RESUMEN
Decision making is often necessary before performing an action. Traditionally, it has been assumed that decision making and motor control are independent, sequential processes. Ogasa et al. challenge this view, and demonstrate that the decision-making process significantly impacts on the formation and retrieval of motor memory by tagging it with the level of confidence.
RESUMEN
Background: Reaching movements are crucial for daily living and rehabilitation, for which Fitts' Law describes a speed-accuracy trade-off that movement time increases with task difficulty. This study aims to investigate whether cortical activation in motor-related areas is directly linked to task difficulty as defined by Fitts' Law. Understanding this relationship provides a physiological basis for parameter selection in therapeutic exercises. Methods: Sixteen healthy subjects performed 2D reaching movements using a rehabilitation robot, with their cortical responses detected using functional near-infrared spectroscopy (fNIRS). Task difficulty was manipulated by varying target size and distance, resulting in 3 levels of index-of-difficulty (ID). Kinematic signals were recorded alongside cortical activity to assess the relationship among movement time, task difficulty, and cortical activation. Results: Our results showed that movement time increased with ID by 0.2974s/bit across all subjects (conditional r2 = 0.6434, p < 0.0001), and all subjects showed individual trends conforming Fitts' Law (all p < 0.001). Neither activation in BA4 nor in BA6 showed a significant correlation with ID (p > 0.05), while both the target size and distance, as well as the interaction between them, showed a significant relationship with BA4 or BA6 activation (all p < 0.05). Conclusion: This study found that although kinematic measures supported Fitts' Law, cortical activity in motor-related areas during reaching movements did not correlate directly with task difficulty as defined by Fitts' Law. Additional factors such as muscle activation may call for different cortical control even when difficulty was identical.
RESUMEN
Musculoskeletal disorders challenge significantly the performance of many daily life activities, thus impacting the quality of life. The efficiency of the traditional physical therapy programs is limited by ecological parameters such as intervention duration and frequency, number of caregivers, geographic accessibility, as well as by subjective factors such as patient's motivation and perseverance in training. The implementation of VR rehabilitation systems may address these limitations, but the technology still needs to be improved and clinically validated. Furthermore, current applications generally lack flexibility and personalization. A VR rehabilitation game simulation is developed, which focuses on the upper-limb movement of reaching, an essential movement involved in numerous daily life activities. Its novelty consists in the integration of a machine learning algorithm, enabling highly adaptive and patient-customized therapeutic intervention. An immersive VR system for the rehabilitation of reaching movement using a bubble popping game is proposed. In the virtual space, the patient is presented with bubbles appearing at different locations and is asked to reach the bubble with the injured limb and pop it. The implementation of a Q-learning algorithm enables the game to adjust the location of the next bubble according to the performance of the patient, represented by his kinematic characteristics. Two test cases simulate the performance of the patient during a training program of 10 days/sessions, in order to validate the effectiveness of the algorithm, demonstrated by the spatial and temporal distribution of the bubbles in each evolving scenario. The results show that the algorithm learns the patient's capabilities and successfully adapts to them, following the reward policy dictated by the therapist; moreover, the algorithm is highly responsive to kinematic features' variation, while demanding a reasonable number of iterations. A novel approach for upper limb rehabilitation is presented, making use of immersive VR and reinforcement learning. The simulation suggests that the algorithm offers adaptive capabilities and high flexibility, needed in the comprehensive personalization of a rehabilitation process. Future work will demonstrate the concept in clinical trials.
Asunto(s)
Movimiento , Humanos , Algoritmos , Masculino , Refuerzo en Psicología , Medicina de Precisión/métodos , Enfermedades Musculoesqueléticas/rehabilitación , Aprendizaje Automático , Juegos de Video , Fenómenos BiomecánicosRESUMEN
Background: In young adults, there is evidence that free arm movements do not help to compensate muscle fatigue-induced deteriorations in dynamic balance performance. However, the postural control system in youth is immature, and as a result, the use of arm movements may provide a compensatory "upper body strategy" to correct fatigue-related balance impairments. Thus, the purpose of the present study was to compare the effects of free vs. restricted arm movement on dynamic balance performance prior and following exercise-induced muscle fatigue. Methods: Forty-three healthy youth (19 females; mean age: 12.8 ± 1.9 years) performed the Y Balance Test-Lower Quarter before and immediately after a fatiguing exercise (i.e., repetitive vertical bipedal box jumps until failure) using two different arm positions: free (move the arms freely) and restricted (keep the arms akimbo) arm movement. Results: Muscle fatigue (p ≤ 0.033; 0.10 ≤ η p 2 ≤ 0.33) and restriction of arm movement (p ≤ 0.005; 0.17 ≤ η p 2 ≤ 0.46) resulted in significantly deteriorated dynamic balance performance. However, the interactions between the two did not reach the level of significance (p ≥ 0.091; 0.01 ≤ η p 2 ≤ 0.07). Conclusion: Our findings indicate that the use of an "upper body strategy" (i.e., free arm position) has no compensatory effect on muscle fatigue-induced dynamic balance deteriorations in healthy youth.
RESUMEN
The acquisition of new motor skills from scratch, also known as de novo learning, is an essential aspect of motor development. In de novo learning, the ability to generalize skills acquired under one condition to others is crucial because of the inherently limited range of motor experiences available for learning. However, the presence of generalization in de novo learning and its influencing factors remain unclear. This study aimed to elucidate the generalization of de novo motor learning by examining the motor exploration process, which is the accumulation of motor experiences. To this end, we manipulated the exploration process during practice by changing the target shape using either a small circular target or a bar-shaped target. Our findings demonstrated that the amount of learning during practice was generalized across different conditions. Furthermore, the extent of generalization is influenced by movement variability in the control space, which is irrelevant to the task, rather than the target shapes themselves. These results confirmed the occurrence of generalization in de novo learning and suggest that the exploration process within the control space plays a significant role in facilitating this generalization.
RESUMEN
Consolidation has been associated with performance gains without additional practice (i.e., off-line learning). However, the movement characteristics improving off-line remain poorly understood. To investigate this question, participants were trained to produce a sequence of planar reaching movements toward four different visual targets. The training session with feedback required them to learn the relative time of the movements, the total movement time and aim accurately at each target. The retention test was performed either 10-min or 24-h after. Results revealed that a 24-h consolidation interval did not result in better temporal or spatial accuracy. This finding suggests that off-line learning may be restricted to sequence production tasks in which the different segments must be regrouped ("chunked") together to accelerate their execution.
Asunto(s)
Aprendizaje , Memoria , Humanos , Movimiento , Desempeño PsicomotorRESUMEN
BACKGROUND: Upper limb motor paresis is a major symptom of stroke, which limits activities of daily living and compromises the quality of life. Kinematic analysis offers an in-depth and objective means to evaluate poststroke upper limb paresis, with anticipation for its effective application in clinical settings. OBJECTIVE: This study aims to compare the movement strategies of patients with hemiparesis due to stroke and healthy individuals in forward reach and hand-to-mouth reach, using a simple methodology designed to quantify the contribution of various movement components to the reaching action. METHODS: A 3D motion analysis was conducted, using a simplified marker set (placed at the mandible, the seventh cervical vertebra, acromion, lateral epicondyle of the humerus, metacarpophalangeal [MP] joint of the index finger, and greater trochanter of the femur). For the forward reach task, we measured the distance the index finger's MP joint traveled from its starting position to the forward target location on the anterior-posterior axis. For the hand-to-mouth reach task, the shortening of the vertical distance between the index finger MP joint and the position of the chin at the start of the measurement was measured. For both measurements, the contributions of relevant upper limb and trunk movements were calculated. RESULTS: A total of 20 healthy individuals and 10 patients with stroke participated in this study. In the forward reach task, the contribution of shoulder or elbow flexion was significantly smaller in participants with stroke than in healthy participants (mean 52.5%, SD 24.5% vs mean 85.2%, SD 4.5%; P<.001), whereas the contribution of trunk flexion was significantly larger in stroke participants than in healthy participants (mean 34.0%, SD 28.5% vs mean 3.0%, SD 2.8%; P<.001). In the hand-to-mouth reach task, the contribution of shoulder or elbow flexion was significantly smaller in participants with stroke than in healthy participants (mean 71.8%, SD 23.7% vs mean 90.7%, SD 11.8%; P=.009), whereas shoulder girdle elevation and shoulder abduction were significantly larger in participants with stroke than in healthy participants (mean 10.5%, SD 5.7% vs mean 6.5%, SD 3.0%; P=.02 and mean 16.5%, SD 18.7% vs mean 3.0%, SD 10.4%; P=.02, respectively). CONCLUSIONS: Compared with healthy participants, participants with stroke achieved a significantly greater distance via trunk flexion in the forward reach task and shoulder abduction and shoulder girdle elevation in the hand-to-mouth reach task, both of these differences are regarded as compensatory movements. Understanding the characteristics of individual motor strategies, such as dependence on compensatory movements, may contribute to tailored goal setting in stroke rehabilitation.
RESUMEN
Complete reaching movements involve target sensing, motor planning, and arm movement execution, and this process requires the integration and communication of various brain regions. Previously, reaching movements have been decoded successfully from the motor cortex (M1) and applied to prosthetic control. However, most studies attempted to decode neural activities from a single brain region, resulting in reduced decoding accuracy during visually guided reaching motions. To enhance the decoding accuracy of visually guided forelimb reaching movements, we propose a parallel computing neural network using both M1 and medial agranular cortex (AGm) neural activities of rats to predict forelimb-reaching movements. The proposed network decodes M1 neural activities into the primary components of the forelimb movement and decodes AGm neural activities into internal feedforward information to calibrate the forelimb movement in a goal-reaching movement. We demonstrate that using AGm neural activity to calibrate M1 predicted forelimb movement can improve decoding performance significantly compared to neural decoders without calibration. We also show that the M1 and AGm neural activities contribute to controlling forelimb movement during goal-reaching movements, and we report an increase in the power of the local field potential (LFP) in beta and gamma bands over AGm in response to a change in the target distance, which may involve sensorimotor transformation and communication between the visual cortex and AGm when preparing for an upcoming reaching movement. The proposed parallel computing neural network with the internal feedback model improves prediction accuracy for goal-reaching movements.
Asunto(s)
Objetivos , Extremidad Superior , Animales , Retroalimentación , Miembro Anterior/fisiología , Movimiento/fisiologíaRESUMEN
Performing exercise before or after motor skill learning is thought to have a positive impact on acquisition and retention of motor memories stored in our nervous system. It has been shown that performing 25 min of moderate-intensity aerobic exercise prior to visuomotor adaptation can enhance both visuomotor adaptation and its retention compared to 25 min of rest before the adaptation. To determine whether a single bout of aerobic exercise could actually facilitate the formation of a neural representation associated with a novel visuomotor condition, we examined aftereffects and savings associated with a visuomotor adaptation task following either an exercise or a rest condition. Sixteen healthy young individuals (18-35 years) first experienced 25 min of moderate-intensity cycling or rest, and then adapted to a 30-degree visuomotor rotation condition. Immediately following that, participants experienced a washout session, which was followed by a readaptation session. Results indicated that all subjects adapted to the visuomotor rotation completely, although no difference was found between the cycling and rest conditions. Aftereffects and savings were also observed in both conditions, but with no difference between the conditions. These findings suggest that compared to a short rest session, a single bout of moderate-intensity cycling may not have a greater impact for enhancing visuomotor adaptation and its retention. Further research is needed, in which the effects of certain factors such as exercise intensity, duration and timing are more systematically investigated.
Asunto(s)
Ejercicio Físico , Aprendizaje , Humanos , Ejercicio Físico/fisiología , Aprendizaje/fisiología , Destreza Motora/fisiología , Adaptación Fisiológica/fisiología , Descanso , Desempeño Psicomotor/fisiologíaRESUMEN
Subthalamic nucleus (STN) beta-triggered adaptive deep brain stimulation (ADBS) has been shown to provide clinical improvement comparable to conventional continuous DBS (CDBS) with less energy delivered to the brain and less stimulation induced side effects. However, several questions remain unanswered. First, there is a normal physiological reduction of STN beta band power just prior to and during voluntary movement. ADBS systems will therefore reduce or cease stimulation during movement in people with Parkinson's disease and could therefore compromise motor performance compared to CDBS. Second, beta power was smoothed and estimated over a time period of 400 ms in most previous ADBS studies, but a shorter smoothing period could have the advantage of being more sensitive to changes in beta power, which could enhance motor performance. In this study, we addressed these two questions by evaluating the effectiveness of STN beta-triggered ADBS using a standard 400 ms and a shorter 200 ms smoothing window during reaching movements. Results from 13 people with Parkinson's disease showed that reducing the smoothing window for quantifying beta did lead to shortened beta burst durations by increasing the number of beta bursts shorter than 200 ms and more frequent switching on/off of the stimulator but had no behavioural effects. Both ADBS and CDBS improved motor performance to an equivalent extent compared to no DBS. Secondary analysis revealed that there were independent effects of a decrease in beta power and an increase in gamma power in predicting faster movement speed, while a decrease in beta event related desynchronization (ERD) predicted quicker movement initiation. CDBS suppressed both beta and gamma more than ADBS, whereas beta ERD was reduced to a similar level during CDBS and ADBS compared with no DBS, which together explained the achieved similar performance improvement in reaching movements during CDBS and ADBS. In addition, ADBS significantly improved tremor compared with no DBS but was not as effective as CDBS. These results suggest that STN beta-triggered ADBS is effective in improving motor performance during reaching movements in people with Parkinson's disease, and that shortening of the smoothing window does not result in any additional behavioural benefit. When developing ADBS systems for Parkinson's disease, it might not be necessary to track very fast beta dynamics; combining beta, gamma, and information from motor decoding might be more beneficial with additional biomarkers needed for optimal treatment of tremor.
Asunto(s)
Estimulación Encefálica Profunda , Enfermedad de Parkinson , Núcleo Subtalámico , Humanos , Enfermedad de Parkinson/terapia , Estimulación Encefálica Profunda/métodos , Temblor/terapia , Movimiento/fisiología , Núcleo Subtalámico/fisiologíaRESUMEN
BACKGROUND: Upper-limb rehabilitation robots provide repetitive reaching movement training to post-stroke patients. Beyond a pre-determined set of movements, a robot-aided training protocol requires optimization to account for the individuals' unique motor characteristics. Therefore, an objective evaluation method should consider the pre-stroke motor performance of the affected arm to compare one's performance relative to normalcy. However, no study has attempted to evaluate performance based on an individual's normal performance. Herein, we present a novel method for evaluating upper limb motor performance after a stroke based on a normal reaching movement model. METHODS: To represent the normal reaching performance of individuals, we opted for three candidate models: (1) Fitts' law for the speed-accuracy relationship, (2) the Almanji model for the mouse-pointing task of cerebral palsy, and (3) our proposed model. We first obtained the kinematic data of healthy (n = 12) and post-stroke (n = 7) subjects with a robot to validate the model and evaluation method and conducted a pilot study with a group of post-stroke patients (n = 12) in a clinical setting. Using the models obtained from the reaching performance of the less-affected arm, we predicted the patients' normal reaching performance to set the standard for evaluating the affected arm. RESULTS: We verified that the proposed normal reaching model identifies the reaching of all healthy (n = 12) and less-affected arm (n = 19; 16 of them showed an R2 > 0.7) but did not identify erroneous reaching of the affected arm. Furthermore, our evaluation method intuitively and visually demonstrated the unique motor characteristics of the affected arms. CONCLUSIONS: The proposed method can be used to evaluate an individual's reaching characteristics based on an individuals normal reaching model. It has the potential to provide individualized training by prioritizing a set of reaching movements.
Asunto(s)
Parálisis Cerebral , Accidente Cerebrovascular , Animales , Ratones , Proyectos Piloto , Extremidad Superior , MovimientoRESUMEN
Based on the principles of neuromechanics, human arm movements result from the dynamic interaction between the nervous, muscular, and skeletal systems. To develop an effective neural feedback controller for neuro-rehabilitation training, it is important to consider both the effects of muscles and skeletons. In this study, we designed a neuromechanics-based neural feedback controller for arm reaching movements. To achieve this, we first constructed a musculoskeletal arm model based on the actual biomechanical structure of the human arm. Subsequently, a hybrid neural feedback controller was developed that mimics the multifunctional areas of the human arm. The performance of this controller was then validated through numerical simulation experiments. The simulation results demonstrated a bell-shaped movement trajectory, consistent with the natural motion of human arm movements. Furthermore, the experiment testing the tracking ability of the controller revealed real-time errors within one millimeter, with the tensile force generated by the controller's muscles being stable and maintained at a low value, thereby avoiding the issue of muscle strain that can occur due to excessive excitation during the neurorehabilitation process.
RESUMEN
Studies confirm that children with cerebral palsy (CwCP) have difficulty with simple, everyday movements like reaching for objects. Accurate reaching requires that shoulder and elbow joints are coordinated to move the hand along a smooth path to the desired target location. Here we examined multijoint coordination by comparing reaching performance in the affected and unaffected limbs of CwCP (nine children, six girls and three boys, aged 8-10 years) to reaching performance in the non-dominant and dominant limbs of typically-developing age- and gender-matched control (CTR) children. The hypothesis was that CwCP would show the effects of coordination deficits in both their affected and unaffected limbs. All children performed two sessions (one session with each arm) of speeded reaching movements to three targets arranged to manipulate the required pattern of shoulder and elbow coordination. The movements were tracked with a motion tracker allowing us to assess the following measures: movement distance, duration, and speed, hand-path deviation from linearity, final position accuracy and precision, and measures of shoulder and elbow excursion. We found that CwCP made reaches that covered a greater distance and took more time, that their shoulder and elbow rotations were larger, and that their movements showed greater deviation from linearity than the movements performed by CTR children. Children with CP were also more variable than CTR children on every measure except movement duration. The pattern of shoulder and elbow rotation observed in the CwCP group represents a coordination pattern that is significantly different from the pattern used by CTR children and may represent a greater reliance by CwCP on proximal muscular control systems. The discussion section considers the role that the cortical-spinal system may play in multijoint coordination.
RESUMEN
The capacity to stop impending or ongoing actions contributes to executive control over behavior. Action-stopping, however, is difficult to directly quantify. It is therefore assayed via computational modeling of behavior in the stop signal task to estimate the latency of stopping (stop signal reaction time, SSRT) and, more recently, the reliability of stopping in terms of the distribution of SSRTs (standard deviation, SD-SSRT) and the frequency with which one outright fails to react to a stop signal (trigger failures, TF). Critically, the validity of computational estimates remains unknown because we currently have no direct readouts of behavior against which to compare them. Here, we developed a method for providing single-trial behavioral readouts of SSRT and trigger failures. The method relies on an adaptation of the stop signal task in which participants respond by moving a computer mouse. In two online experiments, we used movement kinematics to quantify stopping performance (SSRT, SD-SSRT, and TF), and then applied the standard Race Model and recent BEESTS model in order to examine the convergent validity of the methods. Overall, we demonstrate good correspondence between kinematics- and model-based estimates of stopping performance at the group and individual level. We conclude that the new method provides valid estimates of stopping performance that, unlike model-based estimates, can be read out at the level of single trials. Our approach might therefore be useful for interrogating single-trial neurophysiological correlates of stopping and for large-scale, online studies of behavioral stopping.
Asunto(s)
Función Ejecutiva , Inhibición Psicológica , Humanos , Reproducibilidad de los Resultados , Tiempo de Reacción/fisiología , Función Ejecutiva/fisiología , Movimiento , Desempeño Psicomotor/fisiologíaRESUMEN
Objective.Due to the variability of human movements, muscle activations vary among trials and subjects. However, few studies investigated how data organization methods for addressing variability impact the extracted muscle synergies.Approach.Fifteen healthy subjects performed a large set of upper limb multi-directional point-to-point reaching movements. Then, the study extracted muscle synergies under different data settings and investigated how data structure prior to synergy extraction, namely concatenation, averaging, and single trial, the number of considered trials, and the number of reaching directions affected the number and components of muscle synergies.Main results.The results showed that the number and components of synergies were significantly affected by the data structure. The concatenation method identified the highest number of synergies, and the averaging method usually found a smaller number of synergies. When the concatenated trials or reaching directions was lower than a minimum value, the number of synergies increased with the increase of the number of trials or reaching directions; however, when the number of trials or reaching directions reached a threshold, the number of synergies was usually constant or with less variation even when novel directions and trials were added. Similarity analysis also showed a slight increase when the number of trials or reaching directions was lower than a threshold. This study recommends that at least five trials and four reaching directions and the concatenation method are considered in muscle synergies analysis during upper limb tasks.Significance.This study makes the researchers focus on the variability analysis induced by the diseases rather than the techniques applied for synergies analysis and promotes applications of muscle synergies in clinical scenarios.
Asunto(s)
Movimiento , Músculo Esquelético , Humanos , Electromiografía , Fenómenos Biomecánicos , Músculo Esquelético/fisiología , Movimiento/fisiología , Extremidad SuperiorRESUMEN
A hallmark of human reaching movements is that they are appropriately tuned to the task goal and to the environmental context. This was demonstrated by the way humans flexibly respond to mechanical and visual perturbations that happen during movement. Furthermore, it was previously showed that the properties of goal-directed control can change within a movement, following abrupt changes in the goal structure. Such online adjustment was characterized by a modulation of feedback gains following switches in target shape. However, it remains unknown whether the underlying mechanism merely switches between prespecified policies, or whether it results from continuous and potentially dynamic adjustments. Here, we address this question by investigating participants' feedback control strategies in presence of various changes in target width during reaching. More specifically, we studied whether the feedback responses to mechanical perturbations were sensitive to the rate of change in target width, which would be inconsistent with the hypothesis of a single, discrete switch. Based on movement kinematics and surface EMG data, we observed a modulation of feedback response clearly dependent on dynamical changes in target width. Together, our results demonstrate a continuous and online transformation of task-related parameters into suitable control policies.
Asunto(s)
Movimiento , Desempeño Psicomotor , Fenómenos Biomecánicos , Electromiografía , Retroalimentación , Humanos , Movimiento/fisiologíaRESUMEN
In human upper-arm reaching movements, the variance of the hand position increases until the middle of the movement and then decreases toward the endpoint. Such decrease in positional variance has been suggested as an evidence to support the hypothesis that our nervous system uses feedback control, rather than feedforward control, for arm reaching tasks. In this study, we computed the optimal trajectories based on feedforward control under several criteria for a one-link two-muscle arm model with considering the stochastic property of muscle activities in order to reexamine the hypothesis. The results showed that the feedforward control also represents the decrease in positional variance in the latter half of the movement when the control signal is planned under the minimum energy cost and minimum variance models. Furthermore, the optimal trajectory that minimizes energy cost represents not only the decrease in positional variance but also many other characteristics of the human reaching movements, e.g., the three-phasic activity of antagonistic muscle, bell-shaped speed curve, N-shaped equilibrium trajectory, and bimodal profile of joint stiffness. These results suggest that minimum energy cost model well expresses the characteristics of hand reaching movements, and our central nervous system would make use of not only a feedback control but also feedforward control.
Asunto(s)
Modelos Biológicos , Movimiento , Retroalimentación , Mano , Humanos , Movimiento/fisiologíaRESUMEN
Online motor control is often required to correct errors in rapid adjustments during reaching movements. It has been established that the initial arm trajectory during reaching is corrected by a target displacement. Since this corrective response occurs without perception of target perturbation, this is regarded as an automatic response. However, an object rarely "jumps" in daily life, rather it often "moves" as a chronological change of the position that causes visual motion. Therefore, the purpose of this study was to investigate whether the implicit visuomotor response is induced by target motion stimuli and to clarify the effects of target motion velocity on initial arm trajectory. Participants were asked to move a cursor from a start circle to a visual target. The target moved either leftward or rightward when the cursor passed 20 mm from the start circle. Four target velocities (10, 20, 30, 40 deg/s) were randomly presented. Our results showed that the initial velocity (first 50 ms) of the fast corrective response increased with the target velocity. Therefore, it is indicated that the fast corrective response is induced by the target motion stimulus with a short latency and its amplitude is dependent on the target velocity.
Asunto(s)
Movimiento , Desempeño Psicomotor , Humanos , Movimiento/fisiología , Desempeño Psicomotor/fisiologíaRESUMEN
Modulation of gamma oscillations recorded from the human motor cortex and basal ganglia appears to play a key role in movement execution. However, there are still major questions to be answered about the specific role of cortical gamma activity in both the planning and execution of movement features such as the scaling of peak velocity and movement time. In this study, we characterized movement-related gamma oscillatory dynamics and its relationship with kinematic parameters based on 256-channels EEG recordings in 64 healthy subjects while performing fast and uncorrected reaching movements to targets located at three distances. In keeping with previous studies, we found that movement-related gamma synchronization occurred during movement execution. As a new finding, we showed that gamma synchronization occurred also before movement onset, with planning and execution phases involving different gamma peak frequencies and topographies. Importantly, the amplitude of gamma synchronization in both planning and execution increased with target distance and predicted peak velocity and movement time. Additional analysis of phase coherence revealed a gamma-coordinated long-range network involving occipital, frontal and central regions during movement execution that was positively related to kinematic features. This is the first evidence in humans supporting the notion that gamma synchronization amplitude and phase coherence pattern can reliably predict peak velocity amplitude and movement time. Therefore, these findings suggest that cortical gamma oscillations have a crucial role for the selection, implementation and control of the appropriate kinematic parameters of goal-directed reaching movements.
RESUMEN
Detecting human motion and predicting human intentions by analyzing body signals are challenging but fundamental steps for the implementation of applications presenting human-robot interaction in different contexts, such as robotic rehabilitation in clinical environments, or collaborative robots in industrial fields. Machine learning techniques (MLT) can face the limit of small data amounts, typical of this kind of applications. This paper studies the illustrative case of the reaching movement in 10 healthy subjects and 21 post-stroke patients, comparing the performance of linear discriminant analysis (LDA) and random forest (RF) in: (i) predicting the subject's intention of moving towards a specific direction among a set of possible choices, (ii) detecting if the subject is moving according to a healthy or pathological pattern, and in the case of discriminating the damage location (left or right hemisphere). Data were captured with wearable electromagnetic sensors, and a sub-section of the acquired signals was required for the analyses. The possibility of detecting with which arm (left or right hand) the motion was performed, and the sensitivity of the MLT to variations in the length of the signal sub-section were also evaluated. LDA and RF prediction accuracies were compared: Accuracy improves when only healthy subjects or longer signals portions are considered up to 11% and at least 10%, respectively. RF reveals better estimation performance both as intention predictor (on average 59.91% versus the 62.19% of LDA), and health condition detector (over 90% in all the tests).