• Journal of Korea Artificial Intelligence Association
• /
• 제1권1호
• /
• pp.11-16
• /
• 2023

Accurate hospital case modeling and prediction are crucial for efficient healthcare. In this study, we demonstrate the implementation of regression analysis methods in machine learning systems utilizing mathematical statics and machine learning techniques. The developed machine learning model includes Bayesian linear, artificial neural network, decision tree, decision forest, and linear regression analysis models. Through the application of these algorithms, corresponding regression models were constructed and analyzed. The results suggest the potential of leveraging machine learning systems for medical research. The experiment aimed to create an Azure Machine Learning Studio tool for the speedy evaluation of multiple regression models. The tool faciliates the comparision of 5 types of regression models in a unified experiment and presents assessment results with performance metrics. Evaluation of regression machine learning models highlighted the advantages of boosted decision tree regression, and decision forest regression in hospital case prediction. These findings could lay the groundwork for the deliberate development of new directions in medical data processing and decision making. Furthermore, potential avenues for future research may include exploring methods such as clustering, classification, and anomaly detection in healthcare systems.

• 한국응급구조학회지
• /
• 제15권3호
• /
• pp.57-70
• /
• 2011

Purpose : The purpose of this study was to reveal some factors of ROSC & survival for cardiac arrests from sport & leisure activities(CASLs). Methods : A retrospective study of the 1,341 out of hospital cardiac arrests(OHCAs) treated by EMS in Gyeonggi Provincial Fire and Disaster Headquarters from January to December in 2008 was conducted. The primary end-point was admission to emergency room. To clarify the factors through comparison of CASLs(n=58) with ROSCs & survivals(n=58), Answer Tree analysis for data mining with the CHAID algorithm was performed and alpha was set at .05. Mean, median, and percentile of time intervals, distances, and age on the 58 CASLs, 75 ROSCs, and 27 survivals(patients admitted to emergency room) were analysed. Results : Fourteen CASLs(24.1%), 41 ROSCs(54.7%), 16 survivals(59.3%) were treated with CPR within 5 min., and only 2 CASLs(3.4%), 11 ROSCs(14.7%), 10 survivals(37.0%) were treated with defilbrillation within 10 min. from arrest. If time recording from arrest to defilbrillation, the patients were classified 81.0%($X^2=9.83$, p=.005) into ROSCs & survivals. And the patients with no history, 100.0%($X^2=5.44$, p=.020). The other patients with no intention, 87.5%($X^2=7.00$, p=.024). Whereas the other patients with intention, treated with CPR after 4 min. from arrest were classified 67.2%($X^2=3.99$, p=.046) into CASLs. Conclusion : CPR within 4 minutes was the most important factor that discriminates between CASLs and ROSCs & survivals to record cardiac arrests-defilbrillation time. CPR within 4 min. from arrest, no history, and no intention were factors for improved ROSC & survival.

• Asian Pacific Journal of Cancer Prevention
• /
• 제17권2호
• /
• pp.835-838
• /
• 2016

Background: Lung cancer remains one of the most common cancers in the world, both in terms of new cases (about 13% of total per year) and deaths (nearly one cancer death in five), because of the high case fatality. Errors in lung cancer type or malignant growth determination lead to degraded treatment efficacy, because anticancer strategy depends on tumor morphology. Materials and Methods: We have made an attempt to evaluate effectiveness of machine learning algorithms in the task of lung cancer classification based on gene expression levels. We processed four publicly available data sets. The Dana-Farber Cancer Institute data set contains 203 samples and the task was to classify four cancer types and sound tissue samples. With the University of Michigan data set of 96 samples, the task was to execute a binary classification of adenocarcinoma and non-neoplastic tissues. The University of Toronto data set contains 39 samples and the task was to detect recurrence, while with the Brigham and Women's Hospital data set of 181 samples it was to make a binary classification of malignant pleural mesothelioma and adenocarcinoma. We used the k-nearest neighbor algorithm (k=1, k=5, k=10), naive Bayes classifier with assumption of both a normal distribution of attributes and a distribution through histograms, support vector machine and C4.5 decision tree. Effectiveness of machine learning algorithms was evaluated with the Matthews correlation coefficient. Results: The support vector machine method showed best results among data sets from the Dana-Farber Cancer Institute and Brigham and Women's Hospital. All algorithms with the exception of the C4.5 decision tree showed maximum potential effectiveness in the University of Michigan data set. However, the C4.5 decision tree showed best results for the University of Toronto data set. Conclusions: Machine learning algorithms can be used for lung cancer morphology classification and similar tasks based on gene expression level evaluation.

• The Journal of Advanced Prosthodontics
• /
• 제10권6호
• /
• pp.395-400
• /
• 2018

PURPOSE. This study tried to find the most significant factors predicting implant prognosis using machine learning methods. MATERIALS AND METHODS. The data used in this study was based on a systematic search of chart files at Seoul National University Bundang Hospital for one year. In this period, oral and maxillofacial surgeons inserted 667 implants in 198 patients after consultation with a prosthodontist. The traditional statistical methods were inappropriate in this study, which analyzed the data of a small sample size to find a factor affecting the prognosis. The machine learning methods were used in this study, since these methods have analyzing power for a small sample size and are able to find a new factor that has been unknown to have an effect on the result. A decision tree model and a support vector machine were used for the analysis. RESULTS. The results identified mesio-distal position of the inserted implant as the most significant factor determining its prognosis. Both of the machine learning methods, the decision tree model and support vector machine, yielded the similar results. CONCLUSION. Dental clinicians should be careful in locating implants in the patient's mouths, especially mesio-distally, to minimize the negative complications against implant survival.

• Tuberculosis and Respiratory Diseases
• /
• 제56권2호
• /
• pp.198-202
• /
• 2004

저자들은 만성기침을 주소로 내원한 2명의 40대 여자 환자에서 전산화단층촬영과 기관지내시경을 통해 주기관지의 기관골형성증 2예를 경험하였기에 문헌 고찰과 함께 보고하는 바이다.

• Asian Pacific Journal of Cancer Prevention
• /
• 제14권5호
• /
• pp.3315-3318
• /
• 2013

Gall bladder carcinoma is the most common cancer of biliary tree, characterized by rapid progression and a very high mortality rate. Detection at an early stage, however, is indicative of a very good prognosis and prolonged survival. The practice of histopathological examination of gall bladder specimens removed for clinically benign conditions and its usefulness has been a subject of controversy. The present prospective study was carried out over a period of four years in order to find out the incidence of unsuspected gallbladder carcinoma in cholecystectomy specimens received in our histopathology laboratory and to analyze their clinico-pathological features. A total of 4,115 cases were examined. Incidentally detected cases comprised 0.44%, which accounted for 72% of all gall bladder carcinomas detected. The majority were in an early, surgically resectable stage. From the results of this study we recommend that in India and other countries with relatively high incidences of gall bladder carcinoma, all cholecystectomy specimens should be submitted to histopathology laboratory, as this is the only means by which malignancies can be detected at an early, potentially curable stage.

• 경영정보학연구
• /
• 제9권1호
• /
• pp.45-65
• /
• 2007

다른 산업과 달리 병원/의료 산업에서는 건강 보험료 심사 평가라는 독특한 검증 과정이 필수적으로 있게 된다. 건강 보험료 심사 평가는 병원의 수익 문제 뿐 아니라 적정한 진료행위를 하는 병원이라는 이미지와도 맞물려 매우 중요한 분야이며, 특히 대형 종합병원일수록 이 부분에 많은 심사관련 인력들을 투입하여, 병원의 수익과 명예를 위해서 업무를 수행하고 있다. 본 논문은 이러한 건강보험료 청구 심사 과정에서, 사전에 수많은 진료 청구 건 중 심사 평가에서 삭감이 될 수 있는 진료 청구 건을 데이터 마이닝을 통해서 발견하여, 사전의 대비를 철저히 하고자 하는 한 국내 대형 종합병원의 사례를 소개하고자 한다. 데이터 마이닝을 적용함에 있어, 주요한 문제점 중 하나는 바로 지도학습 기법을 적용하기에 곤란한 데이터 불균형 문제가 발생하는 것이다. 이런 불균형 문제를 해소하고, 비교 조건 중에 가장 효율적인 삭감 예상 진료 건 탐지 모델을 만들어 내기 위하여, 데이터 불균형 문제의 기본 해법인 Sampling과 오분류 비용의 다양한 혼합적인 적용을 통하여, 적합한 조건을 가지는 의사결정 나무 모델을 도출하였다.

• Asian Pacific Journal of Cancer Prevention
• /
• 제14권12호
• /
• pp.7601-7605
• /
• 2013

Background: Gambogenic acid is a major active compound of gamboge which exudes from the Garcinia hanburyi tree. Gambogenic acid anti-cancer activity in vitro has been reported in several studies, including an A549 nude mouse model. However, the mechanisms of action remain unclear. Methods: We used nude mouse models to detect the effect of gambogenic acid on breast tumors, analyzing expression of apoptosis-related proteins in vivo by Western blotting. Effects on cell proliferation, apoptosis and apoptosis-related proteins in MDA-MB-231 cells were detected by MTT, flow cytometry and Western blotting. Inhibitors of caspase-3,-8,-9 were also used to detect effects on caspase family members. Results: We found that gambogenic acid suppressed breast tumor growth in vivo, in association with increased expression of Fas and cleaved caspase-3,-8,-9 and bax, as well as decrease in the anti-apoptotic protein bcl-2. Gambogenic acid inhibited cell proliferation and induced cell apoptosis in a concentration-dependent manner. Conclusion: Our observations suggested that Gambogenic acid suppressed breast cancer MDA-MB-231 cell growth by mediating apoptosis through death receptor and mitochondrial pathways in vivo and in vitro.

• Journal of Trauma and Injury
• /
• 제30권4호
• /
• pp.216-219
• /
• 2017

Laparoscopy has been one of the most effective modalities in various surgical situations, although its use in trauma patients has some limitations. The benefits of laparoscopy include cost-effectiveness, shorter length of hospital stay, and less postoperative pain. This report describes diagnostic laparoscopy and laparoscopic diverting sigmoid loop colostomy in penetrating extraperitoneal rectal injury. A 41-year-old male presented with perineal pain following penetrating trauma caused by a tree limb. Computed tomography showed air density in the perirectal space and retroperitoneum. As his vital signs were stable, we performed diagnostic laparoscopy and confirmed no intraperitoneal perforation. Therefore, laparoscopic diverting sigmoid loop colostomy was performed. He was discharged without any complications despite underlying hepatitis C-related cirrhosis. Colostomy closure was performed 3 months later.

• Communications for Statistical Applications and Methods
• /
• 제23권1호
• /
• pp.1-20
• /
• 2016

Causal inference methodologies have been developed for the past decade to estimate the unconfounded effect of an exposure under several key assumptions. These assumptions include, but are not limited to, the stable unit treatment value assumption, the strong ignorability of treatment assignment assumption, and the assumption that propensity scores be bounded away from zero and one (the positivity assumption). Of these assumptions, the first two have received much attention in the literature. Yet the positivity assumption has been recently discussed in only a few papers. Propensity scores of zero or one are indicative of deterministic exposure so that causal effects cannot be defined for these subjects. Therefore, these subjects need to be removed because no comparable comparison groups can be found for such subjects. In this paper, using currently available causal inference methods, we evaluate the effect of arbitrary cutoffs in the distribution of propensity scores and the impact of those decisions on bias and efficiency. We propose a tree-based method that performs well in terms of bias reduction when the definition of positivity is based on a single confounder. This tree-based method can be easily implemented using the statistical software program, R. R code for the studies is available online.