Dataset Information

Machine Learning Methods Predict Individual Upper-Limb Motor Impairment Following Therapy in Chronic Stroke.

ABSTRACT: Background. Accurate prediction of clinical impairment in upper-extremity motor function following therapy in chronic stroke patients is a difficult task for clinicians but is key in prescribing appropriate therapeutic strategies. Machine learning is a highly promising avenue with which to improve prediction accuracy in clinical practice. Objectives. The objective was to evaluate the performance of 5 machine learning methods in predicting postintervention upper-extremity motor impairment in chronic stroke patients using demographic, clinical, neurophysiological, and imaging input variables. Methods. A total of 102 patients (female: 31%, age 61 ± 11 years) were included. The upper-extremity Fugl-Meyer Assessment (UE-FMA) was used to assess motor impairment of the upper limb before and after intervention. Elastic net (EN), support vector machines, artificial neural networks, classification and regression trees, and random forest were used to predict postintervention UE-FMA. The performances of methods were compared using cross-validated R2. Results. EN performed significantly better than other methods in predicting postintervention UE-FMA using demographic and baseline clinical data (median REN2=0.91,RRF2=0.88,RANN2=0.83,RSVM2=0.79,RCART2=0.70; P < .05). Preintervention UE-FMA and the difference in motor threshold (MT) between the affected and unaffected hemispheres were the strongest predictors. The difference in MT had greater importance than the absence or presence of a motor-evoked potential (MEP) in the affected hemisphere. Conclusion. Machine learning methods may enable clinicians to accurately predict a chronic stroke patient's postintervention UE-FMA. Interhemispheric difference in the MT is an important predictor of chronic stroke patients' response to therapy and, therefore, could be included in prospective studies.

SUBMITTER: Tozlu C

PROVIDER: S-EPMC7217740 | biostudies-literature | 2020 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine Learning Methods Predict Individual Upper-Limb Motor Impairment Following Therapy in Chronic Stroke.

Tozlu Ceren C Edwards Dylan D Boes Aaron A Labar Douglas D Tsagaris K Zoe KZ Silverstein Joshua J Pepper Lane Heather H Sabuncu Mert R MR Liu Charles C Kuceyeski Amy A

Neurorehabilitation and neural repair 20200320 5

<i>Background</i>. Accurate prediction of clinical impairment in upper-extremity motor function following therapy in chronic stroke patients is a difficult task for clinicians but is key in prescribing appropriate therapeutic strategies. Machine learning is a highly promising avenue with which to improve prediction accuracy in clinical practice. <i>Objectives</i>. The objective was to evaluate the performance of 5 machine learning methods in predicting postintervention upper-extremity motor impa ...[more]

PMID: 32193984

Similar Datasets

Project description:IntroductionLongitudinal observations of upper limb motor recovery after stroke have suggested that certain subgroups may exhibit distinct recovery patterns. Here we sought to examine whether the predictive ability for post-stroke upper limb motor outcomes could be enhanced by applying conventional stratification strategies.MethodFor 60 individuals who suffered the first stroke, upper limb motor impairment was assessed with the upper extremity Fugl-Meyer assessment (UE-FMA) at 2 weeks as a baseline and then 3 months post-stroke. Brain structural damage at baseline was assessed by MRI data-derived markers ranging from traditional lesion size to the lesion load and to the disconnectome. Linear regression models for predicting upper limb motor outcomes (UE-FMA score at 3 months post-stroke) based on baseline upper limb motor impairment (UE-FMA score at 2 weeks post-stroke), brain structural damage, and their combinations were generated, and those with the best predictive performance were determined for individual subgroups stratified according to initial impairment (severe and non-severe), lesion location (cortical and non-cortical), and neurophysiological status (motor evoked potential-positive and motor evoked potential-negative).ResultsThe best predictions were made by baseline upper limb motor impairment alone for subgroups with less functional impairment (non-severe) or less structural involvement (non-cortical), but by the combination of baseline upper limb motor impairment and brain structural damage for the other subgroups. The predictive models tailored for subgroups determined according to initial impairment and neurophysiological status yielded a smaller overall error than that for the whole group in upper limb motor outcome predictions.DiscussionThe predictive ability for upper limb motor outcomes could be enhanced beyond the one-size-fits-all model for all individuals with stroke by applying specific stratification strategies, with stratification according to initial impairment being the most promising. We expect that predictive models tailored for individual subgroups could lead closer to the personalized prognosis of upper limb motor outcomes after stroke.

Project description:BackgroundEffective rehabilitative therapies are needed for patients with long-term deficits after stroke.MethodsIn this multicenter, randomized, controlled trial involving 127 patients with moderate-to-severe upper-limb impairment 6 months or more after a stroke, we randomly assigned 49 patients to receive intensive robot-assisted therapy, 50 to receive intensive comparison therapy, and 28 to receive usual care. Therapy consisted of 36 1-hour sessions over a period of 12 weeks. The primary outcome was a change in motor function, as measured on the Fugl-Meyer Assessment of Sensorimotor Recovery after Stroke, at 12 weeks. Secondary outcomes were scores on the Wolf Motor Function Test and the Stroke Impact Scale. Secondary analyses assessed the treatment effect at 36 weeks.ResultsAt 12 weeks, the mean Fugl-Meyer score for patients receiving robot-assisted therapy was better than that for patients receiving usual care (difference, 2.17 points; 95% confidence interval [CI], -0.23 to 4.58) and worse than that for patients receiving intensive comparison therapy (difference, -0.14 points; 95% CI, -2.94 to 2.65), but the differences were not significant. The results on the Stroke Impact Scale were significantly better for patients receiving robot-assisted therapy than for those receiving usual care (difference, 7.64 points; 95% CI, 2.03 to 13.24). No other treatment comparisons were significant at 12 weeks. Secondary analyses showed that at 36 weeks, robot-assisted therapy significantly improved the Fugl-Meyer score (difference, 2.88 points; 95% CI, 0.57 to 5.18) and the time on the Wolf Motor Function Test (difference, -8.10 seconds; 95% CI, -13.61 to -2.60) as compared with usual care but not with intensive therapy. No serious adverse events were reported.ConclusionsIn patients with long-term upper-limb deficits after stroke, robot-assisted therapy did not significantly improve motor function at 12 weeks, as compared with usual care or intensive therapy. In secondary analyses, robot-assisted therapy improved outcomes over 36 weeks as compared with usual care but not with intensive therapy. (ClinicalTrials.gov number, NCT00372411.)

Project description:IntroductionStroke-induced upper limb disabilities can be characterized by both motor impairments and activity limitations, commonly assessed using Fugl-Meyer Motor Assessment for Upper Extremity (FMMA-UE) and Action Research Arm Test (ARAT), respectively. The relationship between the two assessments during recovery is largely unstudied. Expectedly they diverge over time when recovery of impairment (restitution) plateaus, but compensation-driven improvements still occur. The objective of this study is to evaluate the alignment between FMMA-UE and ARAT in defining upper limb functional recovery categories by ARAT scores. We aimed to establish cut-off scores for both measures from the acute/early subacute, subacute and chronic stages of stroke recovery.MethodsSecondary analysis of four prospective cohort studies (acute/early subacute: n = 133, subacute: n = 113, chronic: n = 92) stages post-stroke. Receiver operating characteristic curves calculated the area under the curve (AUC) to establish optimal FMMA-UE cut-offs based on predefined ARAT thresholds distinguishing five activity levels from no activity to full activity. Weighted kappa was used to determine agreement between the two assessments. We used minimally clinically important difference (MCID) and minimal detectable change (MDC95) for comparison.ResultsFMMA-UE and ARAT scores showed no relevant divergence across all recovery stages. Results indicated similar cut-off scores in all recovery stages with variability below MCID and MDC95 levels. Cut-off scores demonstrated robust AUC values from 0.77 to 0.86 at every recovery stage. Only in highly functional patients at the chronic stage, we found a reduced specificity of 0.55. At all other times sensitivity ranged between 0.68 and 0.99 and specificity between 0.71 and 0.99. Weighted kappa at the acute/early subacute, subacute and chronic stages was 0.76, 0.83, and 0.81, respectively.DiscussionOur research shows a strong alignment between FMMA-UE and ARAT cut-off scores throughout stroke recovery, except among the subgroup of highly recovered patients at the chronic stage. Discrepancies in specificity potentially stem from fine motor deficits affecting dexterity outcomes that are not captured by FMMA-UE. Additionally, the high congruence of both measures suggests they are not suited to distinguish between restitution and compensation. Calling for more comprehensive assessment methods to better understand upper limb functionality in rehabilitation.

Project description:BackgroundThe recovery of upper extremity motor impairment after stroke remains a challenging task. The clinical effectiveness of repetitive transcranial magnetic stimulation (rTMS), which is believed to aid in the recovery process, is still uncertain.MethodsA systematic search was conducted in Medline (Ovid), Cochrane and Embase electronic databases from March 28, 2014, to March 28, 2023. The inclusion criteria consisted of randomized controlled trials that assessed the effects of rTMS on the recovery of upper limb motor impairment among stroke patients. Various measurements, including the Fugl Meyer Assessment Upper Extremity Scale (FMA-UE), Brunnstrom recovery stage, Action Research Arm Test (ARAT), and Barthel index, were evaluated both before and after the intervention.ResultsNineteen articles with 865 patients were included. When considering only the rTMS parameters, both inhibitory and excitatory rTMS improved FMA-UE (MD = 1.87, 95% CI = [0.88]-[2.86], p < 0.001) and Barthel index (MD = 9.73, 95% CI = [4.57]-[14.89], p < 0.001). When considering only the severity of upper limb hemiplegia, both less severe (MD = 1.56, 95% CI = [0.64]-[2.49], p < 0.001) and severe (MD = 2.05, 95% CI = [1.09]-[3.00], p < 0.001) hemiplegia benefited from rTMS based on FMA-UE. However, when considering the rTMS parameters, severity of hemiplegia and stroke stages simultaneously, inhibitory rTMS was found to be significantly effective for less severe hemiplegia in the acute and subacute phases (MD = 4.55, 95% CI = [2.49]-[6.60], p < 0.001), but not in the chronic phase based on FMA-UE. For severe hemiplegia, inhibitory rTMS was not significantly effective in the acute and subacute phases, but significantly effective in the chronic phase (MD = 2.10, 95% CI = [0.75]-[3.45], p = 0.002) based on FMA-UE. Excitatory rTMS was found to be significantly effective for less severe hemiplegia in the acute and subacute phases (MD = 1.93, 95% CI = [0.58]-[3.28], p = 0.005) based on FMA-UE. The improvements in Brunnstrom recovery stage and ARAT need further research.ConclusionThe effectiveness of rTMS depends on its parameters, severity of hemiplegia, and stroke stages. It is important to consider all these factors together, as any single grouping method is incomplete.

Project description:Upper limb motor deficits in severe stroke survivors often remain unresolved over extended time periods. Novel neurotechnologies have the potential to significantly support upper limb motor restoration in severely impaired stroke individuals. Here, we review recent controlled clinical studies and reviews focusing on the mechanisms of action and effectiveness of single and combined technology-aided interventions for upper limb motor rehabilitation after stroke, including robotics, muscular electrical stimulation, brain stimulation and brain computer/machine interfaces. We aim at identifying possible guidance for the optimal use of these new technologies to enhance upper limb motor recovery especially in severe chronic stroke patients. We found that the current literature does not provide enough evidence to support strict guidelines, because of the variability of the procedures for each intervention and of the heterogeneity of the stroke population. The present results confirm that neurotechnology-aided upper limb rehabilitation is promising for severe chronic stroke patients, but the combination of interventions often lacks understanding of single intervention mechanisms of action, which may not reflect the summation of single intervention's effectiveness. Stroke rehabilitation is a long and complex process, and one single intervention administrated in a short time interval cannot have a large impact for motor recovery, especially in severely impaired patients. To design personalized interventions combining or proposing different interventions in sequence, it is necessary to have an excellent understanding of the mechanisms determining the effectiveness of a single treatment in this heterogeneous population of stroke patients. We encourage the identification of objective biomarkers for stroke recovery for patients' stratification and to tailor treatments. Furthermore, the advantage of longitudinal personalized trial designs compared to classical double-blind placebo-controlled clinical trials as the basis for precise personalized stroke rehabilitation medicine is discussed. Finally, we also promote the necessary conceptual change from 'one-suits-all' treatments within in-patient clinical rehabilitation set-ups towards personalized home-based treatment strategies, by adopting novel technologies merging rehabilitation and motor assistance, including implantable ones.

Project description:BackgroundLower socioeconomic status (SES) is associated with higher mortality rates and the likelihood of receiving less evidence-based treatment after stroke. In contrast, little is known about the impact of SES on recovery after discharge from inpatient rehabilitation. The aim of this study was to investigate the influence of SES on long-term recovery after stroke.Patients and methodsIn a prospective, observational, multicentre study, inpatients were recruited towards the end of rehabilitation. The 12-month follow-up focussed on upper limb motor recovery, measured by the Fugl-Meyer score. A clinically relevant improvement of ≥5.25 points was considered recovery. Patient-centric measures such as the Patient-reported Outcomes Measurement Information System-Physical Health (PROMIS-10 PH) provided secondary outcomes. Information on schooling, vocational training, income and occupational status pre-stroke entered a multidimensional SES index. Multivariate logistic regression models calculating odds ratios (ORs) and corresponding confidence intervals (CIs) were applied. SES was added to an initial model including age, sex and baseline neurological deficit. Additional exploratory analyses examined the association between SES and outpatient treatment.ResultsOne hundred and seventy-six patients were enrolled of whom 98 had SES and long-term recovery data. Model comparisons showed the SES-model superior to the initial model (Akaike information criterion (AIC): 123 vs. 120, Pseudo R2: 0.09 vs. 0.13). The likelihood of motor recovery (OR = 17.12, 95%CI = 1.31; 224.18) and PROMIS-10 PH improvement (OR = 20.76, 95%CI = 1.28; 337.11) were significantly increased with higher SES, along with more frequent use of outpatient therapy (p = .02).ConclusionsHigher pre-stroke SES is associated with better long-term recovery after discharge from rehabilitation. Understanding these factors can improve outpatient long-term stroke care and lead to better recovery.KEY MESSAGEHigher pre-stroke socioeconomic status (SES) is associated with better long-term recovery after discharge from rehabilitation both in terms of motor function and self-reported health status.Higher SES is associated with significantly higher utilization of outpatient therapies.Discharge management of rehabilitation clinics should identify and address socioeconomic factors in order to detect individual needs and to improve outpatient recovery. Article registration: clinicaltrials.gov NCT04119479.

Project description:BackgroundUpper limb kinematic assessments provide quantifiable information on qualitative movement behavior and limitations after stroke. A comprehensive characterization of spatiotemporal kinematics of stroke subjects during upper limb daily living activities is lacking. Herein, kinematic expressions were investigated with respect to different movement types and impairment levels for the entire task as well as for motion subphases.MethodChronic stroke subjects with upper limb movement impairments and healthy subjects performed a set of daily living activities including gesture and grasp movements. Kinematic measures of trunk displacement, shoulder flexion/extension, shoulder abduction/adduction, elbow flexion/extension, forearm pronation/supination, wrist flexion/extension, movement time, hand peak velocity, number of velocity peaks (NVP), and spectral arc length (SPARC) were extracted for the whole movement as well as the subphases of reaching distally and proximally. The effects of the factors gesture versus grasp movements, and the impairment level on the kinematics of the whole task were tested. Similarities considering the metrics expressions and relations were investigated for the subphases of reaching proximally and distally between tasks and subgroups.ResultsData of 26 stroke and 5 healthy subjects were included. Gesture and grasp movements were differently expressed across subjects. Gestures were performed with larger shoulder motions besides higher peak velocity. Grasp movements were expressed by larger trunk, forearm, and wrist motions. Trunk displacement, movement time, and NVP increased and shoulder flexion/extension decreased significantly with increased impairment level. Across tasks, phases of reaching distally were comparable in terms of trunk displacement, shoulder motions and peak velocity, while reaching proximally showed comparable expressions in trunk motions. Consistent metric relations during reaching distally were found between shoulder flexion/extension, elbow flexion/extension, peak velocity, and between movement time, NVP, and SPARC. Reaching proximally revealed reproducible correlations between forearm pronation/supination and wrist flexion/extension, movement time and NVP.ConclusionSpatiotemporal differences between gestures versus grasp movements and between different impairment levels were confirmed. The consistencies of metric expressions during movement subphases across tasks can be useful for linking kinematic assessment standards and daily living measures in future research and performing task and study comparisons.Trial registrationClinicalTrials.gov Identifier NCT03135093. Registered 26 April 2017, https://clinicaltrials.gov/ct2/show/NCT03135093 .

Dataset Information

Machine Learning Methods Predict Individual Upper-Limb Motor Impairment Following Therapy in Chronic Stroke.

Publications

Machine Learning Methods Predict Individual Upper-Limb Motor Impairment Following Therapy in Chronic Stroke.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets