• 한국인터넷방송통신학회논문지
• /
• 제22권2호
• /
• pp.51-62
• /
• 2022

본 논문에서는 환자의 자세를 기반으로 행동을 예측하여, 의료진에 의해 입력된 개인의 병력 중심의 프로파일과 신체정보, 침상의 기본 정보를 모두 조합하여 침대에서의 낙상 위험을 예측하는 모델을 설계하고, 위험의 수준을 판단할 수 있는 알고리즘을 제시한다. 낙상 위험 예측은 크게 환자의 프로파일을 활용한 정성적 낙상 위험 노출도 평가와 실시간 낙상 위험 측정 단계로 구분된다. 정성적 낙상 위험 노출도는 의료진이 낙상 위험과 관련된 환자의 건강 상태를 점검하여 위험 노출도를 평가함으로써 위험 등급이 결정된다. 실시간 낙상 위험 측정 단계에서는 환자의 침대에서의 자세를 인식하고 환자의 정성적 위험등급 정보가 고려된 낙상 위험 측정을 위한 규칙 기반 정보를 추출한다. 인식된 환자 자세 정보와 정성적 위험평가 정보를 모두 조합하여 시그모이드 함수를 활용하여 최종 낙상 위험 수준을 예측한다. 본 연구에서 제시된 절차와 예측 모델은 입원 환자를 위한 낙상 사고 예방과 환자 안전을 위한 개인화 서비스에 크게 기여할 것으로 기대된다.

• 인터넷정보학회논문지
• /
• 제21권2호
• /
• pp.27-37
• /
• 2020

최근 국내외 빅데이터가 4차 산업혁명의 핵심기술로 급부상하고 있다. 또한, 4차 산업혁명의 발달과 더불어 드론에 대한 활용도와 수요가 계속 증가하고 있으며, 이에 관한 결과로 이제 드론은 일상생활과 다양한 산업 활동에 많이 활용되고 있다. 하지만 드론의 활용이 많아지면서 추락의 위험 또한 높아지고 있다. 드론은 비행 시 드론 내부 특성상의 간단한 구조로 인하여 작은 문제에도 쉽게 추락할 수 있는 위험요소를 항상 가지고 있다. 본 논문에서는 이러한 드론 추락 위험요소를 예측하고 추락을 방지하기 위하여 드론의 구동 모터와 일체형으로 ESC(Electronic Speed Control)를 부착하고 그 안에 가속도 센서를 장착해 진동 데이터를 실시간으로 수집 및 저장하고 그 데이터를 실시간으로 처리 및 모니터링 한다. 그리고 모니터링 상황에서 얻어진 빅데이터를 통한 데이터를 고속 푸리에 변환(Fast Fourier Transform,FFT) 알고리즘을 이용하여 수집된 빅데이터를 분석하여 드론 추락의 위험을 최소화하는 예측시스템을 제안하였다.

• 임상간호연구
• /
• 제26권2호
• /
• pp.265-273
• /
• 2020

Purpose: The purpose of this study was to test the predictive validity of the Fall Assessment Scale-Korean version (FAS-K) and to find the most appropriate cutoff score to screen high-risk fall groups in adult patients in general hospitals in Korea. Methods: We performed a prospective evaluation study in medical and surgical ward patients at two major general hospitals in Seoul. Data were collected from Nov. 1, 2018 to Feb. 28, 2019, nurses performed 651 observation series. The researcher measured the fall risk assessment score by applying FAS-K, MFS (Morse Fall Scale), and JHFRAT (Johns Hopkins Hospital Fall Risk Assessment tool) to the patients twice a week between 10 am and 12 noon. Data were analyzed using Pearson's corelation coefficients, and the sensitivity, specificity, predictive value, and the area under the curve (AUC) of the three tools. Results: The FAS-K was positively correlated with the MFS (r=.70, p<.001) and the JHFRAT (r=.82, p<.001). According to the receiver operating characteristics (ROC) curve analysis of the FAS-K, sensitivity, specificity, and positive and negative prediction values were 85.3%, 49.4%, 8.5%, and 98.4%, respectively, when the FAS-K score was 4. Therefore, the cut-off score of the FAS-K to identify groups with high fall risk was 4. Conclusion: The FAS-K is a valid tool for measuring fall risk in adult inpatients. In addition, the FAS-K score, 4, can be used to identify high-risk fall groups and know specific points in time to provide active interventions to prevent falls.

• PNF and Movement
• /
• 제22권1호
• /
• pp.43-54
• /
• 2024

Purpose: Falls among persons older than 65 years are a significant concern due to their frequency and severity. This study aimed to develop a vest-type embedded artificial intelligence (AI) system capable of detecting and predicting falls in various scenarios. Methods: In this study, we established and developed a vest-type embedded AI system to judge and predict falls in various directions and situations. To train the AI, we collected data using acceleration and gyroscope values from a six-axis sensor attached to the seventh cervical and the second sacral vertebrae of the user, considering accurate motion analysis of the human body. The model was constructed using a neural network-based AI prediction algorithm to anticipate the direction of falls using the collected pedestrian data. Results: We focused on developing a lightweight and efficient fall prediction model for integration into an embedded AI algorithm system, ensuring real-time network optimization. Our results showed that the accuracy of fall occurrence and direction prediction using the trained fall prediction model was 89.0% and 78.8%, respectively. Furthermore, the fall occurrence and direction prediction accuracy of the model quantized for embedded porting was 87.0 % and 75.5 %, respectively. Conclusion: The developed fall detection and prediction system, designed as a vest-type with an embedded AI algorithm, offers the potential to provide real-time feedback to pedestrians in clinical settings and proactively prepare for accidents.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권11호
• /
• pp.2987-3002
• /
• 2023

In 2015, the number of senior citizens aged 65 and over in Korea was 6,662,400, accounting for 13.1% of the total population. Along with these social phenomena, risk information related to the elderly is increasing every year. In particular, a fall accident caused by a fall can cause serious injury to an elderly person, so special attention is required. Therefore, in this paper, we implemented a system that monitors fall accidents and informs them in real time to minimize damage caused by falls. To this end, beacon-based indoor location positioning was performed and biometric information based on an integrated module was collected using various sensors. In other words, a multi-functional sensor integration module was designed based on Arduino to collect and monitor user's temperature, heart rate, and motion data in real time. Finally, through the analysis and prediction of measurement signals from the integrated module, damage from fall accidents can be reduced and rapid emergency treatment is possible. Through this, it is possible to reduce the damage caused by a fall accident, and rapid emergency treatment will be possible. In addition, it is expected to lead a new paradigm of safety systems through expansion and application to socially vulnerable groups.

• Physical Therapy Rehabilitation Science
• /
• 제7권2호
• /
• pp.54-60
• /
• 2018

Objective: Persons with chronic stroke fall more often than healthy elderly individuals. The Timed Up and Go test (TUG) is used as a fall prediction tool, but only provides a result for the total measurement time. This study aimed to determine the optimal cut-off values for each of the 6 components of the TUG. Design: Retrospective study. Methods: Thirty persons with chronic stroke participated in the study. TUG evaluation was performed using a wearable miniaturized inertial sensor. Sensitivity, specificity, and predictive values were calculated using the Receiver Operating Characteristic (ROC) curve analysis for the measured values in each section. Optimal values for fall risk classification were determined. Logistic regression analysis was used to investigate the risk of future falls based on TUG. Results: The cut-off values of the 6 sections of the TUG were determined, as follows: sit-to-stand >2.00 seconds (p<0.05), forward gait >4.68 seconds (p<0.05), mid-turn >3.82 seconds (p<0.05), return gait >4.81 seconds (p<0.05), end-turn >2.95 seconds (p<0.05), and stand-to-sit >2.13 seconds (p<0.05). The risk of falling increased by 2.278 times when the mid-turn value was >3.82 seconds (p<0.05). Conclusions: The risk of falls increased by 2.28 times when the value of the mid-turn interval exceeded 3.82 seconds. Therefore, when interpreting TUG results, the predictive accuracy for falls will be higher when the measurement time for each section is analyzed, together with the total time for TUG.

• 간호행정학회지
• /
• 제17권4호
• /
• pp.484-492
• /
• 2011

Purpose: The aims of study were; (1) to evaluate the validity and sensitivity of a fall-risk assessment tool, and (2) to establish continuous quality improvement (CQI) methods to monitor the effective use of the risk assessment tool. Methods: A retrospective case-control cohort design was used. Analysis was conducted for 90 admissions as cases and 3,716 as controls during the 2006 and 2007 calendar years was conducted. Fallers were identified from the hospital’s Accident Reporting System, and non-fallers were selected by randomized selection. Accuracy estimates, sensitivity analysis and logistic regression were used. Results: At the lower cutoff score of one, sensitivity, specificity, and positive and negative predictive values were 82.2%, 19.3%, 0.03%, and 96.9%, respectively. The area under the ROC was 0.60 implying poor prediction. Logistic regression analysis showed that five out of nine constitutional items; age, history of falls, gait problems, and confusion were significantly associated with falls. Based on these results, we suggested a tailored falls CQI process with specific indexes. Conclusion: The fall-risk assessment tool was found to need considerable reviews for its validity and usage problems in practice. It is also necessary to develop protocols for use and identify strategies that reflect changes in patient conditions during hospital stay.

• 재활복지공학회논문지
• /
• 제12권1호
• /
• pp.46-52
• /
• 2018

본 연구는 65세 이상 도시거주 노인 107명을 대상으로 일반적 특성, 만성질환 상태, 낙상 관련 의학적 변수, 균형 관련 자신감, 신체적 능력, 우울감을 평가하는 도구를 통해 낙상에 영향을 미치는 관련 주 요인을 찾고자 하였다. 또한 유의한 차이가 있는 변수들 간의 상관관계를 파악하며, 이 중 낙상을 유발하는 데 높은 영향력이 있는 변수를 도출하여 예측력을 알아보았다. 연구 결과, 낙상군에서 요실금, 발의 통증, 하지근력약화, 만성 질환수 및 복용 약물수 빈도수가 비낙상군에 비해 통계적으로 유의하게 높았다. 또한 ABC (Activities-specific Balance Confidence) 총점, BBS (Berg Balance Scale) 총점, SGDS (Short Geriatric Depression Scale) 총점, FRT(Functional Reach Test) 값에서 통계학적으로 유의한 차이가 있었다. 낙상에 영향을 주는 주요인은 ABC 총점으로 점수가 낮을수록 낙상 위험이 높아짐으로써 균형능력에 대한 자기 확신감이 낮을수록 낙상의 가능성이 높아지는 것으로 나타났으며, ABC, SDGS, BBS 척도가 결합하여 적용될 경우 낙상군과 비낙상군을 구분하는 예측력은 70.1%로 높게 나타났다.

• 한국전문물리치료학회지
• /
• 제21권2호
• /
• pp.37-47
• /
• 2014

Falls are common after stroke and most frequently related to loss of balance while walking. Consequently, preventing falls is one of the goals of acute, rehabilitative, and chronic stroke care. The purpose of this study was to investigate the incidence and risk factors of falls and to determine how well the Falls Efficacy Scale (FES), Timed Up and Go test (TUG), and Berg Balance Scale (BBS) could distinguish between fallers and non-fallers among stroke patients during inpatient rehabilitation. One hundred and fifteen participants with at least 3 months post-stroke and able to walk at least 3 m with or without a mono cane participated in this study. Fifty-four (47%) participants reported falling, and 15 (27.8%) had a recurrent fall. Logistic regression analysis for predicting falls showed that left hemiplegia [odds ratio (OR)=4.68] and fear of falling (OR=5.99) were strong risk factors for falls. Fallers performed worse than non-fallers on the FES, TUG, and BBS (p<.05, p<.01, respectively). In the receiver operator characteristic curve analysis, the TUG demonstrated the best discriminating ability among the three assessment tools. The cut-off score was 22 seconds on the TUG for discriminating fallers from non-fallers (sensitivity=88.9%, specificity=45.9%) and 27 seconds for discriminating recurrent fallers from single fallers and non-fallers (sensitivity=71.4%, specificity=40.2%). Results suggest that there is a need for providing fall prevention and injury minimization programs for stroke patients who record over 22 seconds on the TUG.

• The Journal of Asian Finance, Economics and Business
• /
• 제8권1호
• /
• pp.665-673
• /
• 2021

Default has become an extreme concern in the current world due to the financial crisis. The previous prediction of companies' bankruptcy exhibits evidence of decision assistance for financial and regulatory bodies. Notwithstanding numerous advanced approaches, this area of study is not outmoded and requires additional research. The purpose of this research is to find the best classifier to detect a company's default risk and bankruptcy. This study used secondary data from the Pakistan Stock Exchange (PSX) and it is time-series data to examine the impact on the determinants. This research examined several different classifiers as per their competence to properly categorize default and non-default Pakistani companies listed on the PSX. Additionally, PSX has remained consistent for some years in terms of growth and has provided benefits to its stockholders. This paper utilizes machine learning techniques to predict financial distress in companies listed on the PSX. Our results indicate that most multi-stage mixture of classifiers provided noteworthy developments over the individual classifiers. This means that firms will have to work on the financial variables such as liquidity and profitability to not fall into the category of liquidation. Moreover, Adaptive Boosting (Adaboost) provides a significant boost in the performance of each classifier.