• Journal of Yeungnam Medical Science
• /
• 제33권1호
• /
• pp.8-12
• /
• 2016

Background: The decision to administer oral anticoagulation therapy depends on accurate assessment of stroke risk in patients with atrial fibrillation (AF). Various stroke risk stratification schemes have been developed to help inform clinical decision making. The CHADS2 and CHA2DS2-VASc scores have been used in estimating the risk of stroke in patients with AF. Recently R2CHA2DS2-VASc score was developed. The objective of the current study is to validate the usefulness of the R2CHA2DS2-VASc score and to compare the accuracy of the CHADS2, CHA2DS2-VASc, and R2CHA2DS2-VASc scores in predicting a patient's risk of stroke. Methods: Based on medical records, we conducted a retrospective study of patients hospitalized with AF from March 2011 to July 2013. A total of 448 AF patients were included in this study. The receiver operating characteristic (ROC) curve analysis in MedCalc was used for comparison with respective diagnostic values. Results: The patient characteristics showed male predominance (60.9%). Among the 448 AF patients, 131 (29.2%) patients had strokes during the study. A R2CHA2DS2-VASc score of more than 5 is the optimal cut-off value for prediction of stroke. A risk score of three, the area under the ROC curve (AUC) of R2CHA2DS2-VASc score (AUC 0.631; 95% confidence interval, 0.585-0.679) was the highest. A significant difference was observed between AUC for R2CHA2DS2-VASc, CHADS2, and CHA2DS2-VASc scores, but no meaningful difference between CHADS2 and CHA2DS2-VASc scores. Conclusion: We determined the usefulness of the R2CHA2DS2-VASc score, which showed better association with stroke than the CHADS2 and CHA2DS2-VASc scores.

• 한국지리정보학회지
• /
• 제22권4호
• /
• pp.1-11
• /
• 2019

원격탐사 센서 기술의 발전으로 다양한 분광정보를 지니는 위성영상의 취득이 가능해졌다. 특히, 초분광 영상(hyperspectral image)은 연속적이고 좁은 분광파장대의 영역으로 구성되어 있기 때문에, 토지피복분류, 표적탐지, 환경 모니터링 등 다양한 분야에 효과적으로 활용할 수 있다. 원격탐사자료를 활용한 변화탐지 기법은 일반적으로 동일한 차원을 지닌 자료들의 차분을 통해 수행되기 때문에, 차원이 다른 이종 센서에는 적용하기 어려운 단점을 지니고 있다. 이에 본 연구에서는 다른 차원을 지닌 초분광 영상과 고해상도 위성영상에 적용가능한 변화탐지 기법을 개발하고, 이종 영상 간의 변화탐지기법 적용 가능성을 확인하고자 하였다. 이를 위하여, 변화탐지 기법의 적용을 위해 상관도분석, 주성분분석 등을 활용하여 초분광 영상의 차원을 축소시켜 변화탐지에 사용하였으며, 변화탐지 알고리즘은 CVA(Change Vector Analysis)을 사용하였다. 변화탐지 성능의 평가를 위해 참조자료를 사용하여 ROC(Receiver Operating Characteristics) 곡선과, AUC(Area Under Curve)을 계산하였다. 실험결과, 원 초분광 영상을 활용한 경우보다, 적합한 차원 감소 기법을 통해 제작한 영상을 사용하였을 때의 변화탐지 성능이 더 높은 것으로 나타났다. 이는 차원 감소 기법을 적용하여 초분광 영상이 지니고 있는 잡음을 제거하는 것이 변화탐지 성능에 영향을 미치는 것으로 판단된다. 추후 연구로는 융합기법을 적용한 고해상도 다중분광 영상을 이용하여 공간 해상도의 차이에 따른 변화탐지 성능을 분석할 예정이다.

• 한국지리정보학회지
• /
• 제21권3호
• /
• pp.19-34
• /
• 2018

본 논문의 목적은 농업용 저수지 저수율 계측자료와 기상인자와의 다중선형회귀분석을 통해 저수율 예측 월단위 회귀식을 산정하는데 있다. 2002년부터 2016년까지의 한국농어촌공사 저수지 3,067개에 대한 저수율 관측자료와 기상청 63개 지점 관측자료를 수집하여 저수율 예측 다중선형 회귀식을 도출하였으며, 개발된 월별 회귀식에 대한 $R^2$는 0.51~0.95로 분석되었다. 또한 회귀식의 적용성 평가를 위해 9개 대표저수지에 대해 관측값과 비교한 $R^2$는 0.44~0.81로 나타났다. 회귀식을 이용하여 평년(1976-2005) 대비 저수지 가뭄지수(Reservoir Drought Index, RDI)를 산정하여 ROC 분석을 수행한 결과, 극심한 가뭄의 경우 2년(2015~2016) 평균 적중률은 0.64로 겨울의 적중률이 0.70으로 가장 높았고, 여름의 적중률이 0.58로 가장 낮게 나타났으며, 봄과 가을의 적중률은 각각 0.59, 0.68로 분석되었다. 본 연구에서 도출한 회귀식은 가용한 관측자료 및 1~3개월의 장기 기상전망자료 기반의 월단위 저수율 전망자료 생산이 가능하므로, 이를 기반으로 농업가뭄 전망정보의 생산이 가능할 것으로 판단된다.

• 한국산업융합학회 논문집
• /
• 제24권6_2호
• /
• pp.729-735
• /
• 2021

Recently Artificial Intelligence(AI) has been developed and used in various fields. Especially AI recognition technology can perceive and distinguish images so it should plays a significant role in quality inspection process. For stability of autonomous driving technology, semiconductors inside automobiles must be protected from external electromagnetic wave(EM wave). As a shield film, a thin polymeric material with hole shaped micro-patterns created by a laser processing could be used for the protection. The shielding efficiency of the film can be increased by the hole structure with appropriate pitch and size. However, since the sensitivity of micro-machining for some parameters, the shape of every single hole can not be same, even it is possible to make defective patterns during process. And it is absolutely time consuming way to inspect all patterns by just using optical microscope. In this paper, we introduce a AI inspection system which is based on web site AI tool. And we evaluate the usefulness of AI model by calculate Area Under ROC curve(Receiver Operating Characteristics). The AI system can classify the micro-patterns into normal or abnormal ones displaying the text of the result on real-time images and save them as image files respectively. Furthermore, pressing the running button, the Hardware of robot arm with two Arduino motors move the film on the optical microscopy stage in order for raster scanning. So this AI system can inspect the entire micro-patterns of a film automatically. If our system could collect much more identified data, it is believed that this system should be a more precise and accurate process for the efficiency of the AI inspection. Also this one could be applied to image-based inspection process of other products.

• Nutrition Research and Practice
• /
• 제16권5호
• /
• pp.604-615
• /
• 2022

BACKGROUND/OBJECTIVES: This study aimed to investigate cancer-specific survival (CSS) and associated risk factors in elderly gastric cancer (EGC) patients. SUBJECTS/METHODS: EGC patients (≥ 70 yrs) who underwent curative gastrectomy between January 2013 and December 2017 at our hospital were included. Clinicopathologic characteristics and survival data were collected. Receiver operating characteristic (ROC) analysis was used to extract the best cutoff point for body mass index (BMI). A Cox proportional hazards model was used to determine the risk factors for CSS. RESULTS: In total, 290 EGC patients were included, with a median age of 74.7 yrs. The median follow-up time was 31 (1-77) mon. The postoperative 1-yr, 3-yr and 5-yr CSS rates were 93.7%, 75.9% and 65.1%, respectively. Univariate analysis revealed risk factors for CSS, including age (hazard ratio [HR] = 1.08; 95% confidence interval [CI], 1.01-1.15), intensive care unit (ICU) admission (HR = 1.73; 95% CI, 1.08-2.79), nutritional risk screening (NRS 2002) score ≥ 5 (HR = 2.33; 95% CI, 1.49-3.75), and preoperative prognostic nutrition index score < 45 (HR = 2.06; 95% CI, 1.27-3.33). The ROC curve showed that the best BMI cutoff value was 20.6 kg/m². Multivariate analysis indicated that a BMI ≤ 20.6 kg/m² (HR = 2.30; 95% CI, 1.36-3.87), ICU admission (HR = 1.97; 95% CI, 1.17-3.30) and TNM stage (stage II: HR = 5.56; 95% CI, 1.59-19.43; stage III: HR = 16.20; 95% CI, 4.99-52.59) were significantly associated with CSS. CONCLUSIONS: Low BMI (≤ 20.6 kg/m²), ICU admission and advanced pathological TNM stages (II and III) are independent risk factors for CSS in EGC patients after curative gastrectomy. Nutrition support, better perioperative management and early diagnosis would be helpful for better survival.

• Korean Journal of Radiology
• /
• 제22권5호
• /
• pp.770-781
• /
• 2021

Objective: Chemical exchange-dependent saturation transfer (CEST) MRI is sensitive for detecting solid-like proteins and may detect changes in the levels of mobile proteins and peptides in tissues. The objective of this study was to evaluate the characteristics of chemical exchange proton pools using the CEST MRI technique in patients with dementia. Materials and Methods: Our institutional review board approved this cross-sectional prospective study and informed consent was obtained from all participants. This study included 41 subjects (19 with dementia and 22 without dementia). Complete CEST data of the brain were obtained using a three-dimensional gradient and spin-echo sequence to map CEST indices, such as amide, amine, hydroxyl, and magnetization transfer ratio asymmetry (MTR_asym) values, using six-pool Lorentzian fitting. Statistical analyses of CEST indices were performed to evaluate group comparisons, their correlations with gray matter volume (GMV) and Mini-Mental State Examination (MMSE) scores, and receiver operating characteristic (ROC) curves. Results: Amine signals (0.029 for non-dementia, 0.046 for dementia, p = 0.011 at hippocampus) and MTR_asym values at 3 ppm (0.748 for non-dementia, 1.138 for dementia, p = 0.022 at hippocampus), and 3.5 ppm (0.463 for non-dementia, 0.875 for dementia, p = 0.029 at hippocampus) were significantly higher in the dementia group than in the non-dementia group. Most CEST indices were not significantly correlated with GMV; however, except amide, most indices were significantly correlated with the MMSE scores. The classification power of most CEST indices was lower than that of GMV but adding one of the CEST indices in GMV improved the classification between the subject groups. The largest improvement was seen in the MTR_asym values at 2 ppm in the anterior cingulate (area under the ROC curve = 0.981), with a sensitivity of 100 and a specificity of 90.91. Conclusion: CEST MRI potentially allows noninvasive image alterations in the Alzheimer's disease brain without injecting isotopes for monitoring different disease states and may provide a new imaging biomarker in the future.

• Laboratory Medicine Online
• /
• 제6권1호
• /
• pp.25-30
• /
• 2016

배경: Thymidine kinase 1 (TK1)은 세포 주기의 중요한 조절 효소로 암세포의 증식 시에 증가하는 것으로 알려져 있으며, 현재까지 혈액종양과 다양한 고형암에서 진단 또는 치료 후 모니터링과 예후 예측에 중요한 표지자로 보고되고 있다. 본 연구에서는 한국인에서 혈청 TK1 정량분석을 통하여 건강인의 혈청 TK1 참고치를 설정하고자 하였으며 B세포림프종 환자를 대상으로 임상적 악성도 표지자로서 혈청 TK1 검사의 유용성을 평가하고자 하였다. 방법: 72명의 B세포림프종 환자와 143명의 건강대조군의 혈청검체에서 화학발광면역측정법으로 혈청 TK1 농도를 측정하였다. 건강대조군에서 혈청 TK1의 참고치를 설정하였고, 환자군과 건강대조군에서 측정된 혈청 TK1 결과를 이용해 ROC 분석을 통한 기준치를 구하여 상대적으로 높은 혈청 TK1 정량값과 B세포림프종의 임상 지표들과의 상관성을 비교하였다. 결과: 전체 건강대조군의 혈청 TK1의 95 percentile에 따른 참고범위는 5.4-21.8 U/L였다. B세포림프종 환자군과 건강대조군의 혈청 TK1 수치 비교에서 평균${\pm}$표준편차는 각각 $40.6{\pm}68.5U/L$와 $11.8{\pm}4.4U/L$로 나타났으며 두 그룹 간에 유의한 차이를 보였다(P<0.001). ROC 분석 후, 혈청 TK1 수치 15.2 U/L를 이용하였을 때 민감도 59.7%, 특이도 83.2%, AUC 0.73을 보여 B세포림프종 환자를 선별할 수 있는 기준치로 설정하였다. 상대적으로 높은 혈청 TK1 수치(${\geq}15.2U/L$)는 병기, 골수침범, IPI 점수, LD 수치, 낮은 Hb (<12 g/dL), 림프구 수와 상관성이 있는 것으로 나타났다. 결론: B세포림프종 환자에서 측정되는 혈청 TK1 수치는 B세포림프종의 임상적 악성도 표지자로서 유용하게 사용될 수 있을 것이다.

• Journal of Chest Surgery
• /
• 제41권1호
• /
• pp.55-60
• /
• 2008

배경: 혈 중 BNP 수치는 심부전 환자의 예후에 정보를 제공한다고 잘 알려져 있다. 이 연구의 목적은 술 전과 술 후 BNP 수치가 관상동맥 수술한 환자의 술 후 합병증 및 결과를 예견하는 데 사용될 수 있는지 알아보고자 조사하였다. 대상 및 방법: 2005년 1월 1일부터 1년간 심폐기를 사용한 관상동맥 환자 30명을 전향적으로 자료 수집하였다(남/여=19/11, 평균 나이 $60{\pm}9.6$세). 심폐기를 사용하지 않고 관상동맥우회수술을 한 경우와 동시에 판막 수술을 시행한 경우는 대상에서 제외하였다. BNP 측정은 술 전, 수슬 직 후, 및 술 후 1, 3, 5, 7일째 혈액 채취를 하여 검사하였다. 결과: 술 전 BNP 수치와 술 전 초음파를 통한 심장 구혈률(r=-0.4, p=0.028)과 중환자실에서 5일 이상 치료한 경우(r=0.39, p=0.031)는 통계학적인 의미가 있었다. Receiver operating characteristics (ROC) curve를 사용하여 술 전 BNP 수치가 술 후 합병증을 예견하는 cut-off치는 특이도 90.5%에서 263 pg/mL이었다. 슬 후 합병증이 없는 군과 발생한 군 사이에서 BNP 수치의 변화를 보면 술 전 BNP 수치(평균 BNP=$99{\pm}23\;pg/mL$ vs $296{\pm}74\;pg/mL$, p<0.05)와 술 후 7일째 BNP 수치(평균 $212{\pm}29\;pg/mL$ vs $408{\pm}23\;pg/mL$, p<0.01)에서 통계학적으로 의미 있게 차이가 있었다. 결론: 술 전 BNP 수치가 263 pg/mL 이상이면 술 후 합병증을 예견할 수 있다. BNP 수치는 관상동맥우회술 환자의 술 후 합병증의 가능성을 예견하는 유용한 지표로서 사용될 수 있다.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권11호
• /
• pp.4629-4635
• /
• 2014

Background: Diffusion-weighted imaging (DWI) makes it possible to detect malignant tumors based on the diffusion of water molecules. However, it is uncertain whether DWI has advantages over FDG-PET for distinguishing malignant from benign pulmonary nodules and masses. Materials and Methods: One hundred-forty-three lung cancers, 17 metastatic lung tumors, and 29 benign pulmonary nodules and masses were assessed in this study. DWI and FDG-PET were performed. Results: The apparent diffusion coefficient (ADC) value ($1.27{\pm}0.35{\times}10^{-3}mm^2/sec$) of malignant pulmonary nodules and masses was significantly lower than that ($1.66{\pm}0.58{\times}10^{-3}mm^2/sec$) of benign pulmonary nodules and masses. The maximum standardized uptake value (SUVmax: $7.47{\pm}6.10$) of malignant pulmonary nodules and masses were also significantly higher than that ($3.89{\pm}4.04$) of benign nodules and masses. By using optimal cutoff values for ADC ($1.44{\times}10^{-3}mm^2/sec$) and for SUVmax (3.43), which were determined with receiver operating characteristics curves (ROC curves), the sensitivity (80.0%) of DWI was significantly higher than that (70.0%) of FDG-PET. The specificity (65.5%) of DWI was equal to that (65.5%) of FDG-PET. The accuracy (77.8%) of DWI was not significantly higher than that (69.3%) of FDG-PET for pulmonary nodules and masses. As the percentage of bronchioloalveolar carcinoma (BAC) component in adenocarcinoma increased, the sensitivity of FDG-PET decreased. DWI could not help in the diagnosis of mucinous adenocarcinomas as malignant, and FDG-PET could help in the correct diagnosis of 5 out of 6 mucinous adenocarcinomas as malignant. Conclusions: DWI has higher potential than PET in assessing pulmonary nodules and masses. Both diagnostic approaches have their specific strengths and weaknesses which are determined by the underlying pathology of pulmonary nodules and masses.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권16호
• /
• pp.6805-6810
• /
• 2014

Background: Previous studies have suggested a link between obesity and breast cancer (BC). However, there is no universal consensus, especially in population based studies. Because only few studies have been conducted on African women, we aimed here to assess the relationship between BMI at time of diagnosis and the BC histopathological features among Tunisian patients according to menopausal status using a hospital-based prospective cohort study. Materials and Methods: Clinical and pathological data were collected from 262 patients stratified on four groups according to their BMI. The relationship between BMI and histopathological features at diagnosis was analysed using univariate and multivariate analysis. Receiver-operating characteristic (ROC) curves were used to evaluate the performance of BMI in predicting of high tumor grade, in comparison to ki-67 index of proliferation. Results: Obesity was correlated with larger tumors, advanced grade and with ER-PR-Her2+ BC subtype. An association of BMI with tumor size and tumor grade was observed in both premenopausal and postmenopausal women. Additionally, a significant association between BMI and ER+, ER+PR+Her2+ and ER-PR-Her2+ status was revealed for premenopausal patients, while only ER+PR+Her2+ was associated with BMI for postmenopausal women. Finally, our results showed that compared to Ki67 proliferation index, BMI is a useful prognostic marker of high grade BC tumors. Conclusions: These data are the first to show that in Tunisia obese women suffering from BC have significantly larger tumors and advanced tumor grade and that higher BMI might influence tumor characteristics and behavior.