• Journal of Korea Multimedia Society
• /
• v.13 no.5
• /
• pp.722-729
• /
• 2010

Prediction problems are widely used in medical domains. For example, computer aided diagnosis or prognosis is a key component in a CDSS (Clinical Decision Support System). SVMs with nonlinear kernels like RBF kernels, have shown superior accuracy in prediction problems. However, they are not preferred by physicians for medical prediction problems because nonlinear SVMs are difficult to visualize, thus it is hard to provide intuitive interpretation of prediction results to physicians. Nomogram was proposed to visualize SVM classification models. However, it cannot visualize nonlinear SVM models. Localized Radial Basis Function (LRBF) was proposed which shows comparable accuracy as the RBF kernel while the LRBF kernel is easier to interpret since it can be linearly decomposed. This paper presents a new tool named VRIFA, which integrates the nomogram and LRBF kernel to provide users with an interactive visualization of nonlinear SVM models, VRIFA visualizes the internal structure of nonlinear SVM models showing the effect of each feature, the magnitude of the effect, and the change at the prediction output. VRIFA also performs nomogram-based feature selection while training a model in order to remove noise or redundant features and improve the prediction accuracy. The area under the ROC curve (AUC) can be used to evaluate the prediction result when the data set is highly imbalanced. The tool can be used by biomedical researchers for computer-aided diagnosis and risk factor analysis for diseases.

• Korean Journal of Animal Reproduction
• /
• v.11 no.2
• /
• pp.132-138
• /
• 1987

This study was carried out to evaluate the ability of clinical application of pregnancy diagnosis based upon the determinatin of progesterone in milk, utilizing a chymosin inhibitor labelled with progesterone and monoclonal antibody to progesterone, and its compared with progesterone concentrations in the milk were assayed by radioimmunoassay. 1. The progesterone concentration of the pregnant cows (2.07$\pm$0.54ng/ml) were significantly higher than those of non-pregnant cows (1.04$\pm$0.19 ng/ml), and thereafter began to increase and maintained high levels. 2. During 20 to 22 days after artificial insemination, the accuracy of pregnancy diagnosis from monoclonal antigen of progesterone were 92.9% for non-pregnant cows, and 88.5% for pregnant cows. 3. During 20 to 22 days after artificial inseminatin, the accuracy of pregnancy diagnosis from milk progesterone concentrations were 92.9% for non-pregnant cows(<3.4ng/ml), and 92.3% for pregnant cows( 4.0ng/ml). The average overall accuracy of pregnancy prediction for pregnant and non-pregnant cows were 92.6%. 4. Accordingly, the pregnancy diagnosis from monoclonal antigen of progesterone is thought to be recommendable because this early diagnostic means are simple with accurate result.

• Korean Journal of Animal Reproduction
• /
• v.13 no.3
• /
• pp.157-163
• /
• 1989

This stduy was carried out to evaluate the ability of clinical application of pregnancy diagnosis based upon the determination of progesterone in serum, utilizing EIA-kit of progesterone concentrations in the serum were assayed by radioimmunoassay. 1. The progesterone concentrations of the pregnant cows(2.40$\pm$0.34ng/ml) were significantly higher than those of non-pregnant cows(1.03$\pm$0.09ng/ml), and thereafter began to increase and maintained high levels. 2. During 20 to 22 days after artificial insemination, the accuracy of pregnancy diagnosis from EIA-kit of progesterone were 95.0% for non-pregnant cows, and 92.3% for pregnant cows. 3. During 20 to 22 days after artificial insemination, the accuracy of pregnancy diagnosis from serum progesterone concentrations were 100% for non-pregnant cows(<1.4ng/ml), and 96.2% for pregnant cows( 2.0ng/ml). The average overall accuracy of prediction for pregnant and non-pregnant cows were 98.1%. 4. Accordingly, the pregnancy diagnosis from EIA-kit of progesterone is thought to be recommendable because this early diagnostic means are simple with accurate results.

• Journal of Sasang Constitutional Medicine
• /
• v.34 no.2
• /
• pp.1-14
• /
• 2022

Objectives The aim of this study was to investigate the change in Sasang constitution prediction value in 2 years and find the factors associated with it. Methods Cohort data from Korean medicine data center was used. Using Korean Sasang Constitutional Diagnostic Questionnaire (KS-15) which consist of questions related to body shape, temperament, and symptoms, participants were categorized into Tae-Yang (TY), Tae-Eum (TE), So-Yang (SY), and So-Eum (SE). Sasang constitution was assessed on the baseline and after two years. Result Total 5,784 participants were analyzed. (TE 3, 341; SE 911; SY 1,532). Among them, 1,402 participants (24.2%) showed different prediction value in KS-15 after two years. The proportion of participants showing different prediction value in two years was the highest in SY, and the lowest in TE group. The factors associated with the change in Sasang constitution prediction value were different by constitution type. The change in feeling after sweating was significantly associated with the change in prediction value in TE and SY groups, not in SE group. Although temperament was not significantly associated with the change in prediction value from TE to SE, it was significantly associated with that in the change from TE to SY. The change in BMI and appetite were associated with the change in constitution prediction value in all three constitution types. Conclusion Although the factors associated with the change in prediction value of Sasang constitution were different by each constitution type, BMI and appetite were significant in all three types. These factors could be useful for developing Sasang constitution questionnaire and deciding re-prediction needs of Sasang constitution. Further research about the factors related to Sasang constitution diagnosis need to be conducted.

• Journal of Korean Neurosurgical Society
• /
• v.61 no.3
• /
• pp.376-385
• /
• 2018

Recent discoveries of brain tumor-related genes and fast advances in genomic testing technologies have led to the era of molecular diagnosis of brain tumor. Molecular profiling of brain tumor became the significant step in the diagnosis, the prediction of prognosis and the treatment of brain tumor. Because traditional molecular testing methods have limitations in time and cost for multiple gene tests, next-generation sequencing technologies are rapidly introduced into clinical practice. Targeted sequencing panels using these technologies have been developed for brain tumors. In this article, focused on pediatric brain tumor, key discoveries of brain tumor-related genes are reviewed and cancer panels used in the molecular profiling of brain tumor are discussed.

• Clinical and Experimental Pediatrics
• /
• v.52 no.9
• /
• pp.964-970
• /
• 2009

Neonatal seizures are generally not only brief and subtle but also not easily recognized and are usually untreated. In sick neonates, seizures are frequently not manifested clinically but are detected only by electroencephalography (subclinical EEG seizures). This phenomenon of electroclinical dissociation is fairly common in neonates. On the other hand, neonates frequently show clinical behaviors such as stiffening, apnea, or autonomic manifestations that mimic seizures, which is usually associated with underlying encephalopathy and non-epileptic seizures. Therefore, it might be difficult to confirm the diagnosis of neonatal seizures. Early recognition of neonatal seizures is important to minimize poor neurodevelopmental outcomes, including cognitive, behavioral, and learning disabilities, as well as the development of postnatal epilepsy. EEG is a reliable tool in the determination of neonatal seizures. Continuous EEG monitoring is essential for the identification of seizures, evaluation of treatment efficacy, and prediction of the neurodevelopmental outcome. However, there is not yet a wide consensus on the optimal "standard" lead montage for the continuous EEG monitoring.

• Journal of Biomedical Engineering Research
• /
• v.38 no.5
• /
• pp.271-276
• /
• 2017

Cancer biomarkers are using in the diagnosis, staging, prognosis and prediction of disease progression. But, there are not sufficiently profiled and validated in early detection and risk classification of prostate cancer. In this study, we have devoted to finding a panel of serum biomarkers that are able to detect the diagnosis of prostate cancer. The serum samples were consisted of 111 prostate cancer and 343 control samples and examined. Eleven biomarkers were constructed in this study, and then nine biomarkers were relevant to candidate biomarkers by using t test. Finally, four biomarkers, PSA, ApoA2, CYFRA21.1 and TTR, were selected as the prostate cancer biomarker panel, logistic regression was used to identify algorithms for diagnostic biomarker combinations(AUC = 0.9697). A panel of combination biomarkers is less invasive and could supplement clinical diagnostic accuracy.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.27 no.5
• /
• pp.553-560
• /
• 2003

On-site diagnosis of chiller performance is an essential step fur energy saving business. The main purpose of the on-site diagnosis is to predict the COP of a target chiller. Many models based on thermodynamics background have been proposed for this purpose. However, they have to be modified from chiller to chiller and require deep insight into thermodynamics that most of field engineers are often lacking in. This study focuses on developing an easy-to-use diagnostic technique that is based on adaptive neuro-fuzzy inference system (ANFIS). Quality of the training data for ANFIS, sampled over June through September, is assessed by checking COP prediction errors. The architecture of the ANFIS, its error bounds, and collection of training data are described in detail.

• Genomics & Informatics
• /
• v.17 no.3
• /
• pp.30.1-30.4
• /
• 2019

Neuroblastoma is a major cause of cancer death in early childhood, and its timely and correct diagnosis is critical. Gene expression datasets have recently been considered as a powerful tool for cancer diagnosis and subtype classification. However, no attempts have yet been made to apply deep learning using gene expression to neuroblastoma classification, although deep learning has been applied to cancer diagnosis using image data. Taking the International Neuroblastoma Staging System stages as multiple classes, we designed a deep neural network using the gene expression patterns and stages of neuroblastoma patients. Despite a small patient population (n = 280), stage 1 and 4 patients were well distinguished. If it is possible to replicate this approach in a larger population, deep learning could play an important role in neuroblastoma staging.

• Journal of the Korean Society of Safety
• /
• v.10 no.3
• /
• pp.110-114
• /
• 1995

Bacause of increasing the use in Industrial robots, the accident rate has been increasing now a days. The prediction of accident could be very hard as there are so many factors which occured accident. Removing the accident factors in industrial robots can be diagnosed by the human experts who are very familiar with in those area. The purpose of this study is a development of finger fault diagnosis system for industrial robots. We have many problems such as a long time to get the expert knowledge and the number of expert to be limited. To solve these problems lots of investment and time are required, and then the exepert system to finger fault diagnosis for industrial robots can be applied.