• 한국인공지능학회지
• /
• 제6권1호
• /
• pp.16-20
• /
• 2018

The development of modern information technology has increased the amount of big data about patients' information and diseases. In this study, we developed a prediction model of diabetes using the health examination data provided by the public data portal in 2016. In addition, we graphically visualized diabetes incidence by sex, age, residence area, and income level. As a result, the incidence of diabetes was different in each residence area and income level, and the probability of accurately predicting male and female was about 65%. In addition, it can be confirmed that the influence of X on male and Y on female is highly to affect diabetes. This predictive model can be used to predict the high-risk patients and low-risk patients of diabetes and to alarm the serious patients, thereby dramatically improving the re-admission rate. Ultimately it will be possible to contribute to improve public health and reduce chronic disease management cost by continuous target selection and management.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권9호
• /
• pp.2904-2926
• /
• 2022

Currently, diabetes is the most common chronic disease in the world, affecting 23.7% of the population in the Kingdom of Saudi Arabia. Diabetes may be the cause of lower-limb amputations, kidney failure and blindness among adults. Therefore, diagnosing the disease in its early stages is essential in order to save human lives. With the revolution in technology, Artificial Intelligence (AI) could play a central role in the early prediction of diabetes by employing Machine Learning (ML) technology. In this paper, we developed a diagnosis system using machine learning models for the detection of type 2 diabetes among adults, through the adoption of two different diabetes datasets: one for training and the other for the testing, to analyze and enhance the prediction accuracy. This work offers an enhanced classification accuracy as a result of employing several pre-processing methods before applying the ML models. According to the obtained results, the implemented Random Forest (RF) classifier offers the best classification accuracy with a classification score of 98.95%.

• Genomics & Informatics
• /
• 제14권4호
• /
• pp.138-148
• /
• 2016

The success of genome-wide association studies (GWASs) has enabled us to improve risk assessment and provide novel genetic variants for diagnosis, prevention, and treatment. However, most variants discovered by GWASs have been reported to have very small effect sizes on complex human diseases, which has been a big hurdle in building risk prediction models. Recently, many statistical approaches based on penalized regression have been developed to solve the "large p and small n" problem. In this report, we evaluated the performance of several statistical methods for predicting a binary trait: stepwise logistic regression (SLR), least absolute shrinkage and selection operator (LASSO), and Elastic-Net (EN). We first built a prediction model by combining variable selection and prediction methods for type 2 diabetes using Affymetrix Genome-Wide Human SNP Array 5.0 from the Korean Association Resource project. We assessed the risk prediction performance using area under the receiver operating characteristic curve (AUC) for the internal and external validation datasets. In the internal validation, SLR-LASSO and SLR-EN tended to yield more accurate predictions than other combinations. During the external validation, the SLR-SLR and SLR-EN combinations achieved the highest AUC of 0.726. We propose these combinations as a potentially powerful risk prediction model for type 2 diabetes.

• Journal of Obesity & Metabolic Syndrome
• /
• 제27권4호
• /
• pp.213-222
• /
• 2018

Bariatric surgery has evolved from a surgical measure for treating morbid obesity to an epochal remedy for treating metabolic syndrome as a whole, which is represented by type 2 diabetes mellitus. Numerous clinical trials have advocated bariatric or metabolic surgery over nonsurgical interventions because of markedly superior metabolic outcomes in morbidly obese patients who satisfy traditional criteria for bariatric surgery (body mass index [BMI] >$35kg/m^2$) and in less obese or simply overweight patients. Nevertheless, not all diabetes patients achieve the most desirable outcomes; i.e., diabetes remission after metabolic surgery. Thus, candidates for metabolic surgery should be carefully selected based on comprehensive preoperative assessments of the risk-benefit ratio. Predictors for diabetes remission after metabolic surgery may be classified into two groups based on mechanism of action. The first is indices for preserved pancreatic beta-cell function, including younger age, shorter duration of diabetes, and higher C-peptide level. The second is the potential for an insulin resistance reduction, including higher baseline BMI and visceral fat area. Several prediction models for diabetes remission have been suggested by merging these two to guide the joint decision-making process between clinicians and patients. Three such models, DiaRem, ABCD, and individualized metabolic surgery scores, provide an intuitive scoring system and have been validated in an independent external cohort and can be utilized in routine clinical practice. These prediction models need further validation in various ethnicities to ensure universal applicability.

• 대한지역사회영양학회지
• /
• 제4권3호
• /
• pp.421-430
• /
• 1999

A neural network system was applied in order to analyze the nutritional and other factors influencing chronic diseases. Five different nutrition evaluation methods including SD Score, %RDA, NAR INQ and %RDA-SD Score were utilized to facilitate nutrient data for the system. Observing top three chronic disease prediction ratio, WHR using SD Score was the most frequently quoted factor revealing the highest predication rate as 62.0%. Other high prediction rates using other data processing methods are as follows. Prediction rate with %RDA, NAR, INQ and %RDA-SD Score were 58.5%(diabetes), 53.5%(hyperlipidemia), 51.6%(diabetes), and 58.0%(diabetes)respectively. Higher prediction rate was observed using either NAR or INQ for obesity as 51.7% and 50.9% compared to the previous result using SD Score. After reviewing appearance rate for all chronic disease and for various data processing method used, it was found that iron and vitamin C were the most frequently cited factors resulting in high prediction rate.

• 한국전자통신학회논문지
• /
• 제18권5호
• /
• pp.999-1008
• /
• 2023

당뇨병과 같은 만성 질환의 조기 예측은 중요한 이슈이며, 그중에서도 당뇨 예측의 정확도 향상은 매우 중요하다. 당뇨 예측을 위한 다양한 기계 학습 및 딥 러닝 기반 방법론을 도입하고 있으나, 이러한 기술들은 다른 방법론보다 더 우수한 성능을 위해 대량의 데이터를 필요로 하며, 복잡한 데이터 모델 때문에 학습 비용이 높다. 본 연구에서는 pima 데이터셋과 k-fold 교차 검증을 사용한 DNN이 당뇨 진단 모델의 효율성을 감소시킨다는 주장을 검증하고자 한다. 의사 결정 트리, SVM, 랜덤 포레스트, 로지스틱 회귀, KNN 및 다양한 앙상블 기법과 같은 기계 학습 분류 방법을 사용하여 어떤 알고리즘이 최상의 예측 결과를 내는지 결정하였다. 모든 분류 모델에 대한 훈련 및 테스트 후 제안된 시스템은 ADASYN 방법과 함께 XGBoost 분류기에서 최상의 결과를 제공하였으며, 정확도는 81%, F1 계수는 0.81, AUC는 0.84였다. 또한 도메인 적응 방법이 제안된 시스템의 다양성을 보여주기 위해 구현되었다. LIME 및 SHAP 프레임워크를 사용한 설명 가능한 AI 접근 방식이 모델이 최종 결과를 어떻게 예측하는지 이해하기 위해 구현되었다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2023년도 춘계학술발표대회
• /
• pp.325-327
• /
• 2023

Diabetes is a common chronic disease that threatens human life and health, and its prevalence remains high because its mechanisms are complex, further its etiology remains unclear. According to the International Diabetes Federation (IDF), there are 463 million cases of diabetes in adults worldwide, and the number is growing. This study aims to explore the potential influencing factors of diabetes by learning data from the UCI diabetes dataset, which is a multivariate time series dataset. In this paper we propose the TCN-prophet model for diabetes. The experimental results show that the prediction of insulin concentration by the TCN-prophet model provides a high degree of consistency, compared to the existing LSTM model.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 춘계학술대회 학술발표 논문집
• /
• pp.67-70
• /
• 1998

There were many cases to apply artificial intelligence to medicine. In this paper, we present the prediction model of the development of the NIDDM(noninsulin-dependent diabetes mellitus). It is not difficult that doctor diagnose patient as DM(diabetes mellitus). However NIDDM is usually developmented later on 40 years old and symptom appeares gradually. So screening test or prediction model is needed absolutely. Our model predicts development of NIDDM with still normal data 2 year ago. Prediction models developed are both MLP(multilayer perceptron) with backpropagation training and RBFN(radial basis function network). Performance of both models were evaluated with likelihood ratio. MLP was about two and RBFN was about three. We expect that models developed can prevent development of DM and utilize normal data.

• Journal of Electrical Engineering and information Science
• /
• 제3권3호
• /
• pp.355-365
• /
• 1998

The purposes of this study are to propose a network for the characterizing of the input data and to show how to design predictive neural net재가 expert system which doesn't need previous knowledge base. We derived this network from the radial basis function networks(RBFN), and named it as a statistical EBFN. The proposed network can replace the statistical methods for analyzing dynamic relations between target disease and other parameters in medical studies. We compared statistical RBFN with the probabilistic neural network(PNN) and fuzzy logic(FL). And we testified our method in the diabetes prediction and compared our method with the well-known multilayer perceptron(MLP) neural network one, and showed good performance of our network. At last, we developed the diabetes prediction expert system based on the proposed statistical RBFN without previous knowledge base. Not only the applicability of the characterizing of parameters related to diabetes and construction of the diabetes prediction expert system but also wide applicabilities has the proposed statistical RBFN to other similar problems.

• 한국정보통신학회논문지
• /
• 제25권10호
• /
• pp.1296-1301
• /
• 2021

최근 전 세계적으로 당뇨병 유발률이 증가함에 따라 다양한 머신러닝과 딥러닝 기술을 통해 당뇨병을 예측하려고 는 연구가 이어지고 있다. 본 연구에서는 독일의 Frankfurt Hospital 데이터로 머신러닝 기법을 활용하여 당뇨병을 예측하는 모델을 제시한다. IQR(Interquartile Range) 기법을 이용한 이상치 처리와 피어슨 상관관계 분석을 적용하고 Decision Tree, Random Forest, Knn, SVM, 앙상블 기법인 XGBoost, Voting, Stacking로 모델별 당뇨병 예측 성능을 비교한다. 연구를 진행한 결과 Stacking ensemble 기법의 정확도가 98.75%로 가장 뛰어난 성능을 보였다. 따라서 해당 모델을 이용하여 현대 사회에 만연한 당뇨병을 정확히 예측하고 예방할 수 있다는 점에서 본 연구는 의의가 있다.