Machine learning approaches applied in spinal pain research

Falla, Deborah; Devecchi, Valter; Jiménez-Grande, David; Rügamer, David; Liew, Bernard X. W.

doi:10.1016/j.jelekin.2021.102599

Cited by 9 publications

(10 citation statements)

References 122 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By examining the so far reported results and state-of-the-art, we are hypothesizing that an additional contribution to successful subgrouping of LBP patients can be supported by introducing more comprehensive multifactorial LBP classification models, thus also capturing more individually-biased expressions of LBP occurrences and avoiding a common pitfall of "one-size fits all" approach [5], [34]. A similar view was shared in a recent work by Fallaa et al [35] pointing to possibilities of ML to harness the variability of patients' presentations to enhance clinical predictions, as well as highlighting the importance of not relying on single features when characterizing patients, given the variability of physiological adaptations present in people with spinal pain. Moreover, current procedures leading to subgrouping of LBP patients did not lead to successfully tailored treatment approaches [36], thus, still waiting for more meaningful and interpretable systems able to capture the variability of physiological and neuromotor adaptations in LBP patients.…”

Section: Introductionmentioning

confidence: 74%

Feature Modeling for Interpretable Low Back Pain Classification Based on Surface EMG

2022

View full text Add to dashboard Cite

Low back pain (LBP) is a global health-problem phenomenon. Most patients are categorized as non-specific, thus requiring an individualized approach which still poses a major challenge. In this paper, sEMG recordings from two pairs of lumbar muscle sites were collected during an isometric trunk extension exercise. Ninety-one subjects were included in the study; 29 patients with non-specific chronic LBP (CLBP), 25 patients with radiculopathy (RLBP), and 37 control healthy subjects (HS). Six best-performing timedomain raw features were employed to model contextual secondary feature groups. Neuromuscular LBP characteristics were described with coordination, co-activation, trends, and fatigue measures. Altogether, a set of 327 secondary features was created where inputs into the classification models were further refined by employing neighborhood component analysis (NCA). NCA effectively reduced the number of features (<20 components), alongside preserving them in the original interpretable domain. A set of 23 different classifiers was employed and explored, resulting in classification accuracy of 0.94 for HS vs. LBP, 0.89 for HS vs. CLBP, 0.98 for HS vs. RLBP, and 0.89 for CLBP vs. RLBP differentiation. High median precision (0.97) and sensitivity (0.99) across all classifiers for HS vs. RLBP differentiation was obtained, with only three feature components utilized (out of 327). Support vector machines (SVM) and k-nearest neighbor (kNN) based classifiers consistently demonstrated best classification results. Different profiles of CLBP patients were presented and discussed. The suggested method demonstrated the potential for patients' subgrouping and subsequent more individualized rehabilitation treatments, backed by medical interpretations through feature modeling.

show abstract

Section: Introductionmentioning

confidence: 74%

Feature Modeling for Interpretable Low Back Pain Classification Based on Surface EMG

2022

View full text Add to dashboard Cite

show abstract

“…We also go beyond simple logistic regressions by including all relevant features in a single ML algorithm. Indeed, converting time series into scalar variables may remove a substantial amount of information contained in the original time series that could lead to extra false negative results or inaccurate predictions [ 53 ]. Roijezon et al used linear discriminant analysis to identify neck-pain patients, i.e., the same kind of methodology as ours, but obtained lower sensitivity and specificity than in the present study: they found a sensitivity of 74.6% and a specificity of 73.5% for classification based on head peak yaw angular velocity [ 12 ].…”

Section: Discussionmentioning

confidence: 99%

Head Pitch Angular Velocity Discriminates (Sub-)Acute Neck Pain Patients and Controls Assessed with the DidRen Laser Test

Hage

Buisseret

Houry

et al. 2022

Sensors

View full text Add to dashboard Cite

Understanding neck pain is an important societal issue. Kinematic data from sensors may help to gain insight into the pathophysiological mechanisms associated with neck pain through a quantitative sensorimotor assessment of one patient. The objective of this study was to evaluate the potential usefulness of artificial intelligence with several machine learning (ML) algorithms in assessing neck sensorimotor performance. Angular velocity and acceleration measured by an inertial sensor placed on the forehead during the DidRen laser test in thirty-eight acute and subacute non-specific neck pain (ANSP) patients were compared to forty-two healthy control participants (HCP). Seven supervised ML algorithms were chosen for the predictions. The most informative kinematic features were computed using Sequential Feature Selection methods. The best performing algorithm is the Linear Support Vector Machine with an accuracy of 82% and Area Under Curve of 84%. The best discriminative kinematic feature between ANSP patients and HCP is the first quartile of head pitch angular velocity. This study has shown that supervised ML algorithms could be used to classify ANSP patients and identify discriminatory kinematic features potentially useful for clinicians in the assessment and monitoring of the neck sensorimotor performance in ANSP patients.

show abstract

“…On average, the PCS ratings for the subjects are within the mild range (21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33)(34)(35)(36)(37)(38)(39)(40), and the PASS scores were not significantly different when comparing surgical status or sex differences. In the context of these clinical findings, our results that both CNN and GCNN models constructed based on patients from one subgroup (surgical vs. non-surgical, male vs. female) can decode the patients from the other subgroup can be seen as reflecting shared neural substrate rather than driven by differences in clinical conditions.…”

Section: Clinical Considerationsmentioning

confidence: 96%

“…to differentiate healthy control subjects from patients with fibromyalgia and CBP, 21,22 DNNs have only begun to see applications in chronic pain studies. 23 The present study is the first to train DNN models to predict the fluctuating pain ratings from fMRI data in TN and to derive the signature centers of TN pain from these models. Our overall goal is to seek converging evidence on the neural substrate of TN pain by combining different methods and to uncover novel insights not possible with the conventional methods.…”

Section: Introductionmentioning

confidence: 99%

Imaging the Neural Substrate of Trigeminal Neuralgia Pain Using Deep Learning

Liang

Zhao

et al. 2022

Preprint

View full text Add to dashboard Cite

Trigeminal neuralgia (TN) is a severe and disabling facial pain condition and is characterized by intermittent, severe, electric shock-like pain in one (or more) trigeminal subdivisions. This pain can be triggered by an innocuous stimulus or can be spontaneous. Presently available therapies for TN include both surgical and pharmacological management; however, the lack of a known etiology for TN contributes to the unpredictable response to treatment and the variability in long-term clinical outcomes. Given this, a range of peripheral and central mechanisms underlying TN pain remain to be understood. We acquired functional magnetic resonance imaging (fMRI) data from TN patients who (1) rested comfortably in the scanner during a resting state session and (2) rated their pain levels in real time using a calibrated tracking ball-controlled scale in a pain tracking session. Following data acquisition, the data was analyzed using the conventional correlation analysis and two artificial intelligence (AI)-inspired deep learning methods: convolutional neural network (CNN) and graph convolutional neural network (GCNN). Each of the three methods yielded a set of brain regions related to the generation and perception of pain in TN. There were six regions that were identified by all three methods, including the superior temporal cortex, the insula, the fusiform, the precentral gyrus, the superior frontal gyrus, and the supramarginal gyrus. Additionally, 17 regions, including dorsal anterior cingulate cortex(dACC) and the thalamus, were identified by at least two of the three methods. Collectively, these 23 regions represent signature centers of TN pain and provide target areas for future studies relating to central mechanisms of TN.

show abstract

Machine learning approaches applied in spinal pain research

Cited by 9 publications

References 122 publications

Feature Modeling for Interpretable Low Back Pain Classification Based on Surface EMG

Feature Modeling for Interpretable Low Back Pain Classification Based on Surface EMG

Head Pitch Angular Velocity Discriminates (Sub-)Acute Neck Pain Patients and Controls Assessed with the DidRen Laser Test

Imaging the Neural Substrate of Trigeminal Neuralgia Pain Using Deep Learning

Contact Info

Product

Resources

About