Sung-Taek Chung

Developing a device using accelerometers and EMG for hand movement recognition

In this study, rehabilitation training in motor skills for hemiplegia patients comprises repetitive and passive contents. To reduce boredom of rehabilitation training and provide active training, a movement-recognizing device and contents using it were manufactured. An accelerometer for detecting movement recognition and EMG sensor signals for detecting minute muscle variation were used to recognize hand movement. The signals generated by movement were made to move like a mouse, which was connected to rehabilitation training contents using this function.

On multicast communications with minimum resources

We have developed and evaluated an efficient heuristic algorithm to construct a multicast tree with the minimum cost. Our algorithm uses multiple candidate paths to select a path from source to each destination, and works on directed asymmetric networks. We also show that our proposed algorithm have a perform gain in terms of tree cost for real life networks over existing algorithm. The time complexity of our algorithm is O( D (m + nlogn) ) for a m-arc n-node network with D number of members in the multicast group, and is comparable to well-known algorithm, TM [13], for a multicast tree construction in terms of the time-complexity. We have performed empirical evaluation that compares our algorithms with the others on large networks.

Evaluation of Hand Grip Strength and EMG Signal on Visual Reaction

Hand grip strength has been utilized as an indicator to evaluate the motor ability of hands, responsible for performing multiple body functions. It is however difficult to evaluate other factors (other than hand muscular strength) utilizing the hand grip strength only. The purpose of this study was analyzed the motor ability of hands using EMG and the hand grip strength, simultaneously in order to evaluate concentration, muscular strength reaction time, instantaneous muscular strength change, and agility in response to visual reaction. In results, the average time (and their standard deviations) of muscular strength reaction EMG signal and hand grip strength was found to be 209.6±56.2 ms and 354.3±54.6 ms, respectively. In addition, the onset time which represents accel- eration time to reach 90% of maximum hand grip strength, was 382.9 ± 129.9 ms. Results in visual reaction (on) indicate the dif- ferences in muscular strength agility and concentration of participants in regards to visual reaction.

Measurement of Motion Range to Improve of Body Balance and Its Training Contents

Wrong lifestyle habits can be a major factor for reduced balance ability that may cause difficulties in performing daily activities due to musculoskeletal disorders, gait abnormality, a fall, and other problems. This study assessed range of motion (ROM) using Inertial Measurement Unit (IMU) sensors placed at the waist or upper and lower limbs by measuring the maximum rotation angle of the body to the directions of the sagittal, coronal and transverse planes in real time. Directions with a low body balance are identified using ROM data and based on the analysis results, training content is recommended to improve reduced balance ability in corresponding directions. Furthermore, this content is designed to offer intensive balance training toward a specific direction by providing selection modes of balance training in the desired directions. Motivation for training can be enhanced by comparing changes in measurements assessed before and after balance training. The results of this study are expected to aid improving reduced physical activities in elderly individuals with reduced muscle strength, body balance and walking ability using the content tailored to measure changes in range of motion and improve balance ability.

Experimental Verification of Contents Usability for Upper Limbs Rehabilitation in Patients with Hemiplegia.

Patients with hemiplegia with upper limb impairment have difficulties in performing coordinated bimanual movement of hands and activities of daily living (ADL) due to reduced performance in functional movements. This study developed a serious game for upper limb rehabilitation training using a balance handle specifically designed for patients with hemiplegia. Most current serious games designed for healthy people constitute difficulties for hemiplegia patients to use. For this reason, the simple game interface adapted to accommodate patient disabilities and the content tailored to the needs of these patients are warranted to provide gaming environments according to each patient’s condition. Therefore, this study designed the tailored content that offers an environment appropriate for each patient by evaluating range of motion (inclination of a balance handle) of hemiplegia patients, movement speed and others. This game content is designed based on extension, flexion and movement assessment guidelines for the Fugl-Meyer Assessment (FMA), and measures muscle activities of triceps, biceps and deltoid muscles which demonstrate changes in upper limb motor function while performing the content. The experiment results revealed that triceps, biceps and deltoid muscles were activated. This content is expected to be utilized for rehabilitation purposes in hemiplegia patients with upper limb impairment.

Development of Cognitive Function Evaluation Contents for Mobile Based on MMSE-DS

Mild cognitive impairment (MCI) is considered as the intermediate stage between the cognitive changes of normal aging and dementia, which is marked by decline in memory but preserved ability to perform activities of daily living compared with that of healthy individuals of the same age. These patients need to delay the progression to dementia through early diagnosis and use the standardized Mini-Mental State Examination (MMSE) as an assessment tool.

Personalized Recommendation System for Efficient Integrated Cognitive Rehabilitation Training Based on Bigdata

In this study, a personalized recommendation system for efficient integrated cognitive rehabilitation training based on bigdata was developed. The system consists of 5 main phases (collection, storage, processing, analyzing, visualization). First, in the pre-processing process before the collection phase, resulting scores from multiple cognitive rehabilitation contents and patients’ personal information are saved in database. In the collection/storage phases, the patient information saved in the database is saved in bigdata platform. In the processing phase, the data are processed/refined in a necessary form to be utilized in the analysis and statistical processing program, R. Lastly, in the analysis/visualization phases, personalized contents of integrated cognitive rehabilitation training are recommended to patients using the K-Means method of the unsupervised learning algorithms and spiral model through patients’ personal information, MMSE results, cognitive rehabilitation contents results based on the processed/refined data. Patients can utilize the personalized recommendation system for integrated cognitive rehabilitation training based on bigdata to implement cognitive function evaluation and personalized training at home.

Development of Serious Game and Integrated Management Service Model for the Cognitive Rehabilitation

Cognitive impairment that can be caused by brain disorders or aging affects quality of life by interfering with abilities to perform activities of daily living (ADL). For this reason, this study developed serious games for the purpose of rehabilitation and proposes an integrated management service model that can recommend the difficulty level of games and manage changes in serious results according to user’s cognitive function ability during rehabilitation. The areas of cognitive training are designed in consideration of game elements in relation to attention, concentration, memory, orientation, executive function and ADL. Based on the analysis results, we established a database for recommendation of game level in the next training session. Attention is further subdivided into selective, sustained and divided attention. Memory is subcategorized into topological memory, associative memory, verbal memory, immediate memory, visual short-term memory, working memory and recall memory. Executive function is classified into inhibitory control, cognitive flexibility and working memory. ADL includes physical ADL and instrumental ADL. This model comprises a total of 19 serious games for cognitive rehabilitation and each game consists of 3 levels of increasing difficulty. Game interface is created with a user-centered design which focuses on visualization for elderly-friendly use. Our integrated management service model for cognitive rehabilitation can provide training on cognitive function in different areas as a single program and recommend the difficulty level of games depending on player’s ability.

Development of Serious Game for the Upper Arms Rehabilitation: “Balance Ball Rhythm Game” Case Study

Accuracy, repeatability and activity are very critical factors for rehabilitation of hemiplegic patients. Rehabilitation exercise should be done regardless of space, time and cost. Recently, interesting functional games, which induce active participation, have gained increasing attention. In the current study, a balance ball has been developed that can contract the user muscle and help in natural joint rotation by stimulating the upper arm muscle of hemiplegic patients. Additionally, a functional game was also developed that can engage the patients with rhythm game and training contents. The balance ball can detect the upper arm motion by an acceleration sensor and offered sense of reality and immersion with buttons and haptic sensors. The game applied Fitt’s law to test accurate motion and two tutorial contents that induced their motion based on MFT. The level of difficulty can also be chosen to help intensive training for the motion with low scores from the tutorials and the patients can even do the upper arm rehabilitation exercise while listening to music.

3D modeling of musculoskeletal Cobb's angles

The growth of industries has led to not only the improvement in quality of life but also increased musculoskeletal diseases due to accumulation of micro-sized damages in muscles or other tissues due to simple and repetitive lifestyle. Similarly, having curvature in the vertebrae is referred to as scoliosis, which is increasingly occurring mostly in teenagers. To diagnose such scoliosis easily, the patient was allowed to easily perceive about scoliosis by realizing a 3D backbone data through usage of the tilt sensor and encoder sensor. This is harmless to the human body compared to the expensive X-rays or MRI in addition to being possible to be mobile. The spinal curvature was made easily understandable by mapping the spinal data obtained through the devices that is proposed in this study into 3D modeling spinal image. By utilizing such image, Cobbs angle was able to be easily obtained.