• Genomics & Informatics
• /
• 제16권4호
• /
• pp.31.1-31.7
• /
• 2018

The prevalence of metabolic syndrome (MS) in the nonobese population is not low. However, the identification and risk mitigation of MS are not easy in this population. We aimed to develop an MS prediction model using genetic and clinical factors of nonobese Koreans through machine learning methods. A prediction model for MS was designed for a nonobese population using clinical and genetic polymorphism information with five machine learning algorithms, including naïve Bayes classification (NB). The analysis was performed in two stages (training and test sets). Model A was designed with only clinical information (age, sex, body mass index, smoking status, alcohol consumption status, and exercise status), and for model B, genetic information (for 10 polymorphisms) was added to model A. Of the 7,502 nonobese participants, 647 (8.6%) had MS. In the test set analysis, for the maximum sensitivity criterion, NB showed the highest sensitivity: 0.38 for model A and 0.42 for model B. The specificity of NB was 0.79 for model A and 0.80 for model B. In a comparison of the performances of models A and B by NB, model B (area under the receiver operating characteristic curve [AUC] = 0.69, clinical and genetic information input) showed better performance than model A (AUC = 0.65, clinical information only input). We designed a prediction model for MS in a nonobese population using clinical and genetic information. With this model, we might convince nonobese MS individuals to undergo health checks and adopt behaviors associated with a preventive lifestyle.

• 문화기술의 융합
• /
• 제9권3호
• /
• pp.737-744
• /
• 2023

근감소증은 노인들 사이에서 점점 더 흔하게 발생하고 있어, 최근 주목을 받고 있는 질병이다. 근감소증의 원인은 매우 다양하게 나타나지만, 노화, 식습관, 운동 부족등이 주요한 원인들 중 하나이다. 근감소증은 원인이 다양한 만큼 예방 및 치료에 전략을 개발하는 것이 중요하다. 하지만 요인이 다양한 만큼 사람이 근감소증을 정확하게 예측하기는 어렵다. 여기서 기계학습을 이용해 근감소증 예측의 정확도와 편의를 크게 높일 수 있다. 그러나 생활습관과 생체 데이터의 양은 방대한 만큼, 전처리 없이 데이터를 쓰기에는 시간복잡도와 정확성 측면에서 부적절할 수 있다. 본 논문에서는 근감소증과 그 원인에 대한 최신 문헌을 검토하고, 그에 맞게 기계학습 기만 근감소증 예측에 활용할 데이터를 전처리하는데 초점을 맞춘다.

• 여성건강간호학회지
• /
• 제7권2호
• /
• pp.169-187
• /
• 2001

This study was performed to probe the effect of exercise program on muscle strength, endurance, flexibility, pain, disability level and life satisfaction in female teachers of elementary school who complain of low back pain. For this study, 44 female teachers aged 30-50 years with mechanical low back pain of 6 months' duration, who had the structural normalities in the lumbar spine, were recruited from April 1 to July 10 1999. Twenty three out of them were assigned to the experimental group and twenty one to the control group. The exercise program consisted of education on right postures, the etiology and diagnosis of low back pain, and exercise intervention such as muscle relaxation, elongation and strengthening. With 8 weeks program, the subjects received two sessions of education and six sessions of group exercise in the 1st week, while three sessions of group exercise and four sessions of individual exercise weekly and two sessions of education during the later 7 weeks. The muscle strength and endurance were measured by Cybex 770, the flexibility by flexibility measurement machine, the intensity of pain by Visual Analogue Scale (VAS), the level of disability by Oswestry low back pain disability scale, depression by Beck depression inventory (BDI), and life satisfaction by Life satisfaction index-Z. Study measurements were taken before and after 8 weeks exercise program. Data were analyzed using paired t-test, t-test, and ANCOVA. The results were as follows ; 1. The flexors and extensors peak torque and flexors peak torque per body weight of experimental group were significantly increased at test velocities $30^{\circ}$/sec, $60^{\circ}$/sec compared with those of control group. There was no significant difference in extensors peak torque per flexors peak torque at $30^{\circ}/sec$, $60^{\circ}/sec$ between experimental and control group. 2. The flexors and extensors total work and flexors total work per body weight of experimental group were significantly increased at $120^{\circ}/sec$, compared with those of control group. 3.The flexibility of lumbar spine in experimental group was significantly increased compared with that of control group. The pains in anterior, posterior, left lateral and right lateral bending and in rotation of experimental group were significantly increased compared with those of control group. 4. The Oswestry disability scores of experimental and control group were significantly decreased, and there was no difference in the Oswestry disability score change between experimental and control group. 5. The scores of BDI of experimental group were significantly decreased compared with those of control group. Life satisfaction index-Z scores of experimental group were not changed, but those of control group were significantly decreased. There was no difference in the score change of Life satisfaction index-Z between experimental and control group. 6. ANCOVA analysis for the data variables of inhomogeneous baseline represented that there was no significant difference in extensors peak torque and extensors total work at $120^{\circ}/sec$ and extensor total work per body weight at $120^{\circ}/sec$ change between experimental group and control group. These findings indicate that the exercise program could be effective in increasing the muscle strength, endurance, flexibility and decreasing pain, improving depression in female teachers of elementary school with chronic low back pain. It is suggested that the exercise program could be an essential factor for the effective nursing intervention to the patients suffered from chronic low back pain.

• 대한기계학회논문집 C: 기술과 교육
• /
• 제1권2호
• /
• pp.193-198
• /
• 2013

승마 시뮬레이터는 승마동작을 모사하여 실내에서 신체활동을 수행하기 위한 운동기구이다. 본 연구는 승마 시뮬레이터를 이용한 운동 시 소모되는 에너지 추정에 관한 연구이다. 실험은 총 9개의 동작으로 구분된 승마 동작에 대해 운동부하를 1분 단위로 변화시켜 총 25분 동안 운동을 수행하도록 설계하였으며, 호흡가스측정을 통해 에너지소모량을 수집하였다. 그 결과 신체조건에 따라 각 동작별 에너지소모량에 유의한 차이가 있음을 확인하였다. 이를 통해 사용자의 생체정보 측정 없이 신체정보만을 활용한 승마운동의 에너지소모량 추정방법을 개발하였다. 연구결과는 승마 시뮬레이터를 이용한 운동시 동작의 변화와 주행시간, 성별, 연령을 이용하여 실시간 에너지소모량의 계산이 가능해질 것이다.

• Physical Therapy Rehabilitation Science
• /
• 제3권2호
• /
• pp.93-100
• /
• 2014

Objective: The purpose of this study was to determine the effect of the abdominal bracing (AB) and abdominal bracing combined with ankle dorsiflexion (ABDF) on abdominal muscle thickness and strength in patients with chronic low back pain (LBP). Design: Two group pretest posttest design. Methods: Sixteen subjects were divided randomly into two group: ABDF group (n=8), and alone AB group (n=8). The ABDF group practiced AB exercise with additional ankle dorsiflexion. AB group practiced only AB exercises. Subjects in both groups received ABDF exercise and AB exercise for 40 min per day, three days per week during a period of three weeks, respectively. All the subjects were evaluated for abdominal muscle thickness and strength before and after intervention using ultrasonography and MedX machine. Results: The external oblique (EO), internal oblique, transverse abdominis (TrA) muscle thickness and the strength produced at $48^{\circ}$, $60^{\circ}$, $72^{\circ}$ showed a significant increase in the ABDF group after intervention, with a more significant improvement in EO and TrA muscle thickness in the ABDF group compared with the AB group (p<0.05). Also, the strength at $48^{\circ}$ strength showed a significant improvement in the ABDF group than the AB group (p<0.05). Conclusions: The study results showed that abdominal muscle contraction exercises with AD in patients with LBP had an influence on abdominal muscle thickness and strength. Therefore, these findings suggest that ABDF may be useful approach for enhancement of abdominal muscle thickness and strength in patients with chronic low back pain.

• 재활복지공학회논문지
• /
• 제12권1호
• /
• pp.1-9
• /
• 2018

본 연구는 승마가 노인의 골밀도에 어떠한 영향을 미치는지 알아보기 위하여 실시되었다. 본 연구에 참여한 대상자는 총 31명의 정상 노인으로 승마운동군 15명, 승마기구운동군 16명을 무작위로 나누어 연구에 참여하였다. 승마운동 프로그램은 12주 간 주 2회 25분을 실시하였다. 사전 검사는 승마운동프로그램 시작 전에 실시하고, 사후 검사는 운동 6주 후, 운동 12주 후에 연구 전 검사와 동일한 방법으로 시행하였다. 두 그룹의 운동 전과 운동 6주 후, 운동 12주 후의 측정변인에 대한 변화를 알아보기 위해 이원배치 반복측정 분산분석을 이용하였다. 집단별 골밀도의 변화를 비교한 결과, 두 그룹 모두 허리뼈 L3, L4, 넙다리 돌기에서 시간에 따라 유의한 차이가 있었다. 결론적으로, 승마운동을 통한 말의 3차원적인 움직임을 통해 노인들에게 정상적인 움직임을 유도하여 골밀도에 긍정적인 영향을 미칠 수 있다. 노인 대상자의 건강 상태, 시간적 경제적 여건이나 선호도를 고려하여 적절하게 승마운동프로그램을 활용한다면 노인의 삶의 질 향상에 도움이 될 수 있을 것이다.

• 여성건강간호학회지
• /
• 제6권1호
• /
• pp.117-128
• /
• 2000

The purpose of this study was to examine the effect of low back pain relieving program on back muscle strength, intensity of pain, low back disability level in elementary school teachers who have low back pain. subjects were elementary school women teachers who worked at eight elementary school located in Seoul. Intended subjects size were seventy consist of thirty-four experimental group(three schools) and thirty-six control group(five schools), but actual subjects size was forty-four. Among the forty-four patients subjects, twenty-three were experimental group receiving health education about right postures, etiologies of low back pain, diagnosis of low back pain and exercise program composed of muscle strengthening exercise, stretching exercises and twenty-one were control group. During the 8 weeks program, the subjects were received two times education and six times group exercise practices in 1st week and three times per week group exercise practices, two times education in other 7 weeks. This study as carried out from April 1, 1999 to June 30, 1999. Back muscle strength was measured by back muscle strength measuring machine and the intensity of pain were measured by the Visual Analogue Scale(VAS), and level of disability was measured by Oswestry low back pain disability scale. Study measurements were taken before and after 8 week exercise program. Data were analyzed using paired t-test, and ANCOVA. The results were summarized as follows. 1. After low back relieving program, back muscle strength was increased significantly(p=0.000) and there was significant difference in back muscle strength change between experimental group and control group(p=0.002). 2. After low back pain relieving program, pain on anterior bending, pain on posterior bending were decreased significantly than measurements before the program(p=0.000 p=0.000) and there was significant difference in pain on anterior bending and posterior bending change between experimental group and control group(p=0.000, p=0.000). 3. After low back pain relieving program, Oswestry disability scale scores were decreased significantly(p=0.000, p=0.000) but there was no significant difference in Oswestry disability score change between experimental group and control group.

• PNF and Movement
• /
• 제15권1호
• /
• pp.105-113
• /
• 2017

Purpose: The purpose of this case study was to examine the effects of an intervention based on the concept of proprioceptive neuromuscular facilitation (PNF) on the weight-supporting ability, fear of falling, and stair-walking ability of stroke patients. Methods: One pretest, three intervention sessions, and one posttest were conducted. In the tests, weight-supporting ability, manual muscle strength, the fear of falling, and the time to go up and down a 'ㄱ'-shaped stair-walking machine were measured and compared. The intervention was implemented for 30 minutes per day for three days in combination with general rehabilitation. The intervention was planned and implemented based on the concept of PNF, and the goal was for the subjects, after their consent, to walk backward down the stairs. Results: After the intervention, functional improvement was seen in weight-supporting ability and the fear of falling felt when walking down the stairs. Whereas the subjects could not perform the task of walking down the stairs on the 'ㄱ'-shaped stair-walking machine in the pretest, they could perform the task in the posttest, and their stair-climbing speed was greater than before the intervention. Conclusion: This study verified that a PNF-based intervention can improve stroke patients' weight-supporting ability and stair-walking ability. Therefore, this intervention can be clinically applied to stroke patients.

• 한국전문물리치료학회지
• /
• 제28권3호
• /
• pp.177-185
• /
• 2021

Background: Muscle undergoes change continuously with aging. Sarcopenia, in which muscle mass decrease with aging, is associated with various diseases, the risk of falling, and the deterioration of quality of life. Obesity and sarcopenia also have a synergy effect on the disease of the older adults. Objects: This study examined the risk factors for sarcopenia, sarcopenic obesity, and sarcopenia without obesity and developed prediction models. Methods: This machine-learning study used the 2008-2011 Korea National Health and Nutrition Examination Surveys in the analysis. After data curation, 5,563 older participants were selected, of whom 1,169 had sarcopenia, 538 had sarcopenic obesity, and 631 had sarcopenia without obesity; the remaining 4,394 were normal. Decision tree and random forest models were used to identify risk factors. Results: The risk factors for sarcopenia chosen by both methods were body mass index (BMI) and duration of moderate physical activity; those for sarcopenic obesity were sex, BMI, and duration of moderate physical activity; and those for sarcopenia without obesity were BMI and sex. The areas under the receiver operating characteristic curves of all prediction models exceeded 0.75. BMI could predict sarcopenia-related disease. Conclusion: Risk factors for sarcopenia-related diseases should be identified and programs for sarcopenia-related disease prevention should be developed. Data-mining research using population data should be conducted to enhance the effectiveness of early treatment for people with sarcopenia-related diseases through predictive models.

• 지능정보연구
• /
• 제22권4호
• /
• pp.177-192
• /
• 2016

기계학습(Machine Learning)은 인공 지능의 한 분야로, 데이터를 이용하여 기계를 학습시켜 기계 스스로가 데이터 분석 및 예측을 하게 만드는 것과 관련한 컴퓨터 과학의 한 영역을 일컫는다. 그중에서 SVM(Support Vector Machines)은 주로 분류와 회귀 분석을 목적으로 사용되는 모델이다. 어느 두 집단에 속한 데이터들에 대한 정보를 얻었을 때, SVM 모델은 주어진 데이터 집합을 바탕으로 하여 새로운 데이터가 어느 집단에 속할지를 판단해준다. 최근 들어서 많은 금융전문가는 기계학습과 막대한 데이터가 존재하는 금융 분야와의 접목 가능성을 보며 기계학습에 집중하고 있다. 그러면서 각 금융사는 고도화된 알고리즘과 빅데이터를 통해 여러 금융업무 수행이 가능한 로봇(Robot)과 투자전문가(Advisor)의 합성어인 로보어드바이저(Robo-Advisor) 서비스를 발 빠르게 제공하기 시작했다. 따라서 현재의 금융 동향을 고려하여 본 연구에서는 기계학습 방법의 하나인 SVM을 활용하여 매매성과를 올리는 방법에 대해 제안하고자 한다. SVM을 통한 예측대상은 한국형 변동성지수인 VKOSPI이다. VKOSPI는 금융파생상품의 한 종류인 옵션의 가격에 영향을 미친다. VKOSPI는 흔히 말하는 변동성과 같고 VKOSPI 값은 옵션의 종류와 관계없이 옵션 가격과 정비례하는 특성이 있다. 그러므로 VKOSPI의 정확한 예측은 옵션 매매에서의 수익을 낼 수 있는 중요한 요소 중 하나이다. 지금까지 기계학습을 기반으로 한 VKOSPI의 예측을 다룬 연구는 없었다. 본 연구에서는 SVM을 통해 일 중의 VKOSPI를 예측하였고, 예측 내용을 바탕으로 옵션 매매에 대한 적용 가능 여부를 실험하였으며 실제로 향상된 매매 성과가 나타남을 증명하였다.