machine learning in healthcare: a literature review

INTRODUCTION

The increasing availability of data and computing power has made machine learning (ML) a viable approach to faster, more efficient healthcare delivery.

METHODS

A systematic literature review (SLR) of published SLRs evaluating ML applications in healthcare settings published between1 January 2010 and 27 March 2023 was conducted.

RESULTS

In total 220 SLRs covering 10,462 ML algorithms were reviewed. The main application of AI in medicine related to the clinical prediction and disease prognosis in oncology and neurology with the use of imaging data. Accuracy, specificity, and sensitivity were provided in 56%, 28%, and 25% SLRs respectively. Internal and external validation was reported in 53% and less than 1% of the cases respectively. The most common modeling approach was neural networks (2,454 ML algorithms), followed by support vector machine and random forest/decision trees (1,578 and 1,522 ML algorithms, respectively).

EXPERT OPINION

The review indicated considerable reporting gaps in terms of the ML's performance, both internal and external validation. Greater accessibility to healthcare data for developers can ensure the faster adoption of ML algorithms into clinical practice.

From dawn till dusk: Time-adaptive bayesian optimization for neurostimulation

Stimulation optimization has garnered considerable interest in recent years in order to efficiently parametrize neuromodulation-based therapies. To date, efforts focused on automatically identifying settings from parameter spaces that do not change over time. A limitation of these approaches, however, is that they lack consideration for time dependent factors that may influence therapy outcomes. Disease progression and biological rhythmicity are two sources of variation that may influence optimal stimulation settings over time. To account for this, we present a novel time-varying Bayesian optimization (TV-BayesOpt) for tracking the optimum parameter set for neuromodulation therapy. We evaluate the performance of TV-BayesOpt for tracking gradual and periodic slow variations over time. The algorithm was investigated within the context of a computational model of phase-locked deep brain stimulation for treating oscillopathies representative of common movement disorders such as Parkinson's disease and Essential Tremor. When the optimal stimulation settings changed due to gradual and periodic sources, TV-BayesOpt outperformed standard time-invariant techniques and was able to identify the appropriate stimulation setting. Through incorporation of both a gradual "forgetting" and periodic covariance functions, the algorithm maintained robust performance when a priori knowledge differed from observed variations. This algorithm presents a broad framework that can be leveraged for the treatment of a range of neurological and psychiatric conditions and can be used to track variations in optimal stimulation settings such as amplitude, pulse-width, frequency and phase for invasive and non-invasive neuromodulation strategies.

Machine learning for adaptive deep brain stimulation in Parkinson's disease: closing the loop

Parkinson's disease (PD) is the second most common neurodegenerative disease bearing a severe social and economic impact. So far, there is no known disease modifying therapy and the current available treatments are symptom oriented. Deep Brain Stimulation (DBS) is established as an effective treatment for PD, however current systems lag behind today's technological potential. Adaptive DBS, where stimulation parameters depend on the patient's physiological state, emerges as an important step towards "smart" DBS, a strategy that enables adaptive stimulation and personalized therapy. This new strategy is facilitated by currently available neurotechnologies allowing the simultaneous monitoring of multiple signals, providing relevant physiological information. Advanced computational models and analytical methods are an important tool to explore the richness of the available data and identify signal properties to close the loop in DBS. To tackle this challenge, machine learning (ML) methods applied to DBS have gained popularity due to their ability to make good predictions in the presence of multiple variables and subtle patterns. ML based approaches are being explored at different fronts such as the identification of electrophysiological biomarkers and the development of personalized control systems, leading to effective symptom relief. In this review, we explore how ML can help overcome the challenges in the development of closed-loop DBS, particularly its role in the search for effective electrophysiology biomarkers. Promising results demonstrate ML potential for supporting a new generation of adaptive DBS, with better management of stimulation delivery, resulting in more efficient and patient-tailored treatments.

Deep Brain Stimulation beyond the Clinic: Navigating the Future of Parkinson's and Alzheimer's Disease Therapy

Deep brain stimulation (DBS) is a surgical procedure that uses electrical neuromodulation to target specific regions of the brain, showing potential in the treatment of neurodegenerative disorders such as Parkinson's disease (PD) and Alzheimer's disease (AD). Despite similarities in disease pathology, DBS is currently only approved for use in PD patients, with limited literature on its effectiveness in AD. While DBS has shown promise in ameliorating brain circuits in PD, further research is needed to determine the optimal parameters for DBS and address any potential side effects. This review emphasizes the need for foundational and clinical research on DBS in different brain regions to treat AD and recommends the development of a classification system for adverse effects. Furthermore, this review suggests the use of either a low-frequency system (LFS) or high-frequency system (HFS) depending on the specific symptoms of the patient for both PD and AD.

Discovering Themes in Deep Brain Stimulation Research Using Explainable Artificial Intelligence

Deep brain stimulation is a treatment that controls symptoms by changing brain activity. The complexity of how to best treat brain dysfunction with deep brain stimulation has spawned research into artificial intelligence approaches. Machine learning is a subset of artificial intelligence that uses computers to learn patterns in data and has many healthcare applications, such as an aid in diagnosis, personalized medicine, and clinical decision support. Yet, how machine learning models make decisions is often opaque. The spirit of explainable artificial intelligence is to use machine learning models that produce interpretable solutions. Here, we use topic modeling to synthesize recent literature on explainable artificial intelligence approaches to extracting domain knowledge from machine learning models relevant to deep brain stimulation. The results show that patient classification (i.e., diagnostic models, precision medicine) is the most common problem in deep brain stimulation studies that employ explainable artificial intelligence. Other topics concern attempts to optimize stimulation strategies and the importance of explainable methods. Overall, this review supports the potential for artificial intelligence to revolutionize deep brain stimulation by personalizing stimulation protocols and adapting stimulation in real time.

Radiomic features of amygdala nuclei and hippocampus subfields help to predict subthalamic deep brain stimulation motor outcomes for Parkinson‘s disease patients

Background and purpose

The aim of the study is to predict the subthalamic nucleus (STN) deep brain stimulation (DBS) outcomes for Parkinson’s disease (PD) patients using the radiomic features extracted from pre-operative magnetic resonance images (MRI).

Methods

The study included 34 PD patients who underwent DBS implantation in the STN. Five patients (15%) showed poor DBS motor outcome. All together 9 amygdalar nuclei and 12 hippocampus subfields were segmented using Freesurfer 7.0 pipeline from pre-operative MRI images. Furthermore, PyRadiomics platform was used to extract 120 radiomic features for each nuclei and subfield resulting in 5,040 features. Minimum Redundancy Maximum Relevance (mRMR) feature selection method was employed to reduce the number of features to 20, and 8 machine learning methods (regularized binary logistic regression (LR), decision tree classifier (DT), linear discriminant analysis (LDA), naive Bayes classifier (NB), kernel support vector machine (SVM), deep feed-forward neural network (DNN), one-class support vector machine (OC-SVM), feed-forward neural network-based autoencoder for anomaly detection (DNN-A)) were applied to build the models for poor vs. good and very good STN-DBS motor outcome prediction.

Results

The highest mean prediction accuracy was obtained using regularized LR (96.65 ± 7.24%, AUC 0.98 ± 0.06) and DNN (87.25 ± 14.80%, AUC 0.87 ± 0.18).

Conclusion

The results show the potential power of the radiomic features extracted from hippocampus and amygdala MRI in the prediction of STN-DBS motor outcomes for PD patients.