• Journal of Periodontal and Implant Science
• /
• v.48 no.2
• /
• pp.114-123
• /
• 2018

Purpose: The aim of the current study was to develop a computer-assisted detection system based on a deep convolutional neural network (CNN) algorithm and to evaluate the potential usefulness and accuracy of this system for the diagnosis and prediction of periodontally compromised teeth (PCT). Methods: Combining pretrained deep CNN architecture and a self-trained network, periapical radiographic images were used to determine the optimal CNN algorithm and weights. The diagnostic and predictive accuracy, sensitivity, specificity, positive predictive value, negative predictive value, receiver operating characteristic (ROC) curve, area under the ROC curve, confusion matrix, and 95% confidence intervals (CIs) were calculated using our deep CNN algorithm, based on a Keras framework in Python. Results: The periapical radiographic dataset was split into training (n=1,044), validation (n=348), and test (n=348) datasets. With the deep learning algorithm, the diagnostic accuracy for PCT was 81.0% for premolars and 76.7% for molars. Using 64 premolars and 64 molars that were clinically diagnosed as severe PCT, the accuracy of predicting extraction was 82.8% (95% CI, 70.1%-91.2%) for premolars and 73.4% (95% CI, 59.9%-84.0%) for molars. Conclusions: We demonstrated that the deep CNN algorithm was useful for assessing the diagnosis and predictability of PCT. Therefore, with further optimization of the PCT dataset and improvements in the algorithm, a computer-aided detection system can be expected to become an effective and efficient method of diagnosing and predicting PCT.

• Korean Chemical Engineering Research
• /
• v.51 no.1
• /
• pp.87-92
• /
• 2013

Microalgae have been suggested as a promising feedstock for producing biofuel because of their potential of lipid production. In this study, a first principles ODE model for microalgae growth and neutral lipid synthesis proposed by Surisetty et al. (2010) is investigated for the purpose of maximizing the rate of microalgae growth and the amount of neutral lipid. The model has 6 states and 12 parameters and follows the assumption of Droop model which explains the growth as a two-step phenomenon; the uptake of nutrients is first occurred in the cell, and then use of intra-cellular nutrient to support cells growth. In this study, optimal input design using D-optimality criterion is performed to compute the system input profile and sensitivity analysis is also performed to determine which parameters have a negligible effect on the model predictions. Furthermore, model predictive control based on successive linearization is implemented to maximize the amount of neutral lipid contents.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.13
• /
• pp.5371-5375
• /
• 2014

Background: An accurate assessment of potential lymph node metastasis is important for the appropriate treatment of early gastric cancers. Therefore, this study analyzed predictive factors associated with lymph node metastasis and identified differences between mucosal and submucosal gastric cancers. Materials and Methods: A total of 518 early gastric cancer patients who underwent radical gastrectomy were reviewed in this study. Clinicopathological features were analyzed to identify predictive factors for lymph node metastasis. Results: The rate of lymph node metastasis in early gastric cancer was 15.3% overall, 3.3% for mucosal cancer, and 23.5% for submucosal cancer. Using univariate analysis, risk factors for lymph node metastasis were identified as tumor location, tumor size, depth of tumor invasion, histological type and lymphovascular invasion. Multivariate analysis revealed that tumor size >2 cm, submucosal invasion, undifferentiated tumors and lymphovascular invasion were independent risk factors for lymph node metastasis. When the carcinomas were confined to the mucosal layer, tumor size showed a significant correlation with lymph node metastasis. On the other hand, histological type and lymphovascular invasion were associated with lymph node metastasis in submucosal carcinomas. Conclusions: Tumor size >2 cm, submucosal tumor, undifferentiated tumor and lymphovascular invasion are predictive factors for lymph node metastasis in early gastric cancer. Risk factors are quite different depending on depth of tumor invasion. Endoscopic treatment might be possible in highly selective cases.

• Asian Pacific Journal of Cancer Prevention
• /
• v.17 no.4
• /
• pp.1961-1966
• /
• 2016

Background: To investigate the prevalence of premature ejaculation (PE) and erectile dysfunction (ED) in a sample of patients with prostate cancer and to determine the utility of the previously suggested cutoffs of the Premature Ejaculation Diagnostic Tool (PEDT) for the diagnosis of PE and that of International Index of Erectile Function (IIEF-5) for ED. Materials and Methods: A total of 1,202 men with prostate cancer were invited from urology clinics at the universities of Iran, Tehran, Qazvin, Ahvaz, Guilan and Tabriz. Clinical characteristics were collected through medical records. PE and ED diagnoses were made by trained urologists. In addition to the clinical diagnoses, PE and ED were measured through self-report using the PEDT and the IIEF-5. Questionnaire cutoff scores were determined using receiver operating characteristic (ROC) curves and confirmed by predictive ability using logistic regression. Results: The prevalence of PE was 63.7% and that of ED was 66.2%. Prevalences of PE decreased and that of ED increased with advanced TNM stages. According to ROC, the suggested cutoff for the PEDT to diagnose a PE was ${\geq}11$ (sensitivity=0.988, 1-specificity=0.084, and predictive ability=0.914) and ${\leq}17$ for the IIEF-5 (sensitivity=0.966, 1-specificity=0.031, and predictive ability=0.967). Conclusions: Prevalence of sexual problems was high in prostate cancer patients in Iran, therefore oncologists should take into account these potential problems when deciding on treatment modalities.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.22 no.2
• /
• pp.201-210
• /
• 1996

Predictive tests for the identification of contact sensitizing chemicals have been developed. We measured the sensitization potential with three predictive tests, the in vitro and the in vivo Local Lymph Node Assay(LLNA), ELISA to detect interferon-gamma(IFN-${\gamma}$) from supernatant and flow cytometry to detect change of cell surface proteins, using draining lymph nodes of mice. BALB/c mice were exposed to various chemicals or vehicles on the ears daily for 3 consecutive days in all experiments. With some exceptions of propyl paraben, neomycin sulfate, the in vivo LLNA was able to detect the sensitizing capacity of test chemicals and was more sensitive than the in vitro LLNA for chemicals used in the present study. In another experiment, contact sensitivity was assessed by the ELISA to detect IFN-Υ from the supernatants of the cultured LNCs after sensitization with chemicals. There was a good correlation between the LLNA and the IFN-Υ production for test chemicals. We also examined the change of cell surface proteins on LNCs after sensitization by flow cytometry for some cell adhesion molecules(ICAM-1, E-cadherine, B7 molecule), T cell markers(CD3, CD4, CD8, T$\alpha$$\beta$,T${\gamma}$$\delta$) and B cell markers(LR1, CD45R, I-Ad). The number of ICAM-1 positive cells and B cells in LNCs were increased after sensitization with DNCB, TNCB, isoeugenol and 25%, 50% cinnamic aldehyde compared with that of vehicle as a control. In conclusion, the in vivo LLNA could provide more sensitive screening test for moderate to strong sensitizers and some weak sensitizers including cosmetic raw materials than the in vitro LLNA. The production of IFN-Υ by allergen-activated LNCs might be a values indicators without radioisotopes for the identification of contact allergens. Detection of allergens by testing the increase of ICAM-1 positive cells and B cells in LNCs by flow cytometry might be used as a test method to detect allergens.

• Journal of Korean Society on Water Environment
• /
• v.39 no.6
• /
• pp.465-474
• /
• 2023

The chemical composition and molecular weight characteristics of dissolved organic matter (DOM) exert a profound influence on the efficiency of organic matter removal in water treatment systems, acting as efficiency predictive indicators. This research evaluated the primary chemical and molecular weight properties of DOM derived from diverse sources, including rivers, lakes, and biomasses, and assessed their relationship with the efficiency of coagulation/flocculation treatments. Dissolved organic carbon (DOC) removal efficiency through coagulation/flocculation exhibited significant correlations with DOM's hydrophobic distribution, the ratio of humic-like to protein-like fluorescence, and the molecular weight associated with humic substances (HS). These findings suggest that the DOC removal rate in coagulation/flocculation processes is enhanced by a higher presence of HS in DOM, an increased influence of externally sourced DOM, and more presence of high molecular weight compounds. The results of this study further posit that the efficacy of water treatment processes can be more accurately predicted when considering multiple DOM characteristics rather than relying on a singular trait. Based on major results from this study, a predictive model for DOC removal efficiency by coagulation/flocculation was formulated as: 24.3 - 7.83 × (fluorescence index) + 0.089 × (hydrophilic distribution) + 0.102 × (HS molecular weight). This proposed model, coupled with supplementary monitoring of influent organic matter, has the potential to enhance the design and predictive accuracy for coagulation/flocculation treatments targeting DOC removal in future applications.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.9
• /
• pp.4091-4094
• /
• 2014

Background: This study used the receiver operating characteristic curve (ROC) to analyze Surveillance, Epidemiology and End Results (SEER) bronchioaveolar carcinoma data to identify predictive models and potential disparity in outcomes. Materials and Methods: Socio-economic, staging and treatment factors were assessed. For the risk modeling, each factor was fitted by a Generalized Linear Model to predict cause specific survival. The area under the ROC was computed. Similar strata were combined to construct the most parsimonious models. A random sampling algorithm was used to estimate modeling errors. Risk of cause specific death was computed for the predictors for comparison. Results: There were 7,309 patients included in this study. The mean follow up time (S.D.) was 24.2 (20) months. Female patients outnumbered male ones 3:2. The mean (S.D.) age was 70.1 (10.6) years. Stage was the most predictive factor of outcome (ROC area of 0.76). After optimization, several strata were fused, with a comparable ROC area of 0.75. There was a 4% additional risk of death associated with lower county family income, African American race, rural residency and lower than 25% county college graduate. Radiotherapy had not been used in 2/3 of patients with stage III disease. Conclusions: There are socio-economic disparities in cause specific survival. Under-use of radiotherapy may have contributed to poor outcome. Improving education, access and rates of radiotherapy use may improve outcome.

• Physical Therapy Korea
• /
• v.23 no.4
• /
• pp.63-70
• /
• 2016

Background: The Pediatric Balance Scale (PBS) and the Fullerton Advanced Balance (FAB) scale were used to assess balance function in patients with balance problem. These multidimensional clinical balance scales provide information about potential risk factors for falls. Objects: The purpose of this study was to investigate and compare the predictive properties of the PBS and FAB scales relative to fall risk in children with cerebral palsy (CP) using a receiver operating characteristic analysis. Methods: In total, 49 children with CP (boy=21, girl=28) who were diagnosed with level 1 or 2 according to the Gross Motor Function Classification System participated in this study. The PBS and FAB were performed, and verified cut-off score, sensitivity, specificity, and the area of under the curve (AUC). Results: In this study, the PBS scale was as a predictive measure of fall risk, but the FAB was not significant in children with CP. A cut-off score of 45.5 points provided optimal sensitivity of .90 and specificity of .69 on the PBS, and a cut-off score of 21.5 points provided optimal sensitivity of .90 and specificity of .62 on the FAB. Both scales showed moderately accurate of AUC with .79 and .76, respectively. Conclusion: The PBS is a useful screening tool for predicting fall risk in children with cerebral palsy, and those who score 45.5 or lower indicate a high risk for falls and are in need of balance intervention.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.20-20
• /
• 2000

Back-propagation neural network (BPNN) is the most prevalently used paradigm in modeling semiconductor manufacturing processes, which as a neuron activation function typically employs a bipolar or unipolar sigmoid function in either hidden and output layers. In this study, applicability of another linear function as a neuron activation function is investigated. The linear function was operated in combination with other sigmoid functions. Comparison revealed that a particular combination, the bipolar sigmoid function in hidden layer and the linear function in output layer, is found to be the best combination that yields the highest prediction accuracy. For BPNN with this combination, predictive performance once again optimized by incrementally adjusting the gradients respective to each function. A total of 121 combinations of gradients were examined and out of them one optimal set was determined. Predictive performance of the corresponding model were compared to non-optimized, revealing that optimized models are more accurate over non-optimized counterparts by an improvement of more than 30%. This demonstrates that the proposed gradient-optimized teaming for BPNN with a linear function in output layer is an effective means to construct plasma models. The plasma modeled is a hemispherical inductively coupled plasma, which was characterized by a 24 full factorial design. To validate models, another eight experiments were conducted. process variables that were varied in the design include source polver, pressure, position of chuck holder and chroline flow rate. Plasma attributes measured using Langmuir probe are electron density, electron temperature, and plasma potential.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.2
• /
• pp.742-751
• /
• 2018

Electric vehicles (EV) are emerging as the future transportation vehicle reflecting their potential safe environmental advantages. Vehicle to Grid (V2G) system describes the hybrid system in which the EV can communicate with the utility grid and the energy flows with insignificant effect between the utility grid and the EV. The paper presents an optimal power control and energy management strategy for Plug-In Electric Vehicle (PEV) charging stations using Wind-PV-FC-Battery renewable energy sources. The energy management optimization is structured and solved using Multi-Objective Particle Swarm Optimization (MOPSO) to determine and distribute at each time step the charging power among all accessible vehicles. The Model-Based Predictive (MPC) control strategy is used to plan PEV charging energy to increase the utilization of the wind, the FC and solar energy, decrease power taken from the power grid, and fulfil the charging power requirement of all vehicles. Desired features for EV battery chargers such as the near unity power factor with negligible harmonics for the ac source, well-regulated charging current for the battery, maximum output power, high efficiency, and high reliability are fully confirmed by the proposed solution.