IEEE Transactions on Neural Systems and Rehabilitation Engineering最新文献

{"title":"Ankle Exoskeletons May Hinder Standing Balance in Simple Models of Older and Younger Adults","authors":"Daphna Raz;Varun Joshi;Brian R. Umberger;Necmiye Ozay","doi":"10.1109/TNSRE.2025.3549641","DOIUrl":"10.1109/TNSRE.2025.3549641","url":null,"abstract":"Humans rely on ankle torque to maintain standing balance, particularly in the presence of small to moderate perturbations. Reductions in maximum torque (MT) production and maximum rate of torque development (MRTD) occur at the ankle with age, diminishing stability. Ankle exoskeletons are powered orthotic devices that may assist older adults by compensating for reduced torque and power production capabilities. They may also be able to assist with ankle strategies used for balance. However, their effect on standing balance in older adults is not well understood. Here, we model the effects ankle exoskeletons have on stability in physics-based models of healthy young and old adults, focusing on the potential to mitigate age-related deficits in MT and MRTD. Using backward reachability, a mathematical technique for analyzing the behavior of dynamical systems, we compute the set of stable center of mass positions and velocities for sex and age adjusted nonlinear models of human standing balance with an ankle exoskeleton. We show that an ankle exoskeleton moderately reduces feasible stability boundaries in users who have full ankle strength. For individuals with age-related deficits, there is a trade-off. While exoskeletons augment stability at low center of mass velocities, they reduce stability in some high velocity conditions. Our results suggest that ankle exoskeletons using established control strategies might have unforeseen negative effects on stability, especially for individuals who are most likely to benefit from them.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"1145-1155"},"PeriodicalIF":4.8,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10918710","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143596988","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Weighted Multi-Modal Contrastive Learning Based Hybrid Network for Alzheimer’s Disease Diagnosis","authors":"Renping Yu;Chao Peng;Jixiang Zhu;Mingming Chen;Rui Zhang","doi":"10.1109/TNSRE.2025.3549730","DOIUrl":"10.1109/TNSRE.2025.3549730","url":null,"abstract":"Multiple imaging modalities and specific proteins in the cerebrospinal fluid, providing a comprehensive understanding of neurodegenerative disorders, have been widely used for computer-aided diagnosis of Alzheimer’s disease (AD). Given the proven effectiveness of contrastive learning in aligning multi-modal representation, in this paper, we investigate effective contrastive learning strategies to learn better cross-modal representations for the integration of multi-modal complementary information. To enhance the overall performance in AD diagnosis, we construct a unified hybrid network that integrates feature learning and classifier learning into an end-to-end framework. Specifically, we propose a weighted multi-modal contrastive learning based on hybrid network (WMCL-HN) method. Firstly, an adaptive weighted strategy is implemented on the multi-modal contrastive learning to dynamically regulate the degree of information exchange across modalities. It assigns higher weights to more important modality pair, thus the most important underlying relationships across modalities can be captured. Secondly, we construct a hybrid network, which employs a curriculum learning strategy that gradually transitions the training from feature learning to classifier learning, ensuring that the learned features are tailored to the diagnostic task. Experimental results on ADNI dataset demonstrate the effectiveness of the proposed WMCL-HN in AD-related diagnosis tasks. The source code is available at <uri>https://github.com/pcehnago/WMCL-HN</uri>.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"1135-1144"},"PeriodicalIF":4.8,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10919214","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143596992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Human-in-the-Loop Myoelectric Pattern Recognition Control of an Arm-Support Robot to Improve Reaching in Stroke Survivors","authors":"Joseph V. Kopke;Michael D. Ellis;Levi J. Hargrove","doi":"10.1109/TNSRE.2025.3549376","DOIUrl":"10.1109/TNSRE.2025.3549376","url":null,"abstract":"The objective of this study was to assess the feasibility and efficacy of using real-time human-in-the-loop pattern recognition-based myoelectric control to control vertical support force or vertical position to improve reach in individuals with chronic stroke. This work attempts to move proven lab-based static arm support paradigms towards a controllable wearable device. A machine learning (linear discriminant analysis)-based myoelectric pattern recognition system based on movement intent as determined by real-time muscle activation was used to control incremental changes in either vertical position or vertical support force during a reach and retrieve task, with the goal of improving reaching function. Performance under real-time control of both options was compared to two unchanging static-support conditions (current gold standard) and a no-support condition. Both real-time control paradigms were successfully implemented and resulted in greater forward-reaching performance as demonstrated by increased elbow extension and horizontal shoulder adduction compared to no-support and was not different from the current gold standard static support paradigms. Muscle activation levels with real-time support were lower than the no-support condition and similar to those observed during the static support paradigms. Real-time detection of user intent was successful in controlling both vertical position and vertical support force and enabled greater reaching distance than without it demonstrating both its feasibility and efficacy albeit with some limitations.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"1108-1117"},"PeriodicalIF":4.8,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10916732","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143575566","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Functional Reorganization of White Matter Supporting the Transhemispheric Mechanism of Mirror Therapy After Stroke: A Multimodal MRI Study","authors":"Li Ding;Kexu Zhang;Xu Wang;Shanbao Tong;Xiaoli Guo;Jie Jia","doi":"10.1109/TNSRE.2025.3549380","DOIUrl":"10.1109/TNSRE.2025.3549380","url":null,"abstract":"Mirror therapy (MT) is an effective approach in stroke recovery, but its impact on subcortical neural reorganization remains unclear. Thus, we aimed to investigate the neuroplastic effects on white matter due to MT. In this study, thirty-three participants with stroke were recruited and randomly assigned into the MT group (n = 16) or the control group (n = 17) for a 4-week intervention. Before and after the intervention, motor recovery was evaluated using the Fugl-Meyer Assessment upper limb subscale (FMA-UL), and the white matter structure and function were investigated using DTI and resting-state fMRI, focusing on the corticospinal tract and the corpus callosum. Significant correlations between the improvements of the FMA-UL and the baseline fractional anisotropy of ipsilesional corticospinal tract (<inline-formula> <tex-math>${p} lt 0.001$ </tex-math></inline-formula>) and corpus callosum (<inline-formula> <tex-math>${p} = 0.009$ </tex-math></inline-formula>) were observed only in the MT group. Additionally, no significant structural alterations were found between the two groups after the intervention. The fractional amplitude of low-frequency fluctuation of ipsilesional corticospinal tract (<inline-formula> <tex-math>${p} = 0.003$ </tex-math></inline-formula>) and corpus callosum (<inline-formula> <tex-math>${p} = 0.005$ </tex-math></inline-formula>) were significantly enhanced only in the MT group, which were correlated with the improvements of the FMA-UL (<inline-formula> <tex-math>${p} lt 0.001$ </tex-math></inline-formula>). Furthermore, partial correlation analysis and subsequent mediation model analysis suggested that the changes of fractional amplitude of low-frequency fluctuation in corpus callosum partially mediated the effect of the baseline fractional anisotropy of ipsilesional corticospinal tract on the FMA-UL improvements in the MT group. This study provided neuroimaging evidence on white matter reorganization after MT, specifically the corpus callosum, suggesting a potential interhemispheric transcallosal neuroplastic mechanism of MT.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"1126-1134"},"PeriodicalIF":4.8,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10916696","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143575565","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing Manual Wheelchair Propulsion: Incremental Assistance Levels of Pushrim-Activated Power-Assist Proportionally Reduce Physiological and Biomechanical Demands in Non-Disabled Participants","authors":"Jelmer Braaksma;Sonja de Groot;Han Houdijk;Riemer J. K. Vegter","doi":"10.1109/TNSRE.2025.3547052","DOIUrl":"10.1109/TNSRE.2025.3547052","url":null,"abstract":"This study assessed the effect of increasing assistance levels of a Pushrim-Activated Power-assisted Wheelchair (PAPAW) on the physiological and biomechanical demands in non-disabled participants propelling a manual wheelchair on an instrumented ergometer. This cross-sectional study included twenty-four non-disabled participants (aged <inline-formula> <tex-math>$21.1~pm ~1.4$ </tex-math></inline-formula> years) who performed 4 submaximal trials of 4-minutes propulsion (at 1.11m/s and 0.21W/kg body mass resistance) using no, low, medium and high power-assist modes of a PAPAW in counterbalanced order. Physiological strain, in terms of metabolic energy expenditure, heart rate and perceived exertion, was examined, along with the force and velocity data from the wheelchair ergometer and PAPAW. Repeated measures ANOVA revealed that metabolic energy expenditure decreased significantly with each incremental step of assistance (no: <inline-formula> <tex-math>$299~pm ~43$ </tex-math></inline-formula>W, low: <inline-formula> <tex-math>$250~pm ~37$ </tex-math></inline-formula>W, medium: <inline-formula> <tex-math>$240~pm ~44$ </tex-math></inline-formula>W, high: <inline-formula> <tex-math>$224~pm ~38$ </tex-math></inline-formula>W, p <inline-formula> <tex-math>$le 0.001$ </tex-math></inline-formula>), accompanied by similar reductions in heart rate and perceived exertion. Similarly, work per push decreased with each step (no: <inline-formula> <tex-math>$16.9~pm ~6.2$ </tex-math></inline-formula>J, low: <inline-formula> <tex-math>$7.9~pm ~2.8$ </tex-math></inline-formula>J, medium: <inline-formula> <tex-math>$6.4~pm ~2.5$ </tex-math></inline-formula>J, high: <inline-formula> <tex-math>$5.5~pm ~1.4$ </tex-math></inline-formula>J, p <inline-formula> <tex-math>$le 0.001$ </tex-math></inline-formula>). This can be explained by reductions in propulsive forces, which reached a floor effect with no further reduction between medium and high assistance levels, and a decreased contact angle. The level of PAPAW assistance progressively reduces the metabolic and biomechanical demands of manual wheelchair propulsion, potentially lowering the risk of overuse injuries and enhancing participation in daily activities and society. Further research should explore the optimal assistance level that reduces strain while maintaining physical fitness during everyday mobility.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"1071-1078"},"PeriodicalIF":4.8,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10915609","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143572820","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving Acceptance to Sensory Substitution: A Study on the V2A-SS Learning Model Based on Information Processing Learning Theory","authors":"Kyeong Deok Moon;Yun Kyung Park;Moo Seop Kim;Chi Yoon Jeong","doi":"10.1109/TNSRE.2025.3548942","DOIUrl":"10.1109/TNSRE.2025.3548942","url":null,"abstract":"The visual sensory organ (VSO) serves as the primary channel for transmitting external information to the brain; therefore, damage to the VSO can severely limit daily activities. Visual-to-Auditory Sensory Substitution (V2A-SS), an innovative approach to restoring vision, offers a promising solution by leveraging neuroplasticity to convey visual information via auditory channels. Advances in information technology and artificial intelligence mitigate technical challenges such as low resolution and limited bandwidth, thereby enabling broader applicability of V2A-SS. Despite these advances, integrating V2A-SS effectively into everyday life necessitates extensive training and adaptation. Therefore, alongside addressing technical challenges, investigating effective learning strategies to accelerate the acceptance of V2A-SS is crucial. This study introduces a V2A-SS learning model based on the Information Processing Learning Theory (IPLT), encompassing the stages of “concept acquisition, rehearsal, assessment” to reduce the learning curve and enhance adaptation. The experimental results show that the proposed learning model improves recognition rates, achieving an 11% increase over simple random repetition learning. This improvement is significantly higher than the gain of 2.72% achieved by optimizing the V2A-SS algorithm with Mel-Scaled Frequency Mapping. This study suggests that a structured learning model for sensory substitution technologies can contribute to bridging gaps between technical feasibility and practical application. This underscores the need to develop effective learning models, alongside technological advancements, to accelerate the adoption of V2A-SS and neuroplasticity.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"1097-1107"},"PeriodicalIF":4.8,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10915707","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143572824","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The More, the Better? Evaluating the Role of EEG Preprocessing for Deep Learning Applications","authors":"Federico Del Pup;Andrea Zanola;Louis Fabrice Tshimanga;Alessandra Bertoldo;Manfredo Atzori","doi":"10.1109/TNSRE.2025.3547616","DOIUrl":"10.1109/TNSRE.2025.3547616","url":null,"abstract":"The last decade has witnessed a notable surge in deep learning applications for electroencephalography (EEG) data analysis, showing promising improvements over conventional statistical techniques. However, deep learning models can underperform if trained with bad processed data. Preprocessing is crucial for EEG data analysis, yet there is no consensus on the optimal strategies in deep learning scenarios, leading to uncertainty about the extent of preprocessing required for optimal results. This study is the first to thoroughly investigate the effects of EEG preprocessing in deep learning applications, drafting guidelines for future research. It evaluates the effects of varying preprocessing levels, from raw and minimally filtered data to complex pipelines with automated artifact removal algorithms. Six classification tasks (eye blinking, motor imagery, Parkinson’s, Alzheimer’s disease, sleep deprivation, and first episode psychosis) and four established EEG architectures were considered for the evaluation. The analysis of 4800 trained models revealed statistical differences between preprocessing pipelines at the intra-task level for each model and at the inter-task level for the largest model. Models trained on raw data consistently performed poorly, always ranking last in average scores. In addition, models seem to benefit more from minimal pipelines without artifact handling methods. These findings suggest that EEG artifacts may affect the performance and generalizability of deep neural networks.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"1061-1070"},"PeriodicalIF":4.8,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10909332","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143541847","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Locomotion Joint Angle and Moment Estimation With Soft Wearable Sensors for Personalized Exosuit Control","authors":"Luying Feng;Lianghong Gui;Wenzhu Xu;Xiang Wang;Canjun Yang;Yaochu Jin;Wei Yang","doi":"10.1109/TNSRE.2025.3547361","DOIUrl":"10.1109/TNSRE.2025.3547361","url":null,"abstract":"Recent advancements in flexible sensing and machine learning have positioned soft sensors as promising alternatives to traditional methods for human posture detection. However, most research has centered on calibration, with limited progress in practical applications due to the challenges posed by diverse users and complex scenarios such as human-robot interaction. To address these challenges, this study developed a flexible sensing system capable of accurately predicting joint angles and moments, and validated it through a flexible exosuit. To improve the model’s accuracy and generalization, gait data from eight participants with varying walking patterns were collected. Calibrated data were used as static features and trained alongside dynamic features. The model was pre-trained on a large open-source dataset and then fine-tuned for our own data. Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) models were specifically applied to estimate knee joint angles and hip joint moments, achieving a Mean Absolute Error (MAE) of 4.43° and 0.12 Nm/kg, respectively. A flexible exosuit was then developed to provide assistance based on real-time estimation of hip joint moments, enabling personalized control. Testing with five volunteers showed reduced muscle activation, while user satisfaction surveys indicated significant improvements in mobility and comfort. This research not only enhances the practical application of soft sensors but also demonstrates their potential in advancing human-robot interaction.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"1048-1060"},"PeriodicalIF":4.8,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10909362","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143541937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"LAST-PAIN: Learning Adaptive Spike Thresholds for Low Back Pain Biosignals Classification","authors":"Freek Hens;Mohammad Mahdi Dehshibi;Leila Bagheriye;Ana Tajadura-Jiménez;Mahyar Shahsavari","doi":"10.1109/TNSRE.2025.3546682","DOIUrl":"10.1109/TNSRE.2025.3546682","url":null,"abstract":"Spiking neural networks (SNNs) present the potential for ultra-low-power computation, especially when implemented on dedicated neuromorphic hardware. However, a significant challenge is the efficient conversion of continuous real-world data into the discrete spike trains required by SNNs. In this paper, we introduce Learning Adaptive Spike Thresholds (LAST), a novel, trainable encoding strategy designed to address this challenge. The LAST encoder learns adaptive thresholds to transform continuous signals of varying dimensionality-ranging from time series data to high dimensional tensors-into sparse spike trains. Our proposed encoder effectively preserves temporal dynamics and adapts to the characteristics of the input. We validate the LAST approach in a demanding healthcare application using the EmoPain dataset. This dataset contains multimodal biosignal analysis for assessing chronic lower back pain (CLBP). Despite the dataset’s small sample size and class imbalance, our LAST-driven SNN framework achieves a competitive Matthews Correlation Coefficient of 0.44 and an accuracy of 80.43% in CLBP classification. The experimental results also indicate that the same framework can achieve an F1-score of 0.65 in detecting protective behaviour. Furthermore, the LAST encoder outperforms conventional rate and latency-based encodings while maintaining sparse spike representations. This achievement shows promises for energy-efficient and real-time biosignal processing in resource-limited environments.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"1038-1047"},"PeriodicalIF":4.8,"publicationDate":"2025-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10908225","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143541934","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated Autism Assessment With Multimodal Data and Ensemble Learning: A Scalable and Consistent Robot-Enhanced Therapy Framework","authors":"Iqbal Hassan;Nazmun Nahid;Minhajul Islam;Shahera Hossain;Björn Schuller;Md Atiqur Rahman Ahad","doi":"10.1109/TNSRE.2025.3546519","DOIUrl":"10.1109/TNSRE.2025.3546519","url":null,"abstract":"Navigating the complexities of Autism Spectrum Disorder (ASD) diagnosis and intervention requires a nuanced approach that addresses both the inherent variability in therapeutic practices and the imperative for scalable solutions. This paper presents a transformative Robot-Enhanced Therapy (RET) framework, leveraging an intricate amalgamation of an Adaptive Boosted 3D biomarker approach and Saliency Maps generated through Kernel Density Estimation. By seamlessly integrating these methodologies through majority voting, the framework pioneers a new frontier in automating the assessment of ASD levels and Autism Diagnostic Observation Schedule (ADOS) scores, offering unprecedented precision and efficiency. Drawing upon the rich tapestry of the DREAM Dataset, encompassing data from 61 children, this study meticulously crafts novel features derived from diverse modalities including body skeleton, head movement, and eye gaze data. Our 3D bio-marker approach achieves a remarkable predictive prowess, boasting a staggering 95.59% accuracy and an F1 score of 92.75% for ASD level prediction, alongside an RMSE of 1.78 and an R-squared value of 0.74 for ADOS score prediction. Furthermore, the introduction of a pioneering saliency map generation method, harnessing gaze data, further enhances predictive models, elevating ASD level prediction accuracy to an impressive 97.36%, with a corresponding F1 score of 95.56%. Beyond technical achievements, this study underscores RET’s transformative potential in reshaping ASD intervention paradigms, offering a promising alternative to Standard Human Therapy (SHT) by mitigating therapist variability and providing scalable therapeutic approaches. While acknowledging limitations in the research, such as sample constraints and model generalizability, our findings underscore RET’s capacity to revolutionize ASD management.","PeriodicalId":13419,"journal":{"name":"IEEE Transactions on Neural Systems and Rehabilitation Engineering","volume":"33 ","pages":"1191-1201"},"PeriodicalIF":4.8,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10906665","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143541916","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}