Optimization of a Pre-impact Fall Detection Algorithm and Development of Hip Protection Airbag System (S & M 1631)

Ahn, Soonjae,Choi, Dagyeong,Kim, Jongman,Kim, Seongjung,Jeong, Youngjae,Jo, Min,Kim, Youngho SCIENTIFIC PUBLISHING DIVISION MYU 2018 Sensors and materials Vol.30 No.8

Optimization of a Pre-impact Fall Detection Algorithm and Development of Hip Protection Airbag System

Ahn, Soonjae,Choi, Dagyeong,Kim, Jongman,Kim, Seongjung,Jeong, Youngjae,Jo, Min,Kim, Youngho MYU K.K. 2018 Sensors and materials Vol.30 No.8

<P>In this study, a pre-impact fall detection algorithm using a custom-made inertial sensor was optimized, and a spring-trigger airbag system was developed for preventing injuries from falls. Four different types of simulated falls were performed by 20 healthy volunteers (age 23.4 +/- 4.4 years), and six different daily activities were tested in 14 elderly subjects (age 71.8 +/- 4.0 years). An inertial sensor unit was used to measure acceleration, angular velocity, and vertical angle during all activities. Thresholds of 0.9 g acceleration, 47.3 degrees/s angular velocity, and 24.7 degrees vertical angle were determined on the basis of optimizing lead time and accuracy in pre-impact fall detection. A belt-type airbag system consisted of a polyurethane inner skin, an artificial leather outer shell, and a spring-trigger inflator. To evaluate the accuracy of the airbag system, 10 healthy adult males (age 28.5 +/- 2.7 years) wore the system and performed three sets of simulated falls. Fall detection was achieved 401.9 +/- 46.9 ms before impact on average, and the airbag inflated without fail during the falls (100% sensitivity). In all daily activities, no airbag inflation occurred (100% specificity).</P>

Asymmetrical change in the pelvis and the spine during cross-legged sitting postures

Soonjae Ahn,Seunghyeon Kim,강선영,전혜선,김영호 대한기계학회 2013 JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY Vol.27 No.11

People spend much time with cross-legged sitting by increasing the time spent in the chair. Prolonged cross-legged sitting postures might cause spinal disorders and trunk asymmetry problems. We investigated the biomechanical effects of cross-legged sitting postures on the pelvic and spinal regions. 3D-motion analyses were performed on 26 healthy subjects. Data were collected while the subjects sat in four different postures: the control posture (without crossed legs) of sitting upright on the chair; the knee-on-knee (KoK) posture (the right knee on the left knee), the ankle-on-knee (AoK) posture (the right ankle on the left knee) and the ankle-on-ankle (AoA) posture (the right ankle on the left ankle). EMG signals were also collected for abdominal muscles. Cross-legged sitting postures showed significantly greater kyphotic curves in the lumbar and the thoracic spines and pelvic posterior tilting, when compared with the control posture. The pressure on the right buttock area was greater in the AoK posture than in the control posture, and the pressure on the left buttock area in the KoK posture was greater than in the control posture. Cross-legged sitting postures showed significantly greater external oblique muscle activation when compared with the control posture.

전이학습을 활용한 소규모 비정형 정책데이터 감성분석 모델

안순재(Soonjae Ahn),유승의(Seungui Ryu),홍순구(Soongoo Hong) 한국데이터정보과학회 2020 한국데이터정보과학회지 Vol.31 No.2

최근 빅데이터 기술의 발전에도 불구하고 정책분야에서는 텍스트 등 비정형 데이터의 부족으로 감성분석 연구에 제한이 많았다. 이에 본 연구에서는 전이학습을 기반으로 소규모 비정형 정책 데이터를 활용한 감성분석 모델을 제안한다. 이를 위해 네이버 영화리뷰 20만 건의 댓글로 CNN 모델을 생성하고 지역 리뷰 1만 건의 댓글을 이용하여 전이학습을 수행하였다. 분석결과 본 연구에서 제안한 전이학습 모델은 소규모 데이터만으로 생성된 모델보다 약 10%의 정확도 향상과 1epoch당 1000ms의 학습지간 단축을 보였다. 본 연구의 공헌도로 학술적으로는 한글 텍스트 감성분석에 전이학습을 처음으로 적용하여 향후 소규모 데이터의 감성분석 연구에 활용할 수 있는 이론적 기반을 제공하였다는 점이다. 실무적으로는 데이터가 부족하여 시도하기 어려웠던 정책 분야의 감성분석을 통해 정부사업의 성공여부를 판별할 수 있는 기초자료로 활용할 수 있다. Despite the recent development in big data technologies, the research on sentiment analysis is still facing many limitations due to the lack of unstructured data, including texts, in the policy field. Thus, this study proposes a sentiment analysis model for small-scale unstructured policy data based on transfer learning. As a result, the proposed transfer learning model achieved about 10% better accuracy and a shorter training time of 1000 ms per epoch than the model generated only by using small-scale data. As an academic contribution, this study, which is the first application of transfer learning to Korean text sentiment analysis, provides a theoretical basis for future research on sentiment analysis using small-scale data. For practicality, this study can serve as basic data in determining the success or failure of government projects through sentiment analysis in the policy field, which was difficult to determine previously given the lack of data. the detailed feature of sea/land breeze at each site is closely associated with the local shape of coastline.

Impact Attenuation of the Soft Pads and the Wearable Airbag for the Hip Protection in the Elderly

정영재,Soonjae Ahn,김종만,김성정,구범모,Jeseong Ryu,김영호 한국정밀공학회 2019 International Journal of Precision Engineering and Vol.20 No.2

Hip fractures caused by the falls in the elderly can lead to serious injuries that diminish their quality of life. The soft pads and wearable airbag have been developed and commercialized to prevent hip fractures of the elderly. In this study, the hip impact simulator was developed and the impact forces of six types soft pads and the hip protection airbag were measured by using the hip impact simulator. In addition, the distribution of the impact force and maximum contact area were also measured by attaching the pressure sensor on the surrogate pelvis of the hip impact simulator. Experiments were performed depending on the conditions protecting the surrogate pelvis (unpadded, soft pad, airbag) and the fall from 20 to 40 cm heights were repeated three times depending on each condition. The hip protection airbag decreased the mean pressure applied on the hip joint by decreasing a peak impact force according to the increased maximum contact area compared to the unpadded and soft pad conditions at all fall heights. Therefore, it can be concluded that the hip protection airbag can prevent the hip fracture by effectively attenuating impact force during falls in the elderly.

국내 모래놀이치료 연구동향(2009-2018)

안순자(Ahn, Soonja),장미경(Jang, Mikyung),이여름(Lee, Yeoreum),이세화(Lee, Sewha) 한국임상모래놀이치료학회 2020 상징과 모래놀이치료 Vol.11 No.1

The aim of this study was to investigate research trends in sandplay therapy in South Korea. Thus we analyzed 205 Korean research papers published between 2009 and 2018. We grouped our results into five categories using the research trend analysis method. Those categories were research topic, subject, method, and characteristics of sandplay therapy. The results reveal the current status and limitations of sandplay research. These also suggest future research directions for sandplay therapists and researchers.

SNS 사용자들의 사회적 라포 현상 연구

안창민 ( Changmin Ahn ),권순재 ( Soonjae Kwon ),정현희 ( Hyeonhee Jeong ) 한국지식경영학회 2018 지식경영연구 Vol.19 No.1

While there are lots of studies on examining the effects of social rapport in many research areas, however, there is a little work in examining the effect of the social rapport in social network service (SNS) contexts. Thus, this study attempts to examine the effect of social rapport in SNS settings. To address the research questions, this study has presented its hypotheses and conducted three experimental approaches by collecting 180 data from student subjects who have prior experiences on using SNSs to verify the hypotheses. This study has examined three experiments the effects of characteristics of Facebook(i.e. the number of mutual friends, the number of post likes, and the post personalities) on the social rapport and user responses. This study has conducted two-way ANOVA to verity its proposed research hypotheses. Based on three experiments, this study found that both the effects of the number of post likes and the number of post likes on the social rapport were not significant. Based upon empirical findings, this study has demonstrated how the effects of social rapports in SNSs were different from those of previous studies, and brought more attentions to the relevant literature.

Influence of the Melt Index of Low Density Polyethylene on the Properties in the Blends with Linear Low Density Polyethylenes

Cho, Kyucheol,Ahn, Tae Kwang,Park, I,Lee, Byung H,Choe, Soonja 한국공업화학회 1997 Journal of Industrial and Engineering Chemistry Vol.3 No.2

Melt flow index (MI) dependency of low-density polyethylene (LDPE) on the thermal, mechanical and rheological properties in the blends with linear low density polyethylene (1-octene comonomer) (LLDPE) has been investigated by means of differential scanning calorimeter, dynamic mechanical thermal analyzer, capillary rheometer, and rheotons. Two discrete melting and crystallizing peaks were observed regardless of the MI of LDPE in the crystalline phase. Mechanical properties of the blends were fairly dependent on the MI of LDPE. Complex shear viscosities of the blends were dictated by the LLDPE indicating that severe processing conditions may be necessary. Melt viscosity analyzed in terms of the Cole-Cole plot and the log additive rule resulted that the miscibility was improved at high shear rate and that the MI of LDPE does not affect the miscibility. However synergistic effect in the melt strength was observed for the entire blends systems used in this study. In particular, the blends containing 40∼60% LDPE are suggested to design for highly improved melt strength, processibility, and mechanical properties.

Knee Acoustic Emission Characteristics of the Healthy and the Patients with Osteoarthritis Using Piezoelectric Sensor

Choi, Dagyeong,Ahn, Soonjae,Ryu, Jeseong,Nagao, Mitsuo,Kim, Youngho MYU K.K. 2018 Sensors and materials Vol.30 No.8

<P>Osteoarthritis (OA) is one of the most common causes of disability in elderly individuals. Although X-rays, MRIs, and arthroscopy are widely used to examine OA, they do not provide dynamic information on activity-based joint integrity. In this study, acoustic emission (AE) in healthy individuals and OA patients was determined using piezoelectric sensors. Twenty normal subjects and fourteen OA patients were asked to perform weight-bearing (stand-to-sit) and non-weight-bearing (extension to flexion) exercises for 4 s each. Sensors were attached to the medial and lateral epicondyles of the tibia and the front of the patella. An AE event was defined as the threshold crossing based on AE signals. The results indicated that OA and healthy groups exhibited signals at a frequency range corresponding to approximately 100 Hz-10 kHz and less than 1 kHz, respectively. The OA group exhibited an amplitude of AE signals and a number of AE events that exceeded those of the healthy group (p < 0.05). AE signals from lesions exhibited a higher dB value and a larger number of AE events than those at other locations. However, most patients exhibited significant increases in AE signal characteristics at the front of the patella. The results of this study can be helpful in the early diagnosis or easy monitoring of knee OA in daily lives.</P>

LSTM과 CNN 기반 폭력 행위 감지 알고리즘

우준연(Junyeon Woo),권순재(Soonjae Kwon),송원석(Wonseok Song),이건호(Geonho Lee),안현식(Hyunsik Ahn) 대한전자공학회 2023 대한전자공학회 학술대회 Vol.2023 No.6