• 대한의용생체공학회:의공학회지
• /
• 제33권3호
• /
• pp.148-154
• /
• 2012

Hemorrhagic shock is a primary cause of deaths resulting from injury in the world. Although many studies have tried to diagnose accurately hemorrhagic shock in the early stage, such attempts were not successful due to compensatory mechanisms of humans. The objective of this study was to construct a survival prediction model of rats in acute hemorrhagic shock using a random forest (RF) model. Heart rate (HR), mean arterial pressure (MAP), respiration rate (RR), lactate concentration (LC), and peripheral perfusion (PP) measured in rats were used as input variables for the RF model and its performance was compared with that of a logistic regression (LR) model. Before constructing the models, we performed 5-fold cross validation for RF variable selection, and forward stepwise variable selection for the LR model to examine which variables were important for the models. For the LR model, sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (ROC-AUC) were 0.83, 0.95, 0.88, and 0.96, respectively. For the RF models, sensitivity, specificity, accuracy, and AUC were 0.97, 0.95, 0.96, and 0.99, respectively. In conclusion, the RF model was superior to the LR model for survival prediction in the rat model.

• Bulletin of the Korean Chemical Society
• /
• 제31권8호
• /
• pp.2163-2169
• /
• 2010

G-protein coupled receptors (GPCRs) are involved in a wide variety of physiological processes and are known to be targets for nearly 50% of drugs. The various functions of GPCRs are affected by their cognate ligands which are mainly classified as agonists and antagonists. The purpose of this study is to develop a Bayesian classification model, that can predict a compound as either human GPCR agonist or antagonist. Total 6627 compounds experimentally determined as either GPCR agonists or antagonists covering all the classes of GPCRs were gathered to comprise the dataset. This model distinguishes GPCR agonists from GPCR antagonists by using chemical fingerprint, FCFP_6. The model revealed distinctive structural characteristics between agonistic and antagonistic compounds: in general, 1) GPCR agonists were flexible and had aliphatic amines, and 2) GPCR antagonists had planar groups and aromatic amines. This model showed very good discriminative ability in general, with pretty good discriminant statistics for the training set (accuracy: 90.1%) and a good predictive ability for the test set (accuracy: 89.2%). Also, receiver operating characteristic (ROC) plot showed the area under the curve (AUC) to be 0.957, and Matthew's Correlation Coefficient (MCC) value was 0.803. The quality of our model suggests that it could aid to classify the compounds as either GPCR agonists or antagonists, especially in the early stages of the drug discovery process.

• Asian Pacific Journal of Cancer Prevention
• /
• 제17권1호
• /
• pp.323-333
• /
• 2016

Background: Ovarian cancer remains a major worldwide health care issue due to the lack of satisfactory diagnostic methods for early detection of the disease. Prior studies on the role of serum cancer antigen 125 (CA125) and human epididymis protein 4 (HE4) in detecting ovarian cancer presented conflicting results. New tools to improve the accuracy of identifying malignancy are urgently needed. We here aimed to evaluate the diagnostic utility of tissue CA125 and HE4 gene expression in comparison to serum CA125 and HE4 in discriminating benign from malignant pelvic masses. Materials and Methods: One-hundred Egyptian women were enrolled in this study, including 60 epithelial ovarian cancer (EOC) patients and 20 benign ovarian tumor patients, as well as 20 apparently healthy women. Preoperative serum levels of CA125 and HE4 were measured by immunoassays. Tissue expression levels of genes encoding CA125 and HE4 were determined by quantitative real time polymerase chain reaction (qRT-PCR). The diagnostic performance of CA125 and HE4, measured either as mRNA or protein levels, was evaluated by receiver operating characteristic (ROC) curves. Results: The serum CA125+HE4 combination and serum HE4, with area under the curve (AUC) values of 0.935 and 0.932, respectively, performed significantly better than serum CA125 (AUC=0.592; P<0.001). Tissue CA125 and HE4 (AUC=1) performed significantly better than serum CA125 (P<0.001), serum HE4 (P=0.016) and the serum CA125+HE4 combination (P=0.018). Conclusions: Measurement of tissue CA125 and HE4 gene expression not only improves discriminatory performance, but also broadens the range of differential diagnostic possibilities in distinguishing EOC from benign ovarian tumors.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권13호
• /
• pp.5349-5353
• /
• 2014

Statistical methods to analyze and predict the related risk factors of nosocomial infection in lung cancer patients are various, but the results are inconsistent. A total of 609 patients with lung cancer were enrolled to allow factor comparison using Student's t-test or the Mann-Whitney test or the Chi-square test. Variables that were significantly related to the presence of nosocomial infection were selected as candidates for input into the final ANN model. The area under the receiver operating characteristic (ROC) curve (AUC) was used to evaluate the performance of the artificial neural network (ANN) model and logistic regression (LR) model. The prevalence of nosocomial infection from lung cancer in this entire study population was 20.1% (165/609), nosocomial infections occurring in sputum specimens (85.5%), followed by blood (6.73%), urine (6.0%) and pleural effusions (1.82%). It was shown that long term hospitalization (${\geq}22days$, P= 0.000), poor clinical stage (IIIb and IV stage, P=0.002), older age (${\geq}61days$ old, P=0.023), and use the hormones were linked to nosocomial infection and the ANN model consisted of these four factors. The artificial neural network model with variables consisting of age, clinical stage, time of hospitalization, and use of hormones should be useful for predicting nosocomial infection in lung cancer cases.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권24호
• /
• pp.10739-10743
• /
• 2015

The present study aimed at evaluating and comparing the diagnostic performance of B-mode ultrasound (US), elastography score (ES), and strain ratio (SR) for the differentiation of breast lesions. This retrospective study enrolled 431 lesions from 417 in-hospital patients. All patients were examined with both conventional ultrasound and elastography. Two experienced radiologists reviewed ultrasound and elasticity images. The histopathologic result obtained from ultrasound-guided core biopsy or operation excisions were used as the reference standard. Pathologic examination revealed 276 malignant lesions (64%) and 155 benign lesions (36%). A cut-off point of 4.15 (area under the curve, 0.891) allowed significant differentiation of malignant and benign lesions. ROC (receiver-operating characteristic) curves showed a higher value for combination of B-mode ultrasound and elastography for the diagnosis of breast lesions. Conventional ultrasound combined elastography showed high sensitivity, specificity, and accuracy for group II lesions (10mm${\leq}20mm$). Elastography combined with conventional ultrasound show high specificity and accuracy for differentiation of benign and malignant breast lesions. Elastography is particularly important for the diagnosis of BI-RADS 4 and small breast lesions.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권14호
• /
• pp.5883-5888
• /
• 2014

Background: Breast cancer is the most common cancers in female populations. The exact cause is not known, but is most likely to be a combination of genetic and environmental factors. Log-logistic model (LLM) is applied as a statistical method for predicting survival and it influencing factors. In recent decades, artificial neural network (ANN) models have been increasingly applied to predict survival data. The present research was conducted to compare log-logistic regression and artificial neural network models in prediction of breast cancer (BC) survival. Materials and Methods: A historical cohort study was established with 104 patients suffering from BC from 1997 to 2005. To compare the ANN and LLM in our setting, we used the estimated areas under the receiver-operating characteristic (ROC) curve (AUC) and integrated AUC (iAUC). The data were analyzed using R statistical software. Results: The AUC for the first, second and third years after diagnosis are 0.918, 0.780 and 0.800 in ANN, and 0.834, 0.733 and 0.616 in LLM, respectively. The mean AUC for ANN was statistically higher than that of the LLM (0.845 vs. 0.744). Hence, this study showed a significant difference between the performance in terms of prediction by ANN and LLM. Conclusions: This study demonstrated that the ability of prediction with ANN was higher than with the LLM model. Thus, the use of ANN method for prediction of survival in field of breast cancer is suggested.

• Asian Pacific Journal of Cancer Prevention
• /
• 제14권9호
• /
• pp.5495-5501
• /
• 2013

Background: Alu elements are one of the most common repetitive sequences that now constitute more than 10% of the human genome and potential targets for epigenetic alterations. Correspondingly, methylation of these elements can result in a genome-wide event that may have an impact in cancer. However, studies investigating the genome-wide status of Alu methylation in cancer remain limited. Objectives: Oral squamous cell carcinoma (OSCC) presents with high incidence in South-East Asia and thus the aim of this study was to evaluate the Alu methylation status in OSCCs and explore with the possibility of using this information for diagnostic screening. We evaluated Alu methylation status in a) normal oral mucosa compared to OSCC; b) peripheral blood mononuclear cells (PBMCs) of normal controls comparing to oral cancer patients; c) among oral epithelium of normal controls, smokers and oral cancer patients. Materials and Methods: Alu methylation was detected by combined bisulfite restriction analysis (COBRA) at 2 CpG sites. The amplified products were classified into three patterns; hypermethylation ($^mC^mC$), partial methylation ($^uC^mC+^mC^uC$), and hypomethylation ($^uC^uC$). Results: The results demonstrate that the $%^mC^mC$ value is suitable for differentiating normal and cancer in oral tissues (p=0.0002), but is not significantly observe in PBMCs. In addition, a stepwise decrease in this value was observed in the oral epithelium from normal, light smoker, heavy smoker, low stage and high stage OSCC (p=0.0003). Furthermore, receiver operating characteristic (ROC) curve analyses demonstrated the potential of combined $%^mC$ or $%^mC^mC$ values as markers for oral cancer detection with sensitivity and specificity of 86.7% and 56.7%, respectively. Conclusions: Alu hypomethylation is likely to be associated with multistep oral carcinogenesis, and might be developed as a screening tool for oral cancer detection.

• Imaging Science in Dentistry
• /
• 제44권4호
• /
• pp.301-305
• /
• 2014

Purpose: The purpose of this study was to compare the accuracy of postoperative maxillary cyst (POMC) diagnosis by panoramic radiographs versus computed tomography (CT) and by oral and maxillofacial radiologists versus non-specialists. Materials and Methods: Sixty-five maxillary sinuses with POMCs and 63 without any lesion were assessed using panoramic radiographs and CT images by five oral and maxillofacial radiologists and five non-specialists on a five-point scale. The areas under receiver operating characteristic (ROC) curves were analyzed to determine the differences in diagnostic accuracy between the two imaging modalities and between the two groups of observers. The intra-observer agreement was determined, too. Results: The diagnostic accuracy of CT images was higher than that of panoramic radiographs in both groups of observers (p<0.05). The diagnostic accuracy of oral and maxillofacial radiologists for each method was higher than that of non-specialists (p<0.05). Conclusion: The use of CT improves the diagnosis of POMC, and radiological training and experience leads to more accurate evaluation.

• Journal of Trauma and Injury
• /
• 제27권4호
• /
• pp.139-144
• /
• 2014

Purpose: Detection and determination of tendon injury in the finger or hand is not easy. Therefore, we aimed to study and evaluate the accuracy and the effectiveness of ultrasonography for the diagnosis of finger tendon injury. Methods: In this study, we enrolled patients, regardless of age and sex, with lacerations on their fingers. Patients with invisible wounds were excluded. We evaluated the accuracy and the effectiveness of ultrasonography and compared the results obtained from ultrasonography and with those obtained by visual observation of the injuries. Results: The sensitivity, the specificity and the accuracy of ultrasonography were found to be 66.7%, 100% and 91.3%, respectively (p<0.001) while those of physical examination were 71.4%, 98.3% and 91.3%, respectively. Small differences were observed between the sensitivities and specificities of the two examinations; however, the accuracies were the same (p<0.001). The area under the receiver operating characteristic (ROC) curve, which was used for diagnosis of tendon rupture using ultrasonography, was found to be 0.985 (95% confidence interval CI: 0.929-0.999),while that of physical examination was 0.938 (95% CI: 0.861-0.980). Conclusion: Ultrasonography can be used an effective diagnostic tool for patients with finger tendon injury.

• 대한디지털의료영상학회논문지
• /
• 제6권1호
• /
• pp.31-40
• /
• 2003

Purpose : In a PACS(Picture Archiving Communications System) environment, which is a very important component in a digital medical environment, the compression of digital medical images is a necessary and effective feature. In a current system where JPEG is applied to the compression of medical images, this study is to examine effectiveness and suitability when the JPEG2000, a more advanced compression algorithm for still images, is applied to the compression of medical images. In this thesis, we attempt to address the compressibility for effective clinical usage when compressing medical images, applying the objectivization of clinical evaluation as a function of compressibility. In the experiment al method, the compression was applied at a fixed rate using JPEG2000, and the n the result was compared with compressed images by JPEG. Method : For the performance evaluation, we choose SNR(Signal to Noise Ratio) measurement of an objective evaluation of definition and analyze a subjective evaluation by the ROC(Receiver Operating Characteristic) method. The results of the experiment showed that in the case of JPEG2000 there is hardly any distortion of images, even at high compression ratio(100:1), while regarding noise, the SNR remains around about 40dB, which is also relatively high. Before reading by reference to evaluative materials concerning objective compressed images, it is impossible to apply high compression to images : however, after reading, this can be applied to images that have already existed for some time.