• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.15
• /
• pp.6285-6288
• /
• 2015

Background: Breast cancer is commonly diagnosed at late stages in countries with limited resources. In Morocco, breast cancer is ranked the first female cancer (36.1%) and screening methods could reduce the proportion presenting with a late diagnosis. Morocco is currently adopting a breast cancer screening program based on clinical examination at primary health facilities, diagnosis at secondary level and treatment at tertiary level. So far, there is no systematic information on the performance of the screening program for breast cancer in Morocco. The aim of this study was to analyze early performance indicators. Materials and Methods: A retrospective evaluative study conducted in Temara city. The target population was the entire female population aged between 45-70 years. The study was based on process and performance indicators collected at the individual level from the various health structures in Tamara between 2009 and 2011. Results: A total of 2,350 women participated in the screening program; the participation rate was 35.7%. Of these, 76.8% (1,806) were married and 5.2% (106) of this group had a family history of breast cancer. Of the women who attended screening, 9.3% (190) were found to have an abnormal physical examination findings. A total of 260 (12.7%) were referred for a specialist consultation. The positive predictive value of clinical breast examination versus mammography was 23.0%. Forty four (35.5%) of the lesions found on the mammograms were classified as BI-RADs 3; 4 or 5 category. Cancer was found in 4 (1.95%) of the total number of screened women and benign cases represented 0.58%. Conclusions: These first results of the programme are very encouraging, but there is a need to closely monitor performance and to improve programme procedures with the aim of increasing both the participation rate and the proportion of women eligible to attend screening.

• International journal of advanced smart convergence
• /
• v.12 no.3
• /
• pp.89-103
• /
• 2023

Obstructive sleep apnea (OSA) is one of the most prevalent sleep disorders that can lead to serious consequences, including hypertension and/or cardiovascular diseases, if not treated promptly. Continuous positive airway pressure (CPAP) is widely recognized as the most effective treatment for OSA, which needs the proper titration of airway pressure to achieve the most effective treatment results. However, the process of CPAP titration can be time-consuming and cumbersome. There is a growing importance in predicting personalized CPAP pressure before CPAP treatment. The primary objective of this study was to optimize the CPAP titration process for obstructive sleep apnea patients through EEG feature engineering with machine learning techniques. We aimed to identify and utilize the most critical EEG features to forecast key OSA predictive indicators, ultimately facilitating more precise and personalized CPAP treatment strategies. Here, we analyzed 126 OSA patients' PSG datasets before and after the CPAP treatment. We extracted 29 EEG features to predict the features that have high importance on the OSA prediction index which are AHI and SpO2 by applying the Shapley Additive exPlanation (SHAP) method. Through extracted EEG features, we confirmed the six EEG features that had high importance in predicting AHI and SpO2 using XGBoost, Support Vector Machine regression, and Random Forest Regression. By utilizing the predictive capabilities of EEG-derived features for AHI and SpO2, we can better understand and evaluate the condition of patients undergoing CPAP treatment. The ability to predict these key indicators accurately provides more immediate insight into the patient's sleep quality and potential disturbances. This not only ensures the efficiency of the diagnostic process but also provides more tailored and effective treatment approach. Consequently, the integration of EEG analysis into the sleep study protocol has the potential to revolutionize sleep diagnostics, offering a time-saving, and ultimately more effective evaluation for patients with sleep-related disorders.

• Journal of Preventive Medicine and Public Health
• /
• v.53 no.4
• /
• pp.256-265
• /
• 2020

Objectives: We compared the associations of 3 computed tomography (CT)-based abdominal adiposity indexes with non-alcoholic fatty liver disease (NAFLD) among middle-aged Korean men and women. Methods: The participants were 1366 men and 2480 women community-dwellers aged 30-64 years. Three abdominal adiposity indexes-visceral fat area (VFA), subcutaneous fat area (SFA), and visceral-to-subcutaneous fat ratio (VSR)-were calculated from abdominal CT scans. NAFLD was determined by calculating the Liver Fat Score from comorbidities and blood tests. An NAFLD prediction model that included waist circumference (WC) as a measure of abdominal adiposity was designated as the base model, to which VFA, SFA, and VSR were added in turn. The area under the receiver operating characteristic curve (AUC), integrated discrimination improvement (IDI), and net reclassification improvement (NRI) were calculated to quantify the additional predictive value of VFA, SFA, and VSR relative to WC. Results: VFA and VSR were positively associated with NAFLD in both genders. SFA was not significantly associated with NAFLD in men, but it was negatively associated in women. When VFA, SFA, and VSR were added to the WC-based NAFLD prediction model, the AUC improved by 0.013 (p<0.001), 0.001 (p=0.434), and 0.009 (p=0.007) in men and by 0.044 (p<0.001), 0.017 (p<0.001), and 0.046 (p<0.001) in women, respectively. The IDI and NRI were increased the most by VFA in men and VSR in women. Conclusions: Using CT-based abdominal adiposity indexes in addition to WC may improve the detection of NAFLD. The best predictive indicators were VFA in men and VSR in women.

• Geomechanics and Engineering
• /
• v.25 no.1
• /
• pp.1-16
• /
• 2021

Machine learning models have been widely used for landslide susceptibility assessment (LSA) in recent years. The large number of inputs or conditioning factors for these models, however, can reduce the computation efficiency and increase the difficulty in collecting data. Feature selection is a good tool to address this problem by selecting the most important features among all factors to reduce the size of the input variables. However, two important questions need to be solved: (1) how do feature selection methods affect the performance of machine learning models? and (2) which feature selection method is the most suitable for a given machine learning model? This paper aims to address these two questions by comparing the predictive performance of 13 feature selection-based machine learning (FS-ML) models and 5 ordinary machine learning models on LSA. First, five commonly used machine learning models (i.e., logistic regression, support vector machine, artificial neural network, Gaussian process and random forest) and six typical feature selection methods in the literature are adopted to constitute the proposed models. Then, fifteen conditioning factors are chosen as input variables and 1,017 landslides are used as recorded data. Next, feature selection methods are used to obtain the importance of the conditioning factors to create feature subsets, based on which 13 FS-ML models are constructed. For each of the machine learning models, a best optimized FS-ML model is selected according to the area under curve value. Finally, five optimal FS-ML models are obtained and applied to the LSA of the studied area. The predictive abilities of the FS-ML models on LSA are verified and compared through the receive operating characteristic curve and statistical indicators such as sensitivity, specificity and accuracy. The results showed that different feature selection methods have different effects on the performance of LSA machine learning models. FS-ML models generally outperform the ordinary machine learning models. The best FS-ML model is the recursive feature elimination (RFE) optimized RF, and RFE is an optimal method for feature selection.

• Journal of Gastric Cancer
• /
• v.19 no.2
• /
• pp.183-192
• /
• 2019

Purpose: Due to adverse events, dose reduction or withdrawal of adjuvant chemotherapy is required for some patients. To identify the predictive factors for tolerability to postoperative adjuvant S-1 monotherapy in gastric cancer (GC) patients, we evaluated the predictive values of blood indicators. Materials and Methods: We analyzed 98 patients with pStage II/III GC who underwent postoperative adjuvant S-1 monotherapy. We retrospectively analyzed correlations between 14 parameters obtained from perioperative routine blood tests to assess their influence on the withdrawal of postoperative adjuvant S-1 monotherapy, within 6 months after discontinuation. Results: Postoperative adjuvant chemotherapy was discontinued in 21 patients (21.4%) within 6 months. Univariable analysis revealed that high preoperative albumin-bilirubin (ALBI) scores had the highest odds ratio (OR) for predicting the failure of adjuvant S-1 chemotherapy (OR, 6.47; 95% confidence interval [CI], 2.08-20.1; cutoff value, -2.696). The high ALBI group had a significantly shorter time to failure of postoperative adjuvant S-1monotherapy (hazard ratio, 3.48; 95% CI, 1.69-7.25; P=0.001). Multivariable analysis identified high preoperative ALBI score as an independent prognostic factor for tolerability (OR, 10.3; 95% CI, 2.33-45.8; P=0.002). Conclusions: Preoperative ALBI shows promise as an indicator associated with the tolerability of adjuvant S-1 monotherapy in patients with pStage II/III GC.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.5
• /
• pp.825-832
• /
• 2022

In this paper, we discuss the design and implementation of predictive and diagnostic models for realizing intelligent predictive models by collecting and storing the power output of agricultural photovoltaic power generation systems. Our model predicts the amount of photovoltaic power generation using RNN, LSTM, and GRU models, which are recurrent neural network techniques specialized for time series data, and compares and analyzes each model with different hyperparameters, and evaluates the performance. As a result, the MSE and RMSE indicators of all three models were very close to 0, and the R2 indicator showed performance close to 1. Through this, it can be seen that the proposed prediction model is a suitable model for predicting the amount of photovoltaic power generation, and using this prediction, it was shown that it can be utilized as an intelligent and efficient O&M function in an agricultural photovoltaic system.

• International Journal of Computer Science & Network Security
• /
• v.23 no.8
• /
• pp.17-25
• /
• 2023

The alarming global prevalence of Type 2 Diabetes Mellitus (T2DM) has catalyzed an urgent need for robust, early diagnostic methodologies. This study unveils a pioneering approach to predicting T2DM, employing the Extreme Gradient Boosting (XGBoost) algorithm, renowned for its predictive accuracy and computational efficiency. The investigation harnesses a meticulously curated dataset of 4303 samples, extracted from a comprehensive Chinese research study, scrupulously aligned with the World Health Organization's indicators and standards. The dataset encapsulates a multifaceted spectrum of clinical, demographic, and lifestyle attributes. Through an intricate process of hyperparameter optimization, the XGBoost model exhibited an unparalleled best score, elucidating a distinctive combination of parameters such as a learning rate of 0.1, max depth of 3, 150 estimators, and specific colsample strategies. The model's validation accuracy of 0.957, coupled with a sensitivity of 0.9898 and specificity of 0.8897, underlines its robustness in classifying T2DM. A detailed analysis of the confusion matrix further substantiated the model's diagnostic prowess, with an F1-score of 0.9308, illustrating its balanced performance in true positive and negative classifications. The precision and recall metrics provided nuanced insights into the model's ability to minimize false predictions, thereby enhancing its clinical applicability. The research findings not only underline the remarkable efficacy of XGBoost in T2DM prediction but also contribute to the burgeoning field of machine learning applications in personalized healthcare. By elucidating a novel paradigm that accentuates the synergistic integration of multifaceted clinical parameters, this study fosters a promising avenue for precise early detection, risk stratification, and patient-centric intervention in diabetes care. The research serves as a beacon, inspiring further exploration and innovation in leveraging advanced analytical techniques for transformative impacts on predictive diagnostics and chronic disease management.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.3
• /
• pp.531-538
• /
• 2024

While the importance of the water quality environment is being emphasized, the water quality index for improving the water quality of urban rivers in Gwangju Metropolitan City is an important factor affecting the aquatic ecosystem and requires accurate prediction. In this paper, the XGBoost and LightGBM machine learning algorithms were used to compare the performance of the water quality inspection items of the downstream Pyeongchon Bridge and upstream BanghakBr_Gwangjucheon1 water systems, which are important points of Gwangju Stream, as a result of statistical verification, three water quality indicators, Nitrogen(TN), Nitrate(NO3), and Ammonia amount(NH3) were predicted, and the performance of the predictive model was evaluated by using RMSE, a regression model evaluation index. As a result of comparing the performance after cross-validation by implementing individual models for each water system, the XGBoost model showed excellent predictive ability.

• The Korean Journal of Applied Statistics
• /
• v.33 no.4
• /
• pp.483-497
• /
• 2020

We used a deep neural network model for the prediction of the stock prices of Kia Motors and Shinsegae as listed in the KOSPI 100. We used an emotional variable derived from online news in addition to the various technical indicators most often used. The emotional variable used as a predictor variable was generated from the average of the emotional scores for companies in the industrial group after building an emotional dictionary specific to each industrial group classified in a social network analysis. The study was conducted with various combinations of predictors and confirmed that good predictive and profitable power could be expected when jointly using technical indicators and an emotional variable based on online news by industrial groups.

• Journal of Periodontal and Implant Science
• /
• v.47 no.4
• /
• pp.240-250
• /
• 2017

Purpose: The aim of this study was to analyze the prevalence and risk indicators of peri-implantitis in Korean patients with history of periodontal disease. Methods: A total of 444 patients with 1,485 implants were selected from patients who had been treated at the Department of Periodontology, Chonbuk National University Dental Hospital between July 2014 and June 2015. A group with a history of peri-implantitis (HP) (370 patients with 1,189 implants) and a group with a current peri-implantitis (CP) (318 patients with 1,004 implants) were created based on the radiographic and clinical assessments of implants. The prevalence of peri-implantitis was calculated at both the patient and implant levels. The influence of risk variables on the occurrence of peri-implantitis was analyzed using generalized estimating equations analysis. Results: The prevalence of peri-implantitis in the HP and CP groups ranged from 6.7% to 19.7%. The cumulative peri-implantitis rate in the HP group estimated with the Kaplan-Meier method was higher than that in the CP group over the follow-up period. Among the patient-related risk variables, supportive periodontal therapy (SPT) was the only significant risk indicator for the occurrence of peri-implantitis in both groups. In the analysis of implant-related variables, implants supporting fixed dental prosthesis (FDP) and implants with subjective discomfort was associated with a higher prevalence of peri-implantitis than single implants and implants without subjective discomfort in the HP group. The presence of subjective discomfort was the only significant implant-related variable predictive of peri-implantitis in the CP group. Conclusions: Within the limitations of this study, the prevalence of peri-implantitis in Korean patients with a history of periodontal disease was similar to that reported in other population samples. Regular SPT was important for preventing peri-implantitis. Single implants were found to be less susceptible to peri-implantitis than those supporting FDP. Patients' subjective discomfort was found to be a strong risk indicator for peri-implantitis.