Earticle

본 연구는 전 세계적으로 급증하는 한국어 학습 수요에 대응하여, 스마트폰만으로 정밀한 발음 교정이 가능한 시스템을 개발하는 것을 목적으로 한다. 기존의 2차원 평면 기반 음향 분석이 간과해 온 발화의 시공간적 안정성 (Stability)과 한국어 모음 변별의 핵심 기제인 입술 모양(lip rounding)을 통합적으로 분석한다. 연구 방법으로는 모바일 환경에서 선형 예측 부호화(LPC)를 통해 포먼트를 추출하고 커널 밀도 추정(KDE)을 적용하여 ‘3차원 포먼트 밀도’ 지형도 를 생성하며, 동시에 MediaPipe Face Mesh 기술로 입술의 개구도와 원순성을 정량화한다. 연구 결과, 개발된 시스템은 학습자가 생성하는 모음의 음향적 분산을 3차원 산맥 형태로 시각화하여 발음의 견고성을 인지하게 하였으며, 유사한 포먼 트 값을 가지더라도 입술 모양이 잘못된 경우를 실시간으로 탐지하여 교정 효율을 높였다. 본 연구는 ‘점’ 중심의 발음 평가를 ‘분포’ 중심으로 전환하고, 음성학과 컴퓨터 비전 기술을 융합하여 모바일 컴퓨터 기반 컴퓨터 보조 발음훈련 (computer-assisted pronunciation training; CAPT) 시스템의 새로운 표준을 제시하였다.

The rapid globalization of Korean culture has precipitated a surge in Korean language learners, yet effective pronunciation training remains a significant pedagogical bottleneck. Traditional Computer-Assisted Pronunciation Training (CAPT) systems typically rely on two-dimensional acoustic analysis or automatic speech recognition confidence scores. These methods fundamentally fail to capture the temporal stability of vowel production—treating pronunciation as a static point rather than a dynamic distribution—and neglect the critical articulatory role of lip kinematics. This research presents the development of a novel smartphone-based biofeedback system that integrates 3D Vowel Space Analysis with Real-Time Lip Contour Extraction. By employing Kernel Density Estimation (KDE) on accumulated formant trajectories, the system visualizes the "density" of a learner's pronunciation in three dimensions. Simultaneously, a computer vision pipeline utilizes the MediaPipe Face Mesh to extract facial landmarks, providing immediate visual feedback on lip rounding. This multimodal approach effectively decouples acoustic errors from articulatory misconfigurations.

Objective To Investigate the synergic effects of short-term constraint-induced movement therapy (CIMT) and visual biofeedback training (VBT) in subacute stroke patients.Methods Thirty-two subacute stroke patients were enrolled and randomly assigned to one of three groups: short-term CIMT with VBT, VBT only, and control groups. We applied CIMT for an hour daily during VBT instead of the ordinary restraint time, referred to as ‘short-term’ CIMT. Short-term CIMT with VBT group received simultaneous VBT with CIMT, whereas the VBT the only group received VBT without CIMT for an hour a day for 2 weeks. The control group received conventional occupational therapy (OT) alone. Patients underwent the Purdue Pegboard Test, the JAMAR grip strength test, the Wolf Motor Function Test, the Fugl-Meyer Assessment (upper extremity), Motricity index and the Korean version of Modified Barthel Index test to evaluate motor functions of the hemiplegic upper limb at baseline, post-treatment, and 2 weeks after treatment.Results No significant differences were observed between short-term CIMT with VBT and VBT only groups. Both groups showed significantly higher scores compared to the control group in the WMFT and FMA tests. However, the short-term CIMT with VBT group showed significant improvement (p<0.05) compared with the control group in both grasp and pad pinch at post-treatment and 2 weeks after treatment while the VBT only group did not.Conclusion Short-term CIMT with VBT group did not show significant improvement of hemiplegic upper limb function of subacute stroke patients, compared to VBT only group. Larger sample sizes and different restraint times would be needed to clarify the effect.

Objective: To investigate the effect of dynamic balance training with center of pressure (COP) games using visual biofeedback of COP movement in stroke patients. Method: Thirty-eight stroke patients were randomly assigned to the static or dynamic balance training group, or control group. Static group received weight-shift training with visual biofeedback of weight distribution bar. Dynamic group received dynamic balance training using COP games. Each training session lasted 30 minutes, 3 times in a week during 3 weeks. Control group received traditional rehabilitation only. The ability of balance control was assessed using posturography by total path distance, frequency of sway, and COP weight-spectrums during standing, and during presentation of virtual moving surround. Total path error was also assessed during sine curve and dynamic circle trace test. Results: In static group, there were significant improvement in the total path distance during comfortable standing and path error during sine curve tracing. In dynamic group, significant improvements were found in the path error during sine curve and dynamic circle tracing. Conclusion: Dynamic training group showed more improvement in dynamic balance ability, whereas static group showed more changes in the static balance ability. The visual biofeedback training is more specified for balance rehabilitation in stroke patients.

본 연구의 목적은 근전도 바이오피드백 장비를 사용하여 움직임 조절운동을 수행할 때 엉덩관절 근육에 미치는 영향에 대해 알아보고자 하였다. 본 연구는 유사실험설계 단일집단 검사전-검사후 설계이다. 건강한 성인 21명이 본 연구에 참여하였다. 모든 대상자는 근전도 바이오피드백 장비를 사용하여 엉덩관절의 움직임 조절 운동을 총 20분간 수행하였다. 적은무릎 구부리기 검사 수행동안 엉덩관절의 근활성도 변화를 확인하기 위해 중간볼기근, 큰볼기근, 넙다 리근막긴장근, 넙다리 곧은근의 활성도를 측정하였다. 본 연구 결과에서 중재 후 큰볼기근의 유의한 변화를 확인하였다. 이러한 결과는 근전도 바이오피드백 장비를 적용한 엉덩과절 움직임 조절운동이 엉덩관절의 안쪽돌림을 조절하고 엉덩 관절 가쪽돌림근을 활성화시키는 효과적 중재방법임을 확인할 수 있었고, ICT 융합기술을 접목한 재활운동이 임상에서 효과적인 중재가 될 수 있음을 보여주었다.

The aim of this study was to determine the effects of movement control exercise of hip joint using visual EMG biofeedback on hip joint muscles in healthy adults. This study was nonequivalent one group Pre-post test design. Twenty-one healthy adults were participated in the study. all subjects conducted movemnet control exercise(MCE) using electromyography(EMG) biofeedback of hip joint durng 20 min. The outcome measures included surface eletromyography. Surface electromyography data were collected from the Gluteus medius (Gmed), Gluteus maximus (Gmax),and Tensor fasciae latae(TFL), rectus femoris(RF) during small knee bending (SKB) test. There was a significant difference in Gluteus maximus muscle activity between the pre-test and the post-test (p < 0.05). The findings suggest that Movement control exercise using EMG biofeedback for limiting hip internal rotation is effective in activating the hip external rotator muscles. in addition, this study showed that rehabilitation exercise combined with ICT convergence technology could be an effective intervention in clinical practice.

Background: Chronic ankle instability (CAI) is common in young adults after recurrent ankle sprains and is associated with impaired proprioception, neuromuscular control, and balance. Conventional proprioceptive training is widely used, but visual biofeedback systems such as PhysioSensing, which provide real-time center of pressure feedback, may enhance training effects. This study compared PhysioSensing-based proprioceptive training with conventional ankle stability exercises in young adult males with functional ankle instability. Methods: Forty male university students were randomly assigned to a PhysioSensing-based proprioceptive training (PBP) group or a conventional ankle stability exercise (CAS) group. Both groups trained for 20 minutes, five times per week for four weeks. Outcomes included the Cumberland Ankle Instability Tool (CAIT), Foot and Ankle Ability Measure (FAAM), Star Excursion Balance Test (SEBT), and Time-Based Lateral Balance Test (TBLT), assessed before and after the intervention. Results: Both groups showed significant improvements over time in CAIT, FAAM, SEBT, and TBLT scores (p<.05). Significant group × time interaction effects were observed for SEBT and TBLT (p<.01), indicating greater improvements in dynamic balance and lateral stability in the PBP group compared to the CAS group. Additionally, FAAM-Sport scores showed a significant interaction effect (p=.01), with the PBP group demonstrating superior improvements in sports-related ankle function. Conclusion: Both training programs improved ankle function and balance; however, PhysioSensing-based training produced superior gains in dynamic balance and lateral stability, suggesting that real-time visual biofeedback enhances neuromuscular control in CAI rehabilitation.

Purpose: The purpose of this study is to determine the effects of visual biofeedback training on the recovery of balance function in stroke patients. Methods: A total of 30 patients with stroke were chosen as the subjects of this study. The subjects were randomly divided into either the visual biofeedback balance training group (experimental group; n=15) or the general balance training group (control group; n=15). The visual biofeedback balance training and general balance training were implemented for 30 minutes a day, three times a week, for a total of four weeks. The subjects’ balance ability was measured before and after the interventions. Results: The shift length and surface area of the center of the body decreased in both the experimental group and the control group, with the difference being statistically significant. The shift length and surface area of the center of the body both decreased more in the experimental group than in the control group, and there was a statistically significant between-group difference. Conclusion: The experimental group showed a greater improvement in terms of the balance ability of patients with stroke than the control group. Therefore, we believe that visual biofeedback balance training can be effectively applied for the improvement of balance ability in patients with stroke.

Background: After a stroke, the control of the trunk muscle may be severely impaired. Due to the importance of trunk control in complex daily postures, the ability to adopt a correct sitting posture is considered a determinant of the recovery of independent function after a stroke. Objects: The purposes of this study were to compare differences in buttock pressure between the left and right sides of hemiplegic patients and differences in their pelvic tilting angles (sagittal and coronal planes) after sitting training with visual biofeedback (VBF) in real time. Methods: Twenty-two individuals with unilateral strokes (11 left-side and 11 right-side hemiplegic stroke patients) participated in this study. Buttock pressure was measured using a pressure mat, and pelvic angles were measured using a palpation meter. Results: The asymmetry of pressure between the right and left (first and third chamber) sides was significantly decreased after the VBF training. The measurements obtained using the palpation meter revealed a significant decrease in the pelvic angles pre- versus post-intervention. Conclusion: VBF training may be distribute a patient's buttock pressure equally while in a sitting posture and increase the length of time a stroke patient can maintain a symmetrical sitting posture. It can also improve pelvic control while sitting in a neutral position.

PURPOSE: The purpose of this study is to examine the impact of balance training on a three-dimensional balance trainer that provides the up-and-down vertical movement of the knee joint and left-and-right horizontal movement, along with visual feedback on the functional ability of children with spastic cerebral palsy (CPs). METHODS: 8 CPs participated in this study. The experiment was implemented for 40 minutes, three times a week for a total of six weeks. The subjects received general physiotherapy for 15 minutes in each session focused on balance and walking, as based on the neuro-developmental treatment theory. Balance training was performed for 20 minutes on a three-dimensional balance trainer where knee joint movement providing visual feedback is applied. The evaluations were conducted before and after the test, and posture sway was measured using 10 Meter Walking Test (10MWT), Timed Up & Go Test (TUG), and the Good Balance System to evaluate the functional ability and balance of the subjects RESULTS: 10MWT was not statistically significant (p>.05). On the contrary, TUG and postural sway indicate static balance showed a statistically significant decrease (p<.05). In a static balance test using the Good Balance System, the average moving speed statistically significantly decreased in the AP and ML directions (p<.05), and the mean velocity moment also significantly decreased (p<.05). CONCLUSION: These findings suggested that balance training using the three-dimensional balance trainer, with the features of visual feedback and up-and-down knee joint movement effects on increasing dynamic and static balance.

Psychogenic dysphagia is a deglutition disorder characterized by a fear of swallowing, with no structural or functional causes. This report presents the case of a young male patient who had severe malnutrition due to psychogenic dysphagia and was provided visual biofeedback using fiberoptic endoscopic evaluation of swallowing (FEES). A healthy 25-year-old man presented to our clinic with a complaint of throat discomfort when swallowing that had started 6 months prior. As the symptoms worsened, he became fearful of food spreading to his lungs after swallowing and the development of respiratory difficulties. His food intake gradually decreased, resulting in a weight loss of 20 kg within 2 months. Evaluation of organic and other functional causes of dysphagia was performed, but no abnormalities were detected. The sensation of a lump in his throat, fear of swallowing, and anxiety were transformed into somatic symptoms. The patient was diagnosed with psychogenic dysphagia. After visual biofeedback by a physician who performed FEES, the patient resumed eating normally and increased his food intake. If routine tests do not reveal structural or functional causes of dysphagia, assessment of a psychogenic swallowing disorder should be considered. FEES can help in the diagnosis and management of psychogenic dysphagia.

Background: Ipsilateral pelvic elevation has been reported as a common compensatory movement during side-lying hip abduction. It has been reported that pelvic elevation inhibits sufficient contraction of gluteus medius. However, few studies have identified the effects of controlled pelvic elevation on the trunk and hip muscles. Objective: To examine the effects of controlled pelvic elevation using visual biofeedback on the muscle activity of the trunk and hip muscles. Design: Crossover study. Methods: Twelve healthy males performed side-lying hip abduction exercises with and without visual biofeedback for pelvic elevation. Electromyography (EMG) activities of the gluteus medius, quadratus lumborum, and multifidus were analyzed using a wireless EMG system while the ipsilateral pelvic elevation angle was measured using a motion sensor during side-lying hip abduction exercises. Results: EMG activities of the gluteus medius (p = .002), quadratus lumborum (p = .022), and multifidus (p = .020) were significantly increased and ipsilateral pelvic elevation was significantly decreased (p = .001) during side-lying hip abduction with visual biofeedback compared to without visual biofeedback. Conclusions: The results of this study suggest that the application of biofeedback for pelvic motion could improve the trunk and hip muscle activation pattern and decrease compensatory pelvic motion during side-lying hip abduction exercise.

Objective: The purpose of this study is to investigate the impact of visual biofeedback methods utilizing pressure sensors on the static balance of stroke patients. Design: Randomized crossover study. Methods: A total of 27 patients with hemiparesis participated in this study. The following three feedback conditions were considered: condition 1 (Knowledge of performance feedback), condition 2 (Knowledge of result feedback), and condition 3 (None feedback). A force plate was used to measure static balance. The total sway length, average sway velocity, x-axis excursion, and y-axis excursion of the center of pressure were measured. One-way repeated-measures analysis of variance was employed for comparisons of variables between each condition. The statistical significance level was set at α = 0.05 for all analyses. Results: There was a significant difference in the static balance results between each feedback condition (p<0.05). In the post-hoc results, it was confirmed that the static balance was significant in the order of knowledge of performance feedback, knowledge of result feedback, and none feedback. Conclusions: When comparing the three conditions, it was observed that knowledge of performance feedback showed the most improved effect on static balance ability. As further research progresses, that this approach could be used as an effective intervention method in clinical settings.

The purpose of this study was to investigate the effects of visual electromyography (EMG) biofeedback on the EMG activity of the lower trapezius (LT), serratus anterior (SA), and upper trapezius (UT) muscles, the LT/UT and SA/UT EMG activity ratios, and the scapular upward rotation angle during scapular posterior tilting exercise (SPTE). Twenty-four subjects with round-shoulder posture participated in this study. The EMG activities of the LT, SA, and UT were collected during SPTE both without and with visual EMG biofeedback. The scapular upward rotation angle was measured at the baseline, after SPTE without visual EMG biofeedback, and after SPTE with visual EMG biofeedback. The LT, SA, and UT EMG activities, and the LT/UT and SA/UT EMG activity ratios were analyzed by paired t-test. The scapular upward rotation angle was statistically analyzed using one-way repeated analysis of variance. If a significant difference was found, a Bonferroni correction was performed (p=.05/3=.017). The EMG activities of LT and SA significantly increased, and the EMG activity of UT significantly decreased during SPTE with visual EMG biofeedback compared to SPTE without visual EMG biofeedback (p<.05). In addition, the LT/UT and SA/UT EMG activity ratios significantly increased during SPTE with visual EMG biofeedback compared to SPTE without visual EMG biofeedback (p<.05). Significant increases were found in the scapular upward rotation angle after SPTE without and with visual EMG biofeedback compared to baseline (p<.017), and no significant differences were observed in the scapular upward rotation angle between SPTE without and with visual EMG biofeedback. In conclusion, SPTE using visual EMG biofeedback may be an effective method for increasing LT and SA activities while reducing UT activity.

PURPOSE: The purpose of this study was to investigate that the differences in sternocleidomastoid muscle activity, neck rotation angle, neck lateral bending angle and neck lateral bending onset time between conditions with and without visual biofeedback during neck rotation. METHODS: Ten male and four female adults with condition of forward head posture were recruited in this study. Subjects conducted to left and right maximal neck rotation under the conditions with and without visual biofeedback. During neck rotation, kinematic data of neck rotation, neck lateral bending movement, and electromyography activities of bilateral sternocleidomastold muscles were collected. Differences in dependent variables between conditions with and without visual biofeedback were analyzed using paired t-test. RESULTS: There were significant decreases in lateral bending angle, while lateral bending movement onset time was delayed significantly when applying visual biofeedback (p<.05). However, there were no significant differences in the activation of left and right sternocleidomastoid muscles and neck rotation angle between conditions with and without visual biofeedback (p>.05). CONCLUSION: These findings suggest that visual biofeedback may be effective for axial rotation of cervical spine during neck rotation in adults with forward head posture.

Background: This study aimed to investigate the effect of breathing training using visual biofeedback on respiratory muscle activity and function patients COVID-19 experienced patients. Design: Randomized Methods: Thirty people who experienced COVID-19 participated and randomly deployed 15 control groups that conducted standing breathing training and 15 visual biofeedback standing breathing training to train 3 times a week for 8 weeks. In order to evaluate the impact of each sound insulation breathing exercise on respiratory function, respiratory function before and after arbitration 12 times evaluated. Respiratory function were tested using the were used to measure lung capacity and muscle strength of respiratory muscles, and spirometer were used to evaluate intake volume and cough ability. Results: The results of this study showed a significant improvement(p<0.05) in muscle activity of the left latissimus dorsi of the VBG from pre- to post-training, but no improvement in muscle activity of the right latissimus dorsi. In addition, in the measure of inspiratory function, VBG showed significant improvements in inspiratory pressure in the within-group comparison(p<0.05). Conclusion: The results of the study showed no significant change in blocked breathing training using visual biofeedback. Therefore, breathing training through correction of surrounding factors will be effective breathing training for patients.

PURPOSE: The purpose of this study was to determine the comparison of visual and auditory biofeedback during sit-to-stand training in patients with stroke. METHODS: Thirty-five subjects with chronic stroke were divided into three groups: a visual feedback group (12 subjects), an auditory feedback group (12 subjects) and a control group (11 subjects). All Groups received neurodevelopmental treatment, and sit-to-stand training for 30minutes three times a week for four weeks. During the sit to stand training, the experimental groups received visual feedback and auditory feedback, whereas the control group performed sit-to-stand training without feedback. Five times sit-to-stand test (FTSST), motion analysis and postural sway during sit-to-stand were used to evaluate sit to stand performance ability. In addition, Berg balance scale (BBS) was performed for evaluation of balance function in participants. RESULTS: All groups showed significant increase on FTSST and BBS between pre- and post-intervention. The BBS scores in visual feedback group was significant increase than control group. The motion analysis and postural sway, more improvement was observed in the visual and auditory feedback groups compared with the control group. The only visual feedback group was a better performance of midline excursion during sit-to-stand than control group. CONCLUSION: These findings suggest that sit-to-stand training using a biofeedback may help to improve sit to stand performance and balance ability of stroke patients.