Training Tasks Research Articles

Designing a robot-assisted rehabilitation system for hand function recovery using virtual reality and haptic robot

It is well understood that motor dysfunctions in patients can be restored through a certain amount of repetitive training. Robots and virtual reality offer an excellent solution to incentivize and optimize training intensity while also reducing the workload for therapists. In this paper, leveraging virtual reality technologies and haptic robots, we have developed a cost-effective robot-assisted training system and three training exercises for hand rehabilitation. The Unity game engine was utilized to create training scenarios where hand avatars interact with virtual objects. Off-the-shelf haptic robots were employed to manipulate the hand avatars, enabling completion of training tasks with a realistic tactile experience. Haptic feedback is rendered using impedance control laws. Additionally, to address the needs of both patients and therapists we propose to employ robot-assisted force and feedback force. These mechanisms are designed to provide assistance to patients during tasks if needs and to transmit the patient's force feedback to the therapist, facilitating comprehensive interaction and communication between the two parties. Several experiments were conducted involving six healthy subjects, who used electrode stimulation to simulate impaired limbs, to demonstrate the effectiveness of the proposed training frameworks.

BERT-based log exception detection algorithm LADB

With the development of information systems, they have become increasingly large and complex and have generated a large amount of log information. These log information records the system's health, but the number of log information is huge, and the traditional exception detection algorithm in the case of very large amounts of data is difficult to efficiently and accurately detect anomalies due to poor generalization performance. A BERT-based log anomaly detection algorithm, LADB, is proposed, essentially a semi-supervised, multi-classification algorithm. (1) LADB uses the Transformer encoder as the base component for problems such as feature degradation and gradient explosion. (2) In order to make better use of the bidirectional context, and in view of BERT's excellence in the NLP field, the Masking Log Key Prediction (MLKP) self-monitoring task was designed, drawing on the idea of BERT's Masking Language Model. (3) In order to solve the problem of difficult and slow processing of high-dimensional data, the Deep SVDD algorithm is used for minimum superspheres capacity (VHM) self-supervision training task. Experiments have shown that LADB's combined performance is superior to the four representative log anomaly detection algorithms.

Pedagogical conditions for developing the readiness of master's degree applicants for intercultural interaction in the process of professional (piano) training

The article defines and characterizes the pedagogical conditions for forming the readiness of master's degree applicants for intercultural interaction in the process of professional (piano) training. It is emphasized that the tasks of the modern professional training of the applicants of the piano master's degree are not only to ensure a high level of professional knowledge of students, but also their practical preparation for social life and self-realization as an individual. It is emphasized that the humanization of the process of formation of the readiness for intercultural interaction in applicants of the professional master's degree implies the use of the latest technologies, active teaching methods, which would form the motives of educational and artistic activity, contribute to the discovery and development of the creative potential of the education applicant, self-realization of his/her personality; included in the process of learning the sphere of emotions, contributed to the development of cooperation. It has been proven that in the process of forming value orientations in the conditions of a multicultural artistic and educational environment, individual support of piano master's students is successfully implemented through individual counseling, creation of a positive moral and psychological climate in classes, mentoring support of the artistic and educational process, building an individual educational trajectory. Considering the formation and transformation of music-pedagogical and artistic-aesthetic paradigms in the system of modern music education, it is emphasized that the change in the paradigm of music education involves a shift of emphasis from the development in applicants of a complex of special knowledge, abilities and skills, focused on the formation of professional multicultural competence in the conditions of intercultural interaction, which allows us to interpret the artistic and educational space as a unity of creative personalities, united by channels of musical communication, who feel a sense of empathy for each other, mutual interest, are capable of active creative joint activity, overcoming interaction problems, creating a cross-cultural space

A problem-based approach in the formation of the self-educational competence of schoolchildren (based on the use of digital technology resources)

The article reveals the possibilities of the problem-based approach in solving the problem of the formation of the self-educational competence of schoolchildren. Various ways of influencing the development of their self-educational competence through the preparation of specially selected tasks are offered in lessons, in homework and in individual work with schoolchildren. It is substantiated that in lessons it is expedient from the point of view of the formation of self-educational competence to use typical tasks similar to typical tasks for independent work and creative tasks. It is shown that the use of educational, training and research tasks is productive for organizing homework. It has been revealed that in individual work it will be effective to perform preparatory, trial, basic tasks. At the same time, it has been proven that all the specified types of tasks involve the active use of digital technologies, which will allow for effective organization of educational activities, improve the motivation and cognitive interest of schoolchildren, ensure individualization of learning and create prerequisites for the development of independence of schoolchildren. In lessons, it is advisable to use simple digital applications, digital resources for creating your own content, digital activity environments and digital resources for online communication. It is effective to use digital resources-sources of educational information, tools for self-monitoring and self-diagnosis, as well as resources for creating your own digital content in homework. Digital resources-sources of information, tools for self-monitoring and self-diagnosis, digital activity environments and resources for personalized learning are useful for the individual work of schoolchildren. The article examines the potential of the problem-based approach to the formation of the self-educational competence of schoolchildren. The main components of self-educational competence are considered, namely: motivational-value, substantive-procedural and control-reflective. The impact of the problem-based approach on the development of each of these components is analyzed. It has been determined that the problem-based approach helps to increase students' motivation to study, stimulates their cognitive activity and develops critical thinking

Regional Load Forecasting Scheme for Security Outsourcing Computation

Smart grids generate an immense volume of load data. When analyzed using intelligent technologies, these data can significantly improve power load management, optimize energy distribution, and support green energy conservation and emissions reduction goals. However, in the process of data utilization, a pertinent issue arises regarding potential privacy leakage concerning both regional and individual user power load data. This paper addresses the scenario of outsourcing computational tasks for regional power load forecasting in smart grids, proposing a regional-level load forecasting solution based on secure outsourcing computation. Initially, the scheme designs a secure outsourcing training protocol to carry out model training tasks while ensuring data security. This protocol guarantees that sensitive information, including but not limited to individual power consumption data, remains comprehensively safeguarded throughout the entirety of the training process, effectively mitigating any potential risks of privacy infringements. Subsequently, a secure outsourcing online prediction protocol is devised, enabling efficient execution of prediction tasks while safeguarding data privacy. This protocol ensures that predictions can be made without compromising the privacy of individual or regional power load data. Ultimately, experimental analysis demonstrates that the proposed scheme meets the requirements of privacy, accuracy, and timeliness for outsourcing computational tasks of load forecasting in smart grids.

AssocKD: An Association-Aware Knowledge Distillation Method for Document-Level Event Argument Extraction

Event argument extraction is a crucial subtask of event extraction, which aims at extracting arguments that correspond to argument roles when given event types. The majority of current document-level event argument extraction works focus on extracting information for only one event at a time without considering the association among events; this is known as document-level single-event extraction. However, the interrelationship among arguments can yield mutual gains in their extraction. Therefore, in this paper, we propose AssocKD, an Association-aware Knowledge Distillation Method for Document-level Event Argument Extraction, which enables the enhancement of document-level multi-event extraction with event association knowledge. Firstly, we introduce an association-aware training task to extract unknown arguments with the given privileged knowledge of relevant arguments, obtaining an association-aware model that can construct both intra-event and inter-event relationships. Secondly, we adopt multi-teacher knowledge distillation to transfer such event association knowledge from the association-aware teacher models to the event argument extraction student model. Our proposed method, AssocKD, is capable of explicitly modeling and efficiently leveraging event association to enhance the extraction of multi-event arguments at the document level. We conduct experiments on RAMS and WIKIEVENTS datasets and observe a significant improvement, thus demonstrating the effectiveness of our method.

Effectiveness of Task-Oriented Training in Improving Gait in a Child with Ataxic Cerebral Palsy: A Case Report

This case report examines the impact of task-oriented training on gait improvement in a child with Ataxic Cerebral Palsy (CP), a rare subtype characterized by balance and coordination difficulties. The goal was to determine the effectiveness of this intervention in enhancing gait stability, motor coordination, and overall mobility. A 6-year-old boy with Ataxic CP underwent task-oriented training for five days per week over a four-week period. Each session lasted one hour and included lower limb strengthening, coordination exercises, stretching, and walking practice. The Holden Gait Scale (HG) and Gross Motor Function Measure (GMFM) were used to evaluate changes in the child's gait and motor function pre- and post-intervention. After one week of therapy, the child demonstrated reduced tremors and attempted to climb stairs independently. By the end of the four-week period, the child was able to walk approximately 200 meters indoors without displaying a high-guard posture, although supervision was still required when navigating stairs. Improvements in gait stability and balance were observed, though the scissor gait persisted. Task-oriented training proved to be an effective approach in improving gait stability and motor coordination in a child with Ataxic CP. However, challenges such as upper limb high guarding and scissor gait remained, indicating the need for further personalized rehabilitation strategies. Future research should focus on integrated interventions to address these residual issues and enhance the long-term outcomes for children with Ataxic CP.

Force-Field based assisted control for upper-limb rehabilitation robots

Robot-assisted training with an assist-as-needed (AAN) control strategy has been used in clinical investigations to effectively enhance the subject’s active effort and promote the rehabilitation of individuals with upper-limb motor impairment. In this paper, a force field-based AAN control scheme is proposed for robot-assisted upper limb rehabilitation training and then implemented in a robot-assisted rehabilitation training system (RaRTS). A spatial varying assistance force field is constructed around the predetermined trajectory in this scheme, and then a varying force field coefficient based on the subject’s motor performance is adopted to deliver the AAN features with temporal freedom. The normal force field, tangential force field, and viscous force field are the three components of the constructed spatial varying assistance force field. Furthermore, the proposed scheme contains three training modes that are switched according to the subject’s trajectory tracking error. Finally, task-oriented continuous training experiments with different force field factors are conducted on able-bodied subjects to evaluate and validate the performance and feasibility of the proposed scheme. Preliminary experimental results and statistics analysis suggest that the subjects were able to keep much closer to the predetermined trajectory through an appropriate selection of force-field assistance factors. Furthermore, the robot would apply an assistance force as needed as the motor performance deteriorate, which could prevent the subject from losing confidence due to poor performance.

Artificial intelligence derived large language model in decision-making process in uveitis

BackgroundUveitis is the ophthalmic subfield dealing with a broad range of intraocular inflammatory diseases. With the raising importance of LLM such as ChatGPT and their potential use in the medical field, this research explores the strengths and weaknesses of its applicability in the subfield of uveitis.MethodsA series of highly clinically relevant questions were asked three consecutive times (attempts 1, 2 and 3) of the LLM regarding current uveitis cases. The answers were classified on whether they were accurate and sufficient, partially accurate and sufficient or inaccurate and insufficient. Statistical analysis included descriptive analysis, normality distribution, non-parametric test and reliability tests. References were checked for their correctness in different medical databases.ResultsThe data showed non-normal distribution. Data between subgroups (attempts 1, 2 and 3) was comparable (Kruskal-Wallis H test, p-value = 0.7338). There was a moderate agreement between attempt 1 and attempt 2 (Cohen’s kappa, ĸ = 0.5172) as well as between attempt 2 and attempt 3 (Cohen’s kappa, ĸ = 0.4913). There was a fair agreement between attempt 1 and attempt 3 (Cohen’s kappa, ĸ = 0.3647). The average agreement was moderate (Cohen’s kappa, ĸ = 0.4577). Between the three attempts together, there was a moderate agreement (Fleiss’ kappa, ĸ = 0.4534). A total of 52 references were generated by the LLM. 22 references (42.3%) were found to be accurate and correctly cited. Another 22 references (42.3%) could not be located in any of the searched databases. The remaining 8 references (15.4%) were found to exist, but were either misinterpreted or incorrectly cited by the LLM.ConclusionOur results demonstrate the significant potential of LLMs in uveitis. However, their implementation requires rigorous training and comprehensive testing for specific medical tasks. We also found out that the references made by ChatGPT 4.o were in most cases incorrect. LLMs are likely to become invaluable tools in shaping the future of ophthalmology, enhancing clinical decision-making and patient care.

A passive convex optimal control algorithm for teleoperating a redundant robotic arm in minimally invasive surgery

AbstractIn recent years, the remote center of motion (RCM) constraint has moved from purely mechanical to software implementation, enabling the use of serial manipulators in robotic‐assisted minimal invasive surgery. However, ensuring safety with software‐based RCM presents challenges. This article addresses this issue by introducing a novel control algorithm for a 7‐DOF redundant robotic arm, taking into account a software RCM while considering system passivity and kinematic constraints. The algorithm is both a control optimizer and a robot kinematics inversion, enabling precise and dexterous control for various surgical tasks and other control applications. By formulating a quadratic optimization problem under linear constraints, the algorithm guarantees the robotic arm's performance, safety, and stability under communication delay. Experimental validation demonstrates the effectiveness and accuracy of the control algorithm in executing a surgical training task (peg‐and‐ring) during bilateral teleoperation. The results highlight the successful implementation of the RCM constraint, ensuring patient safety and optimizing the manipulation capabilities of the robotic arm.

Learning to drive as humans do: Reinforcement learning for autonomous navigation

This article introduces an approach aimed at enabling self-driving cars to emulate human-learned driving behavior. We propose a method where the navigation challenge of autonomous vehicles, from starting to ending positions, is framed as a series of decision-making problems encountered in various states negating the requirement for high-precision maps and routing systems. Utilizing high-quality images and sensor-derived state information, we design rewards to guide an agent’s movement from the initial to the final destination. The soft actor-critic algorithm is employed to learn the optimal policy from the interaction between the agent and the environment, informed by these states and rewards. In an innovative approach, we apply the variational autoencoder technique to extract latent vectors from high-quality images, reconstructing a new state space with vehicle state vectors. This method reduces hardware requirements and enhances training efficiency and task success rates. Simulation tests conducted in the CARLA simulator demonstrate the superiority of our method over others. It enhances the intelligence of autonomous vehicles without the need for intermediate processes such as target detection, while concurrently reducing the hardware footprint, even though it may not perform as well as the currently available mature techniques.

Preliminary validation of the virtual bariatric endoscopic simulator

Background and AimsObesity is a global health concern. Bariatric surgery offers reliably effective and durable weight loss and improvements of other comorbid conditions. However, the accessibility of bariatric surgery remains limited. Minimally invasive techniques, including endoscopic sleeve gastroplasty (ESG), have emerged to bridge this gap. To effectively complete the ESG procedure, one requires skill in multiple complex interventional endoscopic maneuvers. This requisite expertise poses challenges for training in this burgeoning field. MethodsWe designed the Virtual Bariatric Endoscopic (ViBE) simulator software to mimic the ESG procedure accurately. It features a detailed simulation of an endoscope equipped with an endoscopic suturing system and a high-resolution stomach, enhancing the visualization of the procedural details. Furthermore, it incorporates performance metrics utilizing a reverse scoring system to evaluate users’ proficiency in tasks such as argon plasma coagulation (APC) marking, suturing, and cinching. To validate the simulator, we conducted a study involving experts and novices at the Indiana University School of Medicine, where participants engaged with the simulation environment in a series of training tasks. ResultsTwelve participants, comprising five experts and seven novices, participated in the study. They were asked to complete a post-questionnaire featuring seven items, rating each on a Likert scale. The APC task realism received the highest score, averaging 3.83. The usefulness of improving endoscopic technical skills averaged 3.08, with the realism of cinching the knot and suturing tasks receiving scores of 3.17 and 3.25, respectively, suggesting a generally positive reception. Automated performance metrics indicated that, on average, experts outperformed novices by 10.83 points. ConclusionThe ViBE simulation strives to replicate the steps of the ESG within a virtual environment. Our primary objective in developing this simulator is to enhance the learning curve for endoscopic suturing and ESG techniques, thereby safely extending these skills to a broader patient base.

Local organization of spatial and shape information in the primate prefrontal cortex.

The current understanding of sensory and motor cortical areas has been defined by the existence of topographical maps across the brain surface, however, higher cortical areas, such as the prefrontal cortex, seem to lack an equivalent organization, and only limited evidence of functional clustering of neurons with similar stimulus properties is evident in them. We thus sought to examine whether neurons that represent similar spatial and object information are clustered in the monkey prefrontal cortex and whether such an organization only emerges as a result of training. To this end, we analyzed neurophysiological recordings from male macaque monkeys before and after training in spatial and shape working memory tasks. Neurons with similar spatial or shape selectivity were more likely than chance to be encountered at short distances from each other. Some aspects of organization were present even in naïve animals, however other changes appeared after cognitive training. Our results reveal that prefrontal microstructure automatically supports orderly representations of spatial and object information.

Quantifying the Transition from Unconscious to Conscious Detection of Temporal Patterns in Vigilance Tasks: A Unique Adaptation of Mackworth's Clock Test.

This study investigates the cognitive mechanisms underlying vigilance and pattern recognition using a novel adaptation of Mackworth's Clock Test. We aimed to quantify the time it takes for temporal patterns detected unconsciously through implicit learning to surface in the conscious mind within a dynamic vigilance task environment. Forty-eight participants detected random and non-disclosed rhythmic anomalous clock hand movements in this setting. Our results indicate significant variability in detection accuracy, reaction times, and the ability to recognize the hidden pattern among participants. Notably, 23% of all participants and 56% of those who consciously reported the pattern exhibited statistically lower reaction times indicative of knowledge of the pattern 40 s before conscious identification. These findings provide valuable insights into the transition from unconscious to conscious detection, highlighting the complexity of sustained attention and pattern recognition. The study's implications extend to designing training programs and tasks for high-stakes professions requiring prolonged vigilance. Future research should further explore the cognitive and neural correlates of these processes and the impact of task complexity on performance.

Variations in External and Internal Intensities and Impact of Maturational Age on Soccer Training Tasks.

During peak height velocity, adjusting training intensity is crucial for optimizing performance and minimizing injury risk. This cross-sectional study compares external and internal intensities in different training tasks (analytical tasks, small-sided games, and training matches) and analyzes their effect on the maturation age of young players. Fifty-five U-15 and U-16 boys from two soccer clubs in southwestern Spain were monitored using inertial movement units and heart rate monitors to report training intensities. Anthropometric data and birthdates were collected to estimate maturation age. The Friedman test and Durbin-Conover post hoc test identified specific differences between groups, and Spearman's rank correlation coefficients assessed variable impacts. Training matches showed significantly higher distance covered, maximum and average speed, and average heart rate compared to small-sided games and analytical tasks. High-intensity actions and sprints were significantly higher (p < 0.0001) during training matches compared to analytical tasks and during small-sided games compared to analytical tasks. Player load per minute was significantly highest (p < 0.05) during training matches, followed by small-sided games, and lowest in analytical tasks. Positive correlations between maturational age and high-intensity actions, accelerations, and decelerations indicated higher intensity (p < 0.05) in more mature players. A negative correlation between player load per minute and maturational age suggested more efficient intensity management in mature players. These findings highlight the importance of considering biological maturation and training task variability in youth athletes' development.

Efficacy of Dry Needling in Enhancing Hand Function and Reducing Spasticity in MCA Stroke Patients: A Prospective Case Report

Stroke, particularly involving the middle cerebral artery (MCA), often leads to persistent disability, significantly impairing hand function and inducing severe spasticity in the forearm flexors due to complex motor pathway involvement. This case report evaluates the effectiveness of combining dry needling therapy (DNT) with task-oriented training (TOT) in reducing spasticity and improving hand function in a post-MCA stroke patient. A 61-year-old male with right-side hemiplegia following a left MCA stroke, compounded by a history of diabetes, hypertension, and recent kidney stone diagnosis, participated in this study. The patient experienced severe functional limitations due to spasticity (grade 3 on the Modified Ashworth Scale) and scored 31 on the CAHAI-13, indicating significant hand dysfunction. DNT was applied to key forearm muscles (flexor carpi radialis, flexor carpi ulnaris, flexor digitorum superficialis, flexor digitorum profundus, and pronator teres), followed by TOT targeting daily tasks. Post-intervention, the patient demonstrated significant improvements: spasticity was reduced (Modified Ashworth Scale score decreased from grade 3 to grade 2), and muscle hypertonicity, as measured by EMG, was reduced in all targeted muscles. Functional hand use improved markedly, with the CAHAI-13 score increasing from 31 to 45. These improvements led to enhanced independence in daily activities. This case report highlights the clinical relevance of integrating DNT with TOT in stroke rehabilitation. The combined intervention not only reduced spasticity and improved neuromuscular function but also led to meaningful gains in hand function, enhancing the patient’s ability to perform daily activities independently. Given the promising results, this combined approach could serve as a practical and effective strategy for improving motor recovery and functional independence in post-stroke patients. Further studies with larger sample sizes are necessary to confirm these findings and explore the broader application of this approach in clinical practice.

The effect of dual-task training on postural and cognitive performances in adolescents with down syndrome

BackgroundThe dual task training (DTT) has been shown to improve postural and cognitive performances in many populations. Therefore, it seems of interest to explore the effect of such training modality in individuals with Down syndrome (DS) presenting both cognitive and postural deficits. AimsThis study explored the effect of a DTT compared to a single task training (STT) on postural and cognitive performances in adolescents with DS. Methods and proceduresThe center of pressure velocity (CoPVm) and the cognitive performance in the selective word recall test were recorded under single task (ST) and DT conditions before and after 8 weeks in the STT group (STTG), the DTTG and the control group (CG). Outcomes and resultsBefore training, CoPVm values increased and cognitive performances decreased (p < 0.001) during the DT compared to ST conditions in all groups. After training, CoPVm decreased (p < 0.001) in the DTTG and the STTG under the ST conditions. However, under DT conditions, these values decreased (p < 0.001) only in the DTTG. The cognitive performance increased (p < 0.001) only in the DTTG in both ST and DT conditions. ConclusionThe DTT is strongly recommended to improve both postural and cognitive performances in individuals with DS notably in DT conditions.

AI Services-Oriented Dynamic Computing Resource Scheduling Algorithm Based on Distributed Data Parallelism in Edge Computing Network of Smart Grid

Massive computational resources are required by a booming number of artificial intelligence (AI) services in the communication network of the smart grid. To alleviate the computational pressure on data centers, edge computing first network (ECFN) can serve as an effective solution to realize distributed model training based on data parallelism for AI services in smart grid. Due to AI services with diversified types, an edge data center has a changing workload in different time periods. Selfish edge data centers from different edge suppliers are reluctant to share their computing resources without a rule for fair competition. AI services-oriented dynamic computational resource scheduling of edge data centers affects both the economic profit of AI service providers and computational resource utilization. This letter mainly discusses the partition and distribution of AI data based on distributed model training and dynamic computational resource scheduling problems among multiple edge data centers for AI services. To this end, a mixed integer linear programming (MILP) model and a Deep Reinforcement Learning (DRL)-based algorithm are proposed. Simulation results show that the proposed DRL-based algorithm outperforms the benchmark in terms of profit of AI service provider, backlog of distributed model training tasks, running time and multi-objective optimization.

Bridging the Gap Between Training and Competition in Elite Rink Hockey: A Pilot Study.

Monitoring training load and competition load is crucial for evaluating and improving athlete performance. This study proposes an applied approach to characterize and classify the training task specificity in relation to competition in a top-level rink hockey team, considering external and internal load from training tasks and competition. Training tasks and game demands have significant dose-response differences, and exercises can be classified successfully based on their physiological and biomechanical demands. Cross-sectional study. Level 5. Ten elite-level male rink hockey players participated in this study. Players were monitored on 6 different task categories during 8 training sessions and 2 official games. A linear mixed model with random intercepts was used to compare training tasks and competition load, accounting for individual repeated measures. A 2-step cluster analysis was performed to classify the training tasks and games based on physiological and biomechanical load, employing log-likelihood as the distance measure and Schwartz's Bayesian criterion. Average heartrate , maximum heartrate, and high-speed skating (18.1-30 km/h) were the best physiological load predictors, while the most effective biomechanical load predictors were impacts [8-10] g(n), decelerations [-10 to -3]m/s²(n), and accelerations [3-10]m/s²(n). Different physiological and biomechanical responses were verified between training tasks and match demands. A 4-quadrant efforts assessment for each task category revealed that training tasks used by the team in the analysis presented lower biomechanical and physiological load demands than competition. Training tasks failed to adequately replicate the specific demands of competition, especially regarding high mechanical stress, such as the absence of high-intensity impacts and decelerations. This method of classification of training tasks may allow coaches to understand further the specificity and contribution of each task to competition demands, consequently improving the capacity of load management and the preparedness and readiness of players for competition.

The Effect of Ten Weeks of Dual-Task Training on Functional Balance and Gait Speed Under Dual-Task Conditions in Bipolar Patients

Background: In mood disorders, disturbance in balance control and walking is a main motor symptom. Objectives: The aim of this study was to explore the impact of ten weeks of dual-task training on balance and gait speed under dual-task conditions in bipolar patients. Methods: Thirty male patients (age mean 43.13 ± 8.91) with bipolar disorder living in Urmia Psychiatric Center were assigned to the control group and the dual task training group. Dual-task training with moderate intensity, including 20 sessions, 2/week, per 75 min. To detect the changes in the balance and the gait speed, a timed up and go Test (TUG) under dual task conditions and a 10-m walking course under dual task conditions were administered before and after the exercise protocol. It was analyzed using analysis of covariance test. Results: The findings showed that the two groups had significant differences in TUG under motor dual task conditions, gait speed under motor dual task conditions, and gait speed under cognitive dual task conditions (P ≤ 0.05). Conclusions: Considering the effectiveness of cognitive dual-task training in improving gait speed and functional balance, dual-task training may be a potentially important supplement to conventional treatment in people identified as being at high risk of falling.