• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.5
• /
• pp.331-337
• /
• 2023

In this paper, we propose a deep learning model to detect lumbar 3 (L3) CT images to determine the occurrence and degree of sarcopenia. In addition, we would like to propose an optimization technique that uses oversampling ratio and class weight as design parameters to address the problem of performance degradation due to data imbalance between L3 level and non-L3 level portions of CT data. In order to train and test the model, a total of 150 whole-body CT images of 104 prostate cancer patients and 46 bladder cancer patients who visited Gangneung Asan Medical Center were used. The deep learning model used ResNet50, and the design parameters of the optimization technique were selected as six types of model hyperparameters, data augmentation ratio, and class weight. It was confirmed that the proposed optimization-based L3 level extraction model reduced the median L3 error by about 1.0 slices compared to the control model (a model that optimized only 5 types of hyperparameters). Through the results of this study, accurate L3 slice detection was possible, and additionally, we were able to present the possibility of effectively solving the data imbalance problem through oversampling through data augmentation and class weight adjustment.

• Journal of Veterinary Clinics
• /
• v.40 no.5
• /
• pp.336-340
• /
• 2023

A dog (2-year old, female, Shih-Tzu) presented with hyperthermia and right-sided facial paralysis characterized by the inability to close the right eye and drooling from the right side of the mouth after H3N2 influenza vaccination [A/Canine/Korea/01/07(H3N2) strain; Caniflu-Max, Bionote, Hwaseong, Gyeonggi-do, ROK]. To determine the cause of the fever and neurological symptoms, physical examination, ophthalmic examination, thoracic and abdominal radiography, abdominal ultrasonography, complete blood counts, serum chemistry values, and electrolyte levels were determined. In addition, Cerebrospinal fluid analysis, antinuclear antibody test, fever of unknown origin polymerase chain reaction (PCR) panel, tick-borne pathogen PCR panel were performed. As a result, hyperthermia, leukocytosis, and elevated C-reactive protein were confirmed. In addition, neurological examination revealed decreased right eyelid reflexes, corneal reflexes, threat response, and facial sensation, it was possible to suspect problems with the trigeminal and facial nerves of the cranial nerve. Magnetic resonance imaging revealed a lesion suggestive of myositis in the right muscular lesion at atlanto-occipital junction level on site of vaccine injection. Therefore, right-sided facial paralysis was tentatively determined to be a secondary cause of nerve damage caused by myositis. The patient was treated with immunosuppressants such as prednisolone and mycophenolate mofetil. After 3 months of immunosuppressant therapy, the patient's symptoms improved.

• Proceedings of the Plant Resources Society of Korea Conference
• /
• 2022.09a
• /
• pp.85-85
• /
• 2022

영농형 태양광발전은 태양의 일사량을 전기발전과 영농에 공유하는(solar-sharing) 방식이다. 본 연구는 신재생에너지의 활용의 극대화를 위하여 추적식 영농형 태양광발전시스템을 구축하고 시설하부에서 일정 기간 재배중인 작물의 하부 환경과 생육을 조사하여 영농형태양광 하부작물개발을 위한 기초자료를 확보하고자 하였다. 구축한 추적식 영농형 태양광발전시스템은 4열 6단의 24장 모듈(8m × 6m)을 가지며, 발전시설 중심축 기둥 간 중심간격 14m로 단일지주식 스크루 공법으로 순천대학교 부속농장 답작포(순천시 죽평리)에 설치하여 하부 환경과 하부작물의 생육을 조사하였다. 태양광발전시설 하부작물의 생육을 조사하기 위하여 순천 농협육묘장에서 벼(신동진)를 육묘하여 2022년 6월 16일 이앙하였다. 태양광발전시스템 하부 지역을 4방위 방향에 따라 강음영(중심축으로부터 1~3m), 중음영(5m), 약음영(7~9m) 구역으로 설정하여 생육을 조사한 결과, 방위에 따른 초장은 남쪽에서 음영간 차이가 상대적으로 낮게 나타났으며, 1번기 태양광 발전시설에 의하여 음영이 중첩된 2번기 시설의 동쪽에서 대조구 대비 초장이 상대적으로 낮은 경향을 나타내었다. 음영강도에 따른 초장은 대체로 강음영구에서 낮게 나타났으며, 약음영구로 갈수록 높게 나타났다. 엽수는 방위에 따라서, 그리고 음영의 강도에 따른 차이가 초장에 비하여 작게 나타났다. 출수기의 경우 방위별로는 남쪽에서 음영별 차이가 작게 나타났으며, 음영강도에 따라서 차이를 보였다. 또한 태양광시설 하부에 데이터수집장치(Model 1650, Spctrum Technonogies, USA)를 설치하여 음영에 따른 토양전도도, 토양함수량, 토양온도, par light 등 생육환경을 조사, 비교하였다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.6
• /
• pp.1077-1086
• /
• 2023

The searchable encryption model allows to selectively search for encrypted data stored on a remote server. In a real-world scenarios, the model must be able to support multiple search keywords, multiple data owners/users. In this paper, these models are referred to as Multi Ranked Searchable Encryption model. However, at the time this paper was written, the proposed models use fully-trusted trapdoor centers, some of which assume that the connection between the user and the trapdoor center is secure, which is unlikely that such assumptions will be kept in real life. In order to improve the practicality and security of these searchable encryption models, this paper proposes a new Multi Ranked Searchable Encryption model which uses random keywords to protect search words requested by the data downloader from an honest-but-curious trapdoor center with an external attacker without the assumptions. The attacker cannot distinguish whether two different search requests contain the same search keywords. In addition, experiments demonstrate that the proposed model achieves reasonable performance, even considering the overhead caused by adding this protection process.

• Journal of Biomedical Engineering Research
• /
• v.44 no.6
• /
• pp.450-464
• /
• 2023

Pulse oximetry, a non-invasive technique for evaluating blood oxygen saturation, conventionally depends on isolated measurements, rendering it vulnerable to factors like illumination profile, spatial blood flow fluctuations, and skin pigmentation. Previous efforts to address these issues through imaging systems often employed red and near-infrared illuminations with distinct profiles, leading to inconsistent ratios of transmitted light and the potential for errors in calculating spatial oxygen saturation distributions. While an integrating sphere was recently utilized as an illumination source to achieve uniform red and near-infrared illumination profiles on the sample surface, its bulkiness presented practical challenges. In this work, we have enhanced the pulse oximetry imaging system by transitioning illumination from an integrating sphere to a multi-wavelength LED configuration. This adjustment ensures simultaneous emission of red and near-infrared light from the same position, creating a homogeneous illumination profile on the sample surface. This approach guarantees consistent patterns of red and near-infrared illuminations that are spatially uniform. The sustained ratio between transmitted red and near-infrared light across space enables precise calculation of the spatial distribution of oxygen saturation, making our pulse oximetry imaging system more compact and portable without compromising accuracy. Our work significantly contributes to obtaining spatial information on blood oxygen saturation, providing valuable insights into tissue oxygenation in peripheral regions.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_4
• /
• pp.1111-1123
• /
• 2023

In recent times, semantic image segmentation methods using deep learning models have been widely used for monitoring changes in surface attributes using remote sensing imagery. To enhance the performance of various UNet-based deep learning models, including the prominent UNet model, it is imperative to have a sufficiently large training dataset. However, enlarging the training dataset not only escalates the hardware requirements for processing but also significantly increases the time required for training. To address these issues, transfer learning is used as an effective approach, enabling performance improvement of models even in the absence of massive training datasets. In this paper we present three transfer learning models, UNet-ResNet50, UNet-VGG19, and CBAM-DRUNet-VGG19, which are combined with the representative pretrained models of VGG19 model and ResNet50 model. We applied these models to building extraction tasks and analyzed the accuracy improvements resulting from the application of transfer learning. Considering the substantial impact of learning rate on the performance of deep learning models, we also analyzed performance variations of each model based on different learning rate settings. We employed three datasets, namely Kompsat-3A dataset, WHU dataset, and INRIA dataset for evaluating the performance of building extraction results. The average accuracy improvements for the three dataset types, in comparison to the UNet model, were 5.1% for the UNet-ResNet50 model, while both UNet-VGG19 and CBAM-DRUNet-VGG19 models achieved a 7.2% improvement.

• Journal of the Korean Society of Radiology
• /
• v.17 no.5
• /
• pp.735-742
• /
• 2023

Non-alcoholic fatty liver disease is an independent risk factor for the development of cardiovascular disease, diabetes, hypertension, and kidney disease, and the clinical importance of non-alcoholic fatty liver disease has recently been increasing. In this study, we aim to extract feature values by applying GLCM, a texture analysis method, to ultrasound images of patients with non-alcoholic fatty liver disease. By applying an artificial neural network model using extracted feature values, we would like to classify the degree of fat deposition in non-alcoholic fatty liver into normal liver, mild fatty liver, moderate fatty liver, and severe fatty liver. As a result of applying the GLCM algorithm, the parameters Autocorrelation, Sum of squares, Sum average, and sum variance showed a tendency for the average value of the feature values to increase as it progressed from mild fatty liver to moderate fatty liver to severe fatty liver. The four parameters of Autocorrelation, Sum of squares, Sum average, and sum variance extracted by applying the GLCM algorithm to ultrasound images of non-alcoholic fatty liver disease were applied as inputs to the artificial neural network model. The classification accuracy was evaluated by applying the GLCM algorithm to the ultrasound images of non-alcoholic fatty liver disease and applying the extracted images to an artificial neural network, showing a high accuracy of 92.5%. Through these results, we would like to present the results of this study as basic data when conducting a texture analysis GLCM study on ultrasound images of patients with non-alcoholic fatty liver disease.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.51 no.4
• /
• pp.46-55
• /
• 2023

This study investigated the public's awareness and purchase behavior of organic farming and environment-friendly agriculture products. This study also analyzed whether awareness affects environment-friendly agriculture products' consumption and price resistance and support for the revitalizing organic farming. This study derived environmental and public interst in organic farming, and a web survey was conducted for statistical analysis. As a result, it was found that the awareness of organic farming did not affect the consumption of environment-friendly agriculture products, but in case of high awareness is high, the resistance to prices is low. In addition, it was found that the stronger the public's awareness, the more positive the support for the expansion of organic agriculture and the willingness to purchase environment-friendly agriculture products. The results of this study are expected to be used as basic data for preparing measures to revitalize organic agriculture in the future.

• The Journal of Internal Korean Medicine
• /
• v.44 no.4
• /
• pp.774-790
• /
• 2023

Objective: This study examined the effectiveness of Korean medicine treatments in two patients with Parkinson's disease complaining of discomfort stemming from postural instability and gait disturbance (PIGD). Methods: Two patients were treated for 3 months. They visited the clinic once a week for the first month and thereafter once every 2 weeks. The Unified Parkinson's Disease Rating Scale (UPDRS) and a three-dimensional gait analysis were performed at the first visit and at 1, 2, and 3 months thereafter. Results: In Case 1, gait speed, stride length, and swing phase increased. Double support decreased until 2 months after treatment but increased slightly after 3 months. Among the kinematic parameters, tilt and rotation increased. The total UPDRS Part III score decreased from 51 points to 29 points after 3 months of treatment. In Case 2, gait speed, stride length, and swing phase increased, but double support decreased. Among the kinematic parameters, tilt, rotation, and obliquity decreased. The total UPDRS Part III score decreased from 11 points to 7 points after 3 months of treatment. Conclusions: This study suggests that Korean medicine can be an effective treatment for patients with Parkinson's disease who experience discomfort due to PIGD.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.5
• /
• pp.219-228
• /
• 2023

The purpose of this study is to find out the competence of porspective speech and language pathologist according to Clinical practicums and to use it as basic data in guiding porspective speech and language pathologist. The porspective speech and language pathologist competence consisted of tasks, knowledge, skills, and language areas, and a total of 36 questionnaires were organized by dividing the language areas into sub-areas of smantics, morphology and pragmatics. A total of 105 questionnaires were collected from students with experience in Clinical practicums. A t-test, Pearson correlation analysis, and simple regression analysis were conducted to analyze the competence of porspective speech and language pathologist according to whether or not they practiced. The results of this study are as follows. First, there were significant differences between groups in all areas of knowledge, tasks, skills, and language in the competence area. Second, there was a very strong correlation between competence and language sub-areas. Third, it was found that it had a significant explanatory power in the sub-area of competence and language areas, and had a positive effect on the competence of porspective speech and language pathologist. This study is meaningful in that it should be based on theoretical knowledge of language elements to enhance the competence of porspective speech and language pathologist, and it can be confirmed that theory affects the competence of porspective speech and language pathologist. It is expected to be meaningfully used as a basis for efficient teaching methods based on the improvement of the capabilities of porspective speech and language pathologist, training training professional language rehabilitators, and theory, and theory.