• Journal of Korea Artificial Intelligence Association
• /
• v.2 no.1
• /
• pp.7-14
• /
• 2024

In this paper, we explore the application of RNA-sequencing data and ensemble machine learning to predict lung cancer and treatment strategies for lung cancer, a leading cause of cancer mortality worldwide. The research utilizes Random Forest, XGBoost, and LightGBM models to analyze gene expression profiles from extensive datasets, aiming to enhance predictive accuracy for lung cancer prognosis. The methodology focuses on preprocessing RNA-seq data to standardize expression levels across samples and applying ensemble algorithms to maximize prediction stability and reduce model overfitting. Key findings indicate that ensemble models, especially XGBoost, substantially outperform traditional predictive models. Significant genetic markers such as ADGRF5 is identified as crucial for predicting lung cancer outcomes. In conclusion, ensemble learning using RNA-seq data proves highly effective in predicting lung cancer, suggesting a potential shift towards more precise and personalized treatment approaches. The results advocate for further integration of molecular and clinical data to refine diagnostic models and improve clinical outcomes, underscoring the critical role of advanced molecular diagnostics in enhancing patient survival rates and quality of life. This study lays the groundwork for future research in the application of RNA-sequencing data and ensemble machine learning techniques in clinical settings.

• Journal of Biomedical Engineering Research
• /
• v.45 no.1
• /
• pp.1-9
• /
• 2024

Communication with others plays an important role in human social interaction and information exchange in modern society. However, some individuals have difficulty in communicating due to dysarthria. Therefore, it is necessary to develop effective diagnostic techniques for early treatment of the dysarthria. In the present study, we propose a mobile device-based methodology that enables to automatically classify dysarthria type. The light-weight CNN model was trained by using the open audio dataset of Korean patients with dysarthria. The trained CNN model can successfully classify dysarthria into related subtype disease with 78.8%~96.6% accuracy. In addition, the user-friendly mobile application was also developed based on the trained CNN model. Users can easily record their voices according to the selected inspection type (e.g. word, sentence, paragraph, and semi-free speech) and evaluate the recorded voice data through their mobile device and the developed mobile application. This proposed technique would be helpful for personal management of dysarthria and decision making in clinic.

• Journal of radiological science and technology
• /
• v.40 no.1
• /
• pp.35-40
• /
• 2017

The purpose of this study was to investigate the relationship between abdominal subcutaneous fat thickness(ASFT) and maternal gestational diabetes mellitus(GDM) measured by ultrasound at period of pregnancy. We compared maternal age, pre-pregnancy body mass index, and weight gain during pregnancy in 286 pregnant women who were diagnosed with early pregnancy ASFT and high GDM screening test(50 g OGTT) of more than 140 mg/dL. ROC curve analysis was used to determine the cut-off value of ASFT for GDM prediction. Maternal age and weight gain during pregnancy were not related to GDM in the mid-trimester and pre-pregnancy body mass index and earely pregnancy ASFT were significantly different between normal and GDM high risk groups. The cut-off value of ASFT for GDM prediction was 2.23 cm(AUC 0.913. Sensitivity 76.19%, Specificity 93.72%). ASFT measured by ultrasound in early pregnancy was useful as an important index for predicting mid-trimester GDM prediction. Therefore, ASFT can be used as an auxiliary diagnostic index for early recognition of GDM.

• Korean Journal of Environmental Biology
• /
• v.37 no.3
• /
• pp.433-445
• /
• 2019

This study was designed to develop simple tools to easily and efficiently predict the occurrence of algal bloom in agricultural lakes. Physicochemical water quality parameters were examined to reflect the phytoplankton productivity in 182 samples collected from 15 agricultural lakes from April to October 2018. Total phytoplankton abundance was significantly correlated with chlorophyll-a (Chl-a) (r=0.666) and Secchi depth (SD) (r= -0.351). The abundances of cyanobacteria and harmful cyanobacteria were also correlated with Chl-a (r=0.664, r=0.353) and SD (r= -0.340, r= -0.338), respectively, but not with total nitrogen (TN) and total phosphorus (TP). The Chl-a concentration was correlated with SD (r= -0.434), showing a higher similarity than phytoplankton abundance. Therefore, Chl-a and SD were selected as diagnostic factors for algal bloom prediction, instead of analyzing the standing crop of harmful cyanobacteria used in algae alarm systems. Specifically, accurate diagnoses were made using realtime SD measurements. The algal bloom diagnostic tool is an inverse cone-shaped container with an algal bloom diagnosis chart that modified SD and turbidity measurement methods. Lake water was collected to observe the number of rings visible in the container or the number indicated in each ring, depending on the degree of algal bloom,and to determine the final stage of algal blooming by comparison to the colorimetric level on the diagnosis chart. For an accurate diagnosis, we presented 4-step diagnostic criteria based on the concentration of Chl-a and the number of rings and a fan-shaped algal bloom diagnosis chart with Hexa code names. This tool eliminated the variables and errors of previous methods and the results were easily interpreted. This study is expected to facilitate the diagnosis of algal bloom in agricultural lakes and the establishment of an efficient algal bloom management plan.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.28 no.1
• /
• pp.16-24
• /
• 2001

The purpose of this study was to evaluate the specificity, sensitivity, and diagnostic power of caries activity test using laser fluorescence. The subjects of this study were 50 children of $7\sim9$ years old. Fluorescence from initial carious lesion of teeth illuminated by an argon laser(480nm) was observed and photographed with barrier filter. Visual examination for the dDfFtT rate and Streptococcus mutans colony counting was done to evaluate correlation with caries activity test using laser fluorescence. Data analysis was accomplished by Axelsson's method. The results from the present study can be summarized as follows: 1. There was positive correlation $(\gamma=0.48)$ between laser fluorescence test and Streptococcus mutans count. And also positive correlation $(\gamma=0.39)$ exists between laser fluorescence test and dDfFtT rate (P<0.01). 2. Positive correlation $(\gamma=0.27)$ between Streptococcus mutans colony count and dDfFtT rate was found(P<0.05). 3. When dDfFtT rate was defined to standard testing method, the specificity, senstivity, and diagnostic power of laser fluorescence test were 44.4%, 85.7%, and 87.8%. 4. When dDfFtT rate was defined to standard testing method, the specificity, senstivity, and diagnostic power of S. mutans colony counting were 77.8%, 92.9%, 84.8%. 5. When S. mutans colony counting was defined to standard testing method, sensitivity, specificity and diagnostic power of laser fluorescence test were 40.0%, 84.8%, 95.1%. In regard to above results, laser fluorescence test considered to be accurate and reliable method for determining caries activity because of it's close relationship with caries susceptibility test and caries experience measurements. And it was also considered to be practical because it would be simple, inexpensive, and time saving method.

• Journal of the Korean Society of Radiology
• /
• v.10 no.6
• /
• pp.443-450
• /
• 2016

In this study, we are using a computer tomography image of the abdomen, as an experimental linear research for the image of the fatty liver patients texture features analysis and computer-aided diagnosis system of implementation using the ROC curve analysis, from the computer tomography image. We tried to provide an objective and reliable diagnostic information of fatty liver to the doctor. Experiments are usually a fatty liver, via the wavelet transform of the abdominal computed tomography images are configured with the experimental image section, shows the results of statistical analysis on six parameters indicating a feature value of the texture. As a result, the entropy, average luminance, strain rate is shown a relatively high recognition rate of 90% or more, the control also, flatness, uniformity showed relatively low recognition rate of about 70%. ROC curve analysis of six parameters are all shown to 0.900 (p = 0.0001) or more, showed meaningful results in the recognition of the disease. Also, to determine the cut-off value for the prediction of disease six parameters. These results are applicable from future abdominal computed tomography images as a preliminary diagnostic article of diseases automatic detection and eventual diagnosis.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.31 no.5
• /
• pp.399-408
• /
• 2005

In the progression of the Temporomandibular Joint Disorder(TMD), not only deformation and perforation of disc occur. But also fibrotic adhesion and inflammatory changes to the retrodiscal tissue can be seen in addition to the condylar degenerative change (e.g. osteoarthritis). However, the correct diagnosis,?planning for appropriate treatment, and prediction of prognosis are limited, because there are no means to stage the progression of the disorder. In this study relative signal intensity of retrodiscal tissue in MRI and the synovial fluid concentration of matrix metalloproteinase-2 (MMP-2), MMP-9, and Interleukin-6(IL-6) in the 23 temporomandibular joints(TMJ), from 17 patients with TMD were evaluated as a possible diagnostic marker. The relative signal intensity of retrodiscal tissue was referenced to brain gray matter with same region of interest(ROI) size. The concentrations of MMP-2, MMP-9, and IL-6 were evaluated by Enzyme Linked Immunosorbent Assay (ELISA). The collected data were compared with condylar degenerative change, joint effusion and disc position observed in MRI. The relative signal intensity of the retrodiscal tissue was increased significantly when degenerative changes were present. In addition, there was significantly high signal intensity in the presence of a disc displaced without reduction. The concentration of IL-6 was significantly increased when condylar degenerative change was no observed. And there were no changes in the levels of IL-6 according to disc position and joint effusion measurement. Moreover, there were no significant relevance between the concentration of total MMP-2 and active MMP-9 in synovial fluid, relative to degenerative changes in the mandibular condyle, to joint effusion, and to disc position observed on MRI images. In conclusion, the relative signal intensity of the retrodiscal tissue can be regarded as a mean of diagnosing the procession of TMD in a non-invasive manner. But more additional studies are required for the levels of MMP-2. MMP-9, and IL-6 to determine their potentials as a diagnostic marker for TMD.

• Asian Pacific Journal of Cancer Prevention
• /
• v.17 no.4
• /
• pp.1861-1864
• /
• 2016

Although early diagnosis of thyroid nodule type is very important, the diagnostic accuracy of standard tests is a challenging issue. We here aimed to find an optimal combination of factors to improve diagnostic accuracy for distinguishing malignant from benign thyroid nodules before surgery. In a prospective study from 2008 to 2012, 345 patients referred for thyroidectomy were enrolled. The sample size was split into a training set and testing set as a ratio of 7:3. The former was used for estimation and variable selection and obtaining a linear combination of factors. We utilized smoothly clipped absolute deviation (SCAD) logistic regression to achieve the sparse optimal combination of factors. To evaluate the performance of the estimated model in the testing set, a receiver operating characteristic (ROC) curve was utilized. The mean age of the examined patients (66 male and 279 female) was $40.9{\pm}13.4years$ (range 15- 90 years). Some 54.8% of the patients (24.3% male and 75.7% female) had benign and 45.2% (14% male and 86% female) malignant thyroid nodules. In addition to maximum diameters of nodules and lobes, their volumes were considered as related factors for malignancy prediction (a total of 16 factors). However, the SCAD method estimated the coefficients of 8 factors to be zero and eliminated them from the model. Hence a sparse model which combined the effects of 8 factors to distinguish malignant from benign thyroid nodules was generated. An optimal cut off point of the ROC curve for our estimated model was obtained (p=0.44) and the area under the curve (AUC) was equal to 77% (95% CI: 68%-85%). Sensitivity, specificity, positive predictive value and negative predictive values for this model were 70%, 72%, 71% and 76%, respectively. An increase of 10 percent and a greater accuracy rate in early diagnosis of thyroid nodule type by statistical methods (SCAD and ANN methods) compared with the results of FNA testing revealed that the statistical modeling methods are helpful in disease diagnosis. In addition, the factor ranking offered by these methods is valuable in the clinical context.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.23 no.3
• /
• pp.540-547
• /
• 2009

There needs disease-predictable signs in order to enable preventive diagnosis and therapy. Then traditional Chinese medicine applies various medical diagnostic equipments used in western medicine to diagnosing sub-healthy state. But such data are not originated from inherent oriental medicine, and not obtained easily in ordinary clinical practice. This paper is to provide synopsis of the ante-disease diagno-therapeutics partly and to show predictable data based on the facial shapes and signs, especially of gall bladder's versus bladder's body and masculine versus feminine shape. Ante-disease means not only the complete healthy state, but also the state unseen any symptoms in macrographically in the course of outbreak of disease. It contains two stages, first one is the former state of disease and second one is untransmitted state of disease. The patterns of ante-disease consist of latent disease, pre-disease, transmission type like senescent syndrome, abnormal reactive syndrome(變證), syndrome of transmission and transmutation. The classification with gall bladder and bladder type manifests the differences of shape, color and size of each organ in comparison of the universal and standard figures of the human being. On the other hand, the classification with masculine and feminine shape contrasts the innate sexual difference and the shape, characteristics originated from in itself. These two classification theories have their own pathologic types and syndrome types with each disease so that disease-predictable data can be constructed based on such a relationship. In addition, this diagnostic method by facial shapes and signs is able to be applied to whole stages from prenatal to present state of disease even if the cause and inducement are not clear. Ante-disease diagno-theraputic system by Gall Bladder's versus Bladder's Body and Masculine versus Feminine Shape is getting more important in the chronic and internal disease in comparison of the acute and traumatic disease. So this study is able to make up for the limit of diagnosis on ante-disease in the field of oriental medicine clinic.

• Tuberculosis and Respiratory Diseases
• /
• v.71 no.3
• /
• pp.180-187
• /
• 2011

Background: In mediastinal lymph node sampling in non-small cell lung cancer (NSCLC) it is important to determine the appropriate treatment as well as to predict an outcome. Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) is a recently developed, accurate, safe technique in patients with NSCLC for sampling mediastinal lymph nodes. We sought to determine the usefulness of EBUS-TBNA in mediastinal staging with NSCLC considered to be operable. Methods: We retrospectively reviewed the records of 142 patients who underwent EBUS-TBNA for mediastinal staging in the Asan Medical Center, Korea from July 2008 to July 2010. If patients were in an operable state, they underwent subsequent surgical staging. Diagnoses based on biopsy results were compared with those based on surgical results. Results: We performed EBUS-TBNA in 184 mediastinal lymph nodes in 142 NSCLC patients. Almost all of the EBUS-TBNA samples were from the lower paratracheal (112, 60.9%) and subcarinal (57, 31.0%) lymph nodes. In 142 patients, 51 patients (35.9%) were confirmed with malignant invasion of the mediastinal lymph node by EBUS-TBNA and 91 (64.1%) patients were not confirmed. Among the 91 patients, 64 patients (70.3%) underwent surgical staging. 3 patients (4.7%) who were misdiagnosed by the EBUS-TBNA were confirmed by surgery. After Diagnostic sensitivity of EBUS-TBNA, the prediction of mediastinal metastatsis was 94.4% and specificity was 100%. The procedures were performed safely and no serious complications were observed. Conclusion: We demonstrated the high diagnostic value of EBUS-TBNA for mediastinal staging.