• 대한물리치료과학회지
• /
• 제26권1호
• /
• pp.1-8
• /
• 2019

Background: PNF patterns are the basis of human motion and can be expected to improve joint motion and coordination. Combined physical training with CLX training and PNF patterns can help to improve balance and perform functional mobility in the lower limb. The purpose of this study is to confirm the effect of CLX training combined with PNF pattern on balance ability. Design: Randomized Controlled Trial. Methods: Total 16 persons participate in this study and were randomly divided in two groups the experimental group and control group. In the experimental group, exercise program with PNF pattern and CLX was performed total 24 times for 8 weeks. In the exercise program, the PNF pattern composed of D1F and D2F was applied with CLX in five positions. Single limb hop test, Y-balance test and Balance Error scoring system were performed to evaluate the balance ability according to the interventions. Results: In the single limb hop, the experimental group revealed a significant difference than a control group (p<0.05).The result of balance error scoring system, experimental group revealed significant differences between before and after training and revealed significant differences than a control group (p<0.05). In the Y-balance test, the experimental group revealed significant differences than a control group in both side. Conclusion: The results of this study showed that the CLX exercise in combination with the PNF pattern had a positive effect on enhancing the balance ability of the normal adult and performing the functional mobility of the lower limb.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.619-623
• /
• 1992

In pattern recognition, the conventional neural networks contain a large number of weights and require considerable training times and preprocessor to classify a transformed patterns. In this paper, we propose a constrained pattern recognition method which is insensitive to rotation of input pattern by various degrees and does not need any preprocessing. Because these neural networks can not be trained by the conventional training algorithm such as error back propagation, a novel training algorithm is suggested. As such a system is useful in problem related to calssify overse side and reverse side of 500 won coin. As an illustrative example, identification problem of overse and reverse side of 500 won coin is shown.

• 대한물리의학회지
• /
• 제2권2호
• /
• pp.173-182
• /
• 2007

Purpose : The purpose of this study was carried out to review for the importance of breathing pattern training for the spinal stabilization. Methods : This is a literature study with books and thesis. Results : Breathing with normal respiratory mechanics has a potent role in neuro-musculo-skeletal system. The evaluation of respiratory mechanics should be a routine part of every physical examination. And respiratory mechanics must be intact for both normal posture and spinal stabilization to be possible. Conclusion : The spinal stabilization exercise with the breathing pattern training is more efficient therapeutic exercise program for the patient with neuro-musculo-skeletal system disorder.

• PNF and Movement
• /
• 제17권1호
• /
• pp.119-127
• /
• 2019

Purpose: The purpose of this study is to show the effect of elastic band on balance and functional ability in chronic stroke patients living in community. Methods: The subjects who participated in the study were 9 patients with chronic stroke. One of them gave up during the study, finally 8 patients performed. The intervention was conducted once a week for 10 weeks. In this study functional reach test (FRT), timed up and go test (TUG), Tinetti performance oriented mobility assessment (Tinetti-POMA) were measured for balance. The coordination training of arms and legs using the elastic band was performed in three positions as supine, side lying, sitting. One arm performed flexion-adduction- external rotation with elbow flexion pattern and the opposite side(diagonal) leg was performed flexion-adduction-external rotation with knee flexion pattern, the other arm's pattern was extension-abduction-internal rotation with elbow extension and the opposite side (diagonal) leg was in extension-abduction-internal rotation with knee extension pattern. The training was performed in each position for 15 minutes in per position. The participants had a five minute break after each training. Results: The results are as follows. FRT and Tinetti-POMA showed significant increase statistically in each position. The TUG showed significant decrease statistically in each position. Conclusion: Even though the coordination training with elastic band had performed once a week, it showed positive effects on balance in chronic stroke patients. Therefore, if we can suggest the appropriate frequencies of coordination training of arms and legs using the elastic band, it can be a method to improve daily life and life quality to patients with chronic stroke.

• 대한물리의학회지
• /
• 제18권1호
• /
• pp.59-66
• /
• 2023

PURPOSE: This study examined the foot pressure distribution using the sprinter and skater patterns of coordinative locomotor training. METHODS: Thirty healthy adults, comprising 11 men and 19 women, participated in the study. All the participants performed patterns in sprinter pattern conditions 1-3 and skater pattern conditions 1-3, and were measured using a pedoscan to determine the applied foot pressure distribution. RESULTS: The participants significantly differed in the big toe during the sprinter pattern. As a result of the post hoc test, opposite and opposite sprinters showed a significant difference from the same sprinters (same sprinter; 21.33 ± 5.92, opposite sprinter; 23.54 ± 5.41, and reopposite sprinter; 24.14 ± 6.46). There was a significant difference in the lateral side during the skater pattern. As a result of the post hoc test, reopposite and same skaters showed a significant difference from opposite skaters (same skater; 49.88 ± 5.75, opposite skater; 48.78 ± 5.64, and reopposite skater; 51.15 ± 5.37). CONCLUSION: The foot pressure was distributed toward the hallux and fifth toe according to the sprinter and skater patterns of coordinative locomotor training.

• 대한인간공학회:학술대회논문집
• /
• 대한인간공학회 1996년도 추계학술대회논문집
• /
• pp.197-205
• /
• 1996

Unification of neural network with a hierarchical pattern recognition is presented for recognizing large set of objects. A two-step identification procedure is developed for pattern recognition: coarse and fine identification. The coarse identification is designed for finding a class of object while the fine identification procedure is to identify a specific object. During the training phase a course neural network is trained for clustering larger set of reference objects into a number of groups. For training a fine neural network, expert neural network is also trained to identify a specific object within a group. The presented idea can be interpreted as two step identification. Experimental results are given to verify the proposed methodology.

• 한국구조물진단유지관리공학회 논문집
• /
• 제6권4호
• /
• pp.169-179
• /
• 2002

This paper presents an efficient algorithm for the estimation of damage location and severity in structure using Probabilistic Neural Network (PNN). Artificial neural network has been being used for damage assessment by many researchers, but there are still some barriers that must be overcome to improve its accuracy and efficiency. The major problems with the conventional neural network are the necessity of many training data for neural network learning and ambiguity in the relation of neural network architecture with convergence of solution. In this paper, PNN is used as a pattern classifier to overcome those problems in the conventional neural network. The basic idea of damage assessment algorithm proposed in this paper is that modal characteristics from a damaged structure are compared with the training patterns which represent the damage in specific element to determine how close it is to training patterns in terms of the probability from PNN. The training pattern that gives a maximum probability implies that the element used in producing the training pattern is considered as a damaged one. The proposed damage assessment algorithm using PNN is applied to a 2-span continuous beam model structure to verify the algorithm.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제30권3_4호
• /
• pp.339-346
• /
• 2003

본 논문에서는 지역적 특성을 가지는 작은 인식기(마이크로 인식기)의 모음으로 인식기를 구현하는 다중 인식기 시스템을 제안한다. 각 학습패턴에서 k개의 이웃한 학습패턴을 추출해서 학습한 인식기를 마이크로인식기라고 한다. 각 학습패턴에는 한개 이상의 마이크로 인식기를 부여한다. 본 논문에서는 선형 커널을 사용한 SVM과 RBF 커널을 사용한 SVM등 두 가지 형태의 마이크로 인식기를 사용한다. 테스트 패턴이 인가되면 테스트패턴 주변의 마이크로인식기들 중에서 성능이 가장 좋은 것 하나를 선택한 후 선택된 인식기로 최종 클래스를 결정한다. 테스트패턴 주변에 있는 학습패턴들을 인식한 결과를 성능 측정 척도로 사용한다. Elena 데이터 베이스를 사용하여 기존의 단일 인식기, 다중 인식기 결합, 다중 인식기 선택 방법들과 인식률을 비교한 결과 제안된 방법이 우수함을 알 수 있다.

• 전자공학회논문지B
• /
• 제31B권12호
• /
• pp.108-115
• /
• 1994

Most pattern classifiers have been designed based on the ML (Maximum Likelihood) training algorithm which is simple and relatively powerful. The ML training is an efficient algorithm to individually estimate the model parameters of each class under the assumption that all class models in a classifier are statistically independent. That assumption, however, is not valid in many real situations, which degrades the performance of the classifier. In this paper, we propose a minimum-error-rate training algorithm based on the MAP (Maximum a Posteriori) approach. The algorithm regards the normalized outputs of the classifier as estimates of the a posteriori probability, and tries to maximize those estimates. According to Bayes decision theory, the proposed algorithm satisfies the condition of minimum-error-rate classificatin. We apply this algorithm to NPM (Neural Prediction Model) for speech recognition, and derive new disrminative training algorithms. Experimental results on ten Korean digits recognition have shown the reduction of 37.5% of the number of recognition errors.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.2037-2040
• /
• 2004

In this paper we introduce a new prototype based classifiers for overlapping data, where training pattern can be overlap on the feature space. The proposed classifier is based on the prototype from neural network classifier (NNC)[1] for overlap data. The method automatically chooses the initial center and two radiuses for each class. The center is used as a mean representative of training data for each class. The unclassified pattern is classified by measure distance from the class center. If the distance is in the lower (shorter radius) the unknown pattern has the high percentage of being in this class. If the distance is between the lower and upper (further radius), the pattern has the probability of being in this class or others. But if the distance is outside the upper, the pattern is not in this class. We borrow the words upper and lower from the rough set to represent the region of certainty [3]. The training algorithm to find number of cluster and their parameters (center, lower, upper) is presented. The clustering result is tested using patterns from Thai handwritten letter and the clustering result is very similar to human eyes clustering.