| 1. |
- Lendaro, Eva, 1989, et al.
(author)
-
Common Spatial Pattern EEG decomposition for Phantom Limb Pain detection
- 2021
-
In: Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. - 1557-170X. ; , s. 726-729
-
Conference paper (peer-reviewed)abstract
- Phantom Limb Pain (PLP) is a chronic condition frequent among individuals with acquired amputation. PLP has been often investigated with the use of functional MRI focusing on the changes that take place in the sensorimotor cortex after amputation. In the present study, we investigated whether a different type of data, namely electroencephalographic (EEG) recordings, can be used to study the condition. We acquired resting state EEG data from people with and without PLP and then used machine learning for a binary classification task that differentiates the two. Common Spatial Pattern (CSP) decomposition was used as the feature extraction method and two validation schemes were followed for the classification task. Six classifiers (LDA, Log, QDA, LinearSVC, SVC and RF) were optimized through grid search and their performance compared. Two validation approaches, namely all-subjects validation and leave-one-out cross-validation (LOOCV), resulted in high classification accuracy. Most notably, the 93.7% accuracy achieved with SVC in LOOCV holds promise for good diagnostic capabilities using EEG biomarkers. In conclusion, our findings indicate that EEG data is a promising target for future research aiming at elucidating the neural mechanisms underlying PLP and its diagnosis.
|
|
| 2. |
- Lendaro, Eva, 1989
(author)
-
Investigating Phantom Motor Execution as treatment of Phantom Limb Pain
- 2021
-
Doctoral thesis (other academic/artistic)abstract
- Phantom Limb Pain (PLP) is commonly suffered by people with amputations and even though it has been studied for centuries, it remains a mysterious object of debate among researchers. For one thing, despite the vast number of proposed PLP treatments, no therapy has so far proved to be reliably effective. For another, studies attempting to provide a mechanistic explanation of the condition have produced mixed and inconsistent results, thus providing unreliable guidance for devising new treatment approaches. Phantom Motor Execution (PME) – exertion of voluntary phantom limb movements – aims at restoring control over the phantom limb and the exercise of such control has been hypothesized to reverse neural changes implicated in PLP. Preliminary evidence supporting this hypothesis has been provided by clinical investigations on upper limb amputees. The main purpose of this doctoral thesis was to provide high quality and unbiased evidence for the use of PME as a treatment of PLP, by probing its efficacy with a Randomized Controlled Trial (RCT) on both upper and lower limb amputees. However, the implementation of this clinical investigation required of additional technology development related the extraction of motor volition via Myoelectric Pattern Recognition (MPR). In practice, this doctoral work consisted in the extension of PME technology to lower limb amputations by proposing and validating a new and more user-friendly recording method to acquire myoelectric signals. The use of PME was then shown to be efficacious in relieving PLP even in the lower limb population with a case study. Another necessity for providing unbiased evidence was to ensure that the highest standards were met when designing, conducting, analysing and reporting the results of the RCT. For this reason, the protocol for the RCT and the prospective Statistical Analysis Plan (SAP) were designed and published. The RCT was established as an international, multi-center effort in 2017 and it is expected to reach its conclusion in September 2021. Preliminary results of the RCT regarding the primary outcome showed reduction of PLP above what is considered clinically relevant, and whereas a higher reduction was obtained with PME, this was not statistically significant over the control treatment. The available evidence at this stage indicates that the RCT will not be able to rule out the role of contextual factors other than PME in providing pain relief. Having at hand a way to alleviate PLP provided a unique opportunity to investigate and identify its neural correlates, therefore this became a secondary aim of this thesis. In particular, patients suffering from PLP were followed regarding their pain trajectory through the therapy and brain imaging studies with functional Magnetic Resonance Imaging (fMRI) and electroencephalography (EEG) were performed. The present doctoral thesis reports part of this work by showing the early results of a cross-sectional study on the EEG correlates of PLP. The results show that it is possible to use machine-learning techniques to discriminate EEG recorded from patients with and without PLP. The findings further point to this technique as a promising target for future longitudinal research aiming at elucidating the neural mechanisms underlying PLP.
|
|
