• 한국임상수의학회지
• /
• 제31권4호
• /
• pp.288-292
• /
• 2014

A critical assumption of the standard sample size calculation is that the response (outcome) for an individual patient is completely independent to that for any other patient. However, this assumption no longer holds when there is a lack of statistical independence across subjects seen in cluster randomized designs. In this setting, patients within a cluster are more likely to respond in a similar manner; patient outcomes may correlate strongly within clusters. Thus, direct use of standard sample size formulae for cluster design, ignoring the clustering effect, may result in sample size that are too small, resulting in a study that is under-powered for detecting the desired level of difference between groups. This paper revisit worked examples for sample size calculation provided in a previous paper using nomogram to easy to access. Then we present the concept of cluster design illustrated with worked examples, and introduce design effect that is a factor to inflate the standard sample size estimates.

• 대한한의학회지
• /
• 제25권1호
• /
• pp.1-20
• /
• 2004

Objectives : Based on the analysis of Yin-Yang[음양] characteristics and symptoms, each person is classified into Yin-Yang. Also the validity of the result is statistically analized. Methods : From Feb. to May. 2003, the data were collected through a questionnaire given to 690 patients. The questionnaire was composed of 34 items which were about personality, habit, sweat, response to coldness, thirst, bowel, urine, physical shape, and menstruation for women only. SD(Semantic Differential Technique) used for each item, each item is measured as a contrast of two opposite symptoms. Reliability analysis was used to select items and categories. Based on means of items in each category the Yin-Yang index was developed. The validity of Yin-Yang index was investigated using classification and clustering analysis. In statistical analysis, SPSS V10.0.7 PC was used. Results : The obtained results are summarized as follows: 1) We constructed Yin-Yang index based on the middle point of the sum of categorical means. Then we classified each person into Yin or Yang. 2) To investigate the validity of the distribution of personal Yin-Yang degree, the crosstabulation of results from clustering and classification was used. The hit ratio for classification was much higher than Maximum Chance Criterion($C_{max}$), and concurrence in crosstabulation was successful. Therefore we can infer that the distribution of Yin-Yang was valid. Conclusions : Based on Yin-Yang characteristics and symptoms, we was analyzed personal degree of Yin-Yang, and confirmed the validity of its distribution. Therefore this index can be used further for Bian-Zheng [변증] and classification of the constitution.

• 서비스연구
• /
• 제2권2호
• /
• pp.45-51
• /
• 2012

U-Healthcare는 홈 네트워크, 휴대용 장치 등에 기반한 정보통신기술과 의료시스템이 서로 융합되어 개인의 생체정보 등을 실시간으로 모니터링하고, 자동으로 병원 및 의사와 연결되어 시공간의 제약을 줄임으로써 언제 어디서나 건강을 관리하고 질병을 예방하는 새로운 형태의 의료서비스이다. 본 논문에서는 진료 중심에서 예방 중심으로 변화되어가고 있는 최근의 U-Healthcare 시스템의 기술 발전 추세에 맞추어 조기 대응이 가능한 Healthcare 정보시스템 구축을 위한 요구분석 사항들에 대해 정리하고, 이를 기반으로 u-Healthcare의 실현을 위한 기존의 단위 시스템인 PACS, OCS, EMR, 응급의료시스템을 통합한 환자중심의 클라이언트 시스템을 설계한다. 특히, 온톨로지는 특정분야의 정보 모델에 이용되어 그 분야에서 공통의 어휘를 제공하고, 그 용어의 의미와 용어간의 관계를 다양한 수준의 형식성을 가지고 제공한다. 본 논문에서는 이러한 온톨로지 및 무질서한 데이터에 대한 관계를 정의하고, 보다 체계적으로 데이터를 군집화하는 클러스터링의 개념을 포함한 환자중심의 서비스를 위한 온톨로지 기반의 시스템을 제안한다.

• 한국정보통신학회논문지
• /
• 제24권12호
• /
• pp.1618-1624
• /
• 2020

X-ray를 기반으로 하는 장 폐색 진단 방법은 검사자의 주관적인 요소가 포함되기 때문에 객관적 진단에 영향을 미칠 수 있다. 따라서 본 논문에서는 허프 변환과 PCM 클러스터링 기법을 적용하여 장폐색 영역을 추출하는 방법을 제안한다. 제안된 방법은 X-ray 장폐색 영상에서 ROI 영역을 추출한 후, 허프 변환 기법을 이용하여 ROI 영역에서 직선을 검출하고, 검출된 직선을 이용하여 공기 액체층의 형태학적 특징을 이용하여 대장 폐색을 추출한다. 그리고 추출된 ROI 영역을 PCM 클러스터링을 적용하여 ROI 영역을 양자화 한다. 양자화된 ROI 영역 중에서 대장 폐색의 특징이 포함된 클러스터의 그룹을 선정하고, 선정된 클러스터의 그룹에서 객체를 탐색하여 소장 장폐색 영역을 추출한다. 장폐색 환자의 X-ray 영상 30개를 대상으로 PCM 클러스터링을 적용한 결과, PCM의 초기 클러스터의 수를 4개로 설정한 경우가 장폐색 검출 성능이 우수하였고 TPR은 81.47%로 나타났다.

• Journal of Biosystems Engineering
• /
• 제33권5호
• /
• pp.340-347
• /
• 2008

Dipstick urinalysis is used as a routine test for a screening test of UTI (urinary tract infection) in primary practice because urine dipstick test is simple. The result of dipstick urinalysis brings medical professionals to make a microscopic examination and urine culture for exact UTI diagnosis, therefore it is emphasized on a role of screening test. The objective of this study was to the classification between UTI patients and normal subjects using hybrid neural network classifier with enhanced clustering performance in urine dipstick screening test. In order to propose a classifier, we made a hybrid neural network which combines with RBF layer, summation & normalization layer and L VQ artificial neural network layer. For the demonstration of proposed hybrid neural network, we compared proposed classifier with various artificial neural networks such as back-propagation, RBFNN and PNN method. As a result, classification performance of proposed classifier was able to classify 95.81% of the normal subjects and 83.87% of the UTI patients, total average 90.72% according to validation dataset. The proposed classifier confirms better performance than other classifiers. Therefore the application of such a proposed classifier expect to utilize telemedicine to classify between UTI patients and normal subjects in the future.

• 대한치과교정학회지
• /
• 제47권4호
• /
• pp.248-255
• /
• 2017

Objective: The purpose of this study was to investigate whether the craniofacial patterns of Korean children with snoring and adenotonsillar hypertrophy (ATH) could be categorized into characteristic clusters according to age. Methods: We enrolled 236 children with snoring and ATH (age range, 5-12 years) in this study. They were subdivided into four age groups: 5-6, 7-8, 9-10, and 11-12 years. Based on cephalometric analysis, the sagittal and vertical skeletal patterns of each individual were divided into Class I, II, and III, as well as the normodivergent, hypodivergent, and hyperdivergent patterns, respectively. Cluster analysis was performed using cephalometric principal components in addition to the age factor. Results: Three heterogeneous clusters of craniofacial patterns were obtained in relation to age: cluster 1 (41.9%) included patients aged 5-8 years with a skeletal Class I or mild Class II and hyperdivergent pattern; cluster 2 (45.3%) included patients aged 9-12 years with a Class II and hyperdivergent pattern; and cluster 3 (12.8%) included patients aged 7-8 years with a Class III and hyperdivergent pattern. Conclusions: This study found that the craniofacial patterns of Korean children with snoring and ATH could be categorized into three characteristic clusters according to age groups. Although no significantly dominant sagittal skeletal discrepancy was observed, hyperdivergent vertical discrepancy was consistently evident in all clusters.

• 응용통계연구
• /
• 제18권1호
• /
• pp.1-13
• /
• 2005

최근 많은 연구자와 실무자들이 모집단에 내재해 있는 여러 다른 그룹(class, segment)간의 이질성을 밝혀내고 객체들을 그룹별로 세분화하는 방법 중 하나로 잠재그룹 모델(Latent class model)을 고려하고 있다. 이 논문에서는 2000년도에 국립 암 센터에 접수된 한국 내 연령별 전립선암 사망자수 자료를 기반으로, 잠재그룹 포아송 모형을 이용하여 전립선암 환자의 연령에 따른 그룹화를 시도한다. 최우추정법 등 고전적 추론방법의 한계를 극복하기 위하여 Markov Chain Monte Carlo (MCMC) 방법을 도구로 한 베이지안 추정 방법을 제안한다. 제안된 베이지안 방법의 장점은 용이한 모수추정과 추정오차의 제공, 그리고 각 객체의 소속그룹의 판정과 이에 따르는 오차, 즉, 객체의 각 군집에 속할 확률, 도 구할 수 있다는 것이다. 또한 주어진 자료들에 대해 가장 적합한 그룹의 수를 결정하는 방법을 제시하여 그룹의 수나 세분화의 근거를 사전에 제공하지 않아도 자료가 주는 정보로부터 이들을 자동으로 결정하는 방법을 제시한다.

• 보건행정학회지
• /
• 제32권4호
• /
• pp.347-355
• /
• 2022

Background: Since November 2019, long-term care hospitals have been able to provide patients with discharging programs to support the elderly in the community. This study aimed to identify both patient- and hospital-level factors that affect successful community discharge from long-term care hospitals. Methods: A multilevel logistic regression model was performed using hospitals as a clustering unit. The dependent variable was whether a patient stayed in the community for at least 30 days after discharge from a long-term care hospital. As for the patient-level independent variables, an agreement between a patient and the family about discharge, length of hospital stay, patient category, and residence at discharge were included. The number of beds and the ratio of long-stay patients were selected for the hospital-level factors. The sample size was 1,428 patients enrolled in the discharging program from November 2019 to December 2020. Results: The number of patients who were discharged to the community and stayed at least for 30 days was 532 (37.3%). The intraclass correlation coefficient was 22.9%, indicating that hospital-level factors had a significant impact on successful community discharge. The odds ratio (OR) of successful community discharge increased by 1.842 times when the patients and their families agreed on discharge. The ORs also increased by 3.020 or 2.681 times, respectively when the patients planned to discharge to their own house or their child's house compared to those who didn't have a plan for residence at discharge. The ORs increased by 1.922 or 2.250 times when the hospitals were owned by corporate or private property compared to publicly owned hospitals. The ORs decreased by 0.602 or 0.520 times when the hospital was sized over 400 beds or located in small and medium-sized cities compared to less than 200 bedded hospitals or located in metropolitan cities. Conclusion: The results of the study showed that the patients' and their family's willingness for discharge had a great impact on successful community discharge and the hospital-level factors played a significant role in it. Therefore, it is important to acknowledge and support long-term care hospitals to involve active in the patient discharge planning process.

• 응용통계연구
• /
• 제33권6호
• /
• pp.697-712
• /
• 2020

최근 20년간 세계적으로 새로운 전염병이 반복해서 등장해왔으며 코로나-19에 들어서는 일상에까지 큰 변화와 피해를 주고 있다. 이에 더해 앞으로도 새로운 전염병의 등장을 간과할 수 없게 되면서 경제 타격에 대응하기 위한 정책 발굴이 지속적으로 요구되고 있다. 이러한 상황에서 생활인구는 시민들의 생활 패턴 변화를 드러내는 중요한 지표이다. 본 논문에서는 코로나-19에 의한 일상의 변화를 유동인구 관점에서 감지 및 분류하여 시간적 및 사회환경적 특징을 분석한다. 시간 단위로 측정된 서울시 424개 행정동별 생활인구 데이터를 분류하기 위해 k-shape clustering을 사용하였고, 이후에는 각 군집에 개입분석, One-way ANOVA 등을 적용하여 코로나-19 진행 여파에 따른 군집별 특성 및 생활인구 변화 양상을 자세히 살펴보았다. 결론적으로 국내 코로나 환자 발생 전후의 인구 유출입 변동에 있어 각 군집별로 뚜렷한 특징을 확인하였으며, 코로나-19 관련 사건을 바탕으로 지정한 개입 시점에 대해서도 민감하게 반응하는 군집과 그렇지 않은 군집을 구분할 수 있었다.

• Asian Pacific Journal of Cancer Prevention
• /
• 제16권8호
• /
• pp.3543-3549
• /
• 2015

Background: Bladder cancer is one of the most common cancers worldwide. Gene expression profiling using microarray technologies improves the understanding of cancer biology. The aim of this study was to determine the gene expression profile in Egyptian bladder cancer patients. Materials and Methods: Samples from 29 human bladder cancers and adjacent non-neoplastic tissues were analyzed by cDNA microarray, with hierarchical clustering and multidimensional analysis. Results: Five hundred and sixteen genes were differentially expressed of which SOS1, HDAC2, PLXNC1, GTSE1, ULK2, IRS2, ABCA12, TOP3A, HES1, and SRP68 genes were involved in 33 different pathways. The most frequently detected genes were: SOS1 in 20 different pathways; HDAC2 in 5 different pathways; IRS2 in 3 different pathways. There were 388 down-regulated genes. PLCB2 was involved in 11 different pathways, MDM2 in 9 pathways, FZD4 in 5 pathways, p15 and FGF12 in 4 pathways, POLE2 in 3 pathways, and MCM4 and POLR2E in 2 pathways. Thirty genes showed significant differences between transitional cell cancer (TCC) and squamous cell cancer (SCC) samples. Unsupervised cluster analysis of DNA microarray data revealed a clear distinction between low and high grade tumors. In addition 26 genes showed significant differences between low and high tumor stages, including fragile histidine triad, Ras and sialyltransferase 8 (alpha) and 16 showed significant differences between low and high tumor grades, like methionine adenosyl transferase II, beta. Conclusions: The present study identified some genes, that can be used as molecular biomarkers or target genes in Egyptian bladder cancer patients.