Receiver Operating Characteristic ROC Research Articles

Identification and validation of tryptophan metabolism-related lncRNAs in lung adenocarcinoma prognosis and immune response

BackgroundTryptophan (Trp) is an essential amino acid. Increasing evidence suggests that tryptophan metabolism plays a complex role in immune escape from Lung adenocarcinoma (LUAD). However, the role of long non-coding RNAs (lncRNAs) in tryptophan metabolism remains to be investigated.MethodsThis study uses The Cancer Genome Atlas (TCGA)-LUAD dataset as the training cohort, and several datasets from the Gene Expression Omnibus (GEO) database are merged into the validation cohort. Genes related to tryptophan metabolism were identified from the Molecular Signatures Database (MSigDB) database and further screened for lncRNAs with Trp-related expression. Subsequently, a prognostic signature of lncRNAs related to tryptophan metabolism was constructed using Cox regression analysis, (Least absolute shrinkage and selection operator regression) and LASSO analysis. The predictive performance of this risk score was validated by Kaplan–Meier (KM) survival analysis, (receiver operating characteristic) ROC curves, and nomograms. We also explored the differences in immune cell infiltration, immune cell function, tumor mutational load (TMB), tumor immune dysfunction and exclusion (TIDE), and anticancer drug sensitivity between high- and low-risk groups. Finally, we used real-time fluorescence quantitative PCR, CCK-8, colony formation, wound healing, transwell, flow cytometry, and nude mouse xenotransplantation models to elucidate the role of ZNF8-ERVK3-1 in LUAD.ResultsWe constructed 16 tryptophan metabolism-associated lncRNA prognostic models in LUAD patients. The risk score could be used as an independent prognostic indicator for the prognosis of LUAD patients. Kaplan–Meier survival analysis, ROC curves, and risk maps validated the prognostic value of the risk score. The high-risk and low-risk groups showed significant differences in phenotypes, such as the percentage of immune cell infiltration, immune cell function, gene mutation frequency, and anticancer drug sensitivity. In addition, patients with high-risk scores had higher TMB and TIDE scores compared to patients with low-risk scores. Finally, we found that ZNF8-ERVK3-1 was highly expressed in LUAD tissues and cell lines. A series of in vitro experiments showed that knockdown of ZNF8-ERVK3-1 inhibited cell proliferation, migration, and invasion, leading to cell cycle arrest in the G0/G1 phase and increased apoptosis. In vivo experiments with xenografts have shown that knocking down ZNF8-ERVK3-1 can significantly inhibit tumor size and tumor proliferation.ConclusionWe constructed a new prognostic model for tryptophan metabolism-related lncRNA. The risk score was closely associated with common clinical features such as immune cell infiltration, immune-related function, TMB, and anticancer drug sensitivity. Knockdown of ZNF8-ERVK3-1 inhibited LUAD cell proliferation, migration, invasion, and G0/G1 phase blockade and promoted apoptosis.

Predictive Value of the Modified GRACE Scoring System for All-Cause Mortality in Patients with Acute Myocardial Infarction.

To establish a modified Global Registry of Acute Coronary Events (GRACE) scoring system with an improved predictive performance compared with the traditional GRACE scoring system. We identified 5512 patients who were hospitalized with a definite diagnosis of acute myocardial infarction (AMI) from January 1, 2015, to December 31, 2020, at the Heart Center of the First Affiliated Hospital of Xinjiang Medical University through the hospital's electronic medical record system. A total of 4561 patients were enrolled after the inclusion and exclusion criteria were applied. The mean follow-up was 51.8 23.4 months. The patients were divided into dead and alive groups by endpoint events. The differences between the two groups were compared using the two-sample t test and chi-square test. Adjusted traditional risk factors as well as LogBNP (B-type natriuretic peptide precursor, BNP) and the modified GRACE scoring system were included in a multifactorial COX regression model. The predictive performance of the traditional and modified GRACE scoring systems was compared by (Receiver Operating Characteristic) ROC curves. Significant differences in age, heart rate, creatinine, uric acid, LogBNP, traditional GRACE score, and modified GRACE score were found between the dead and alive groups by the two-sample t test. Comparison of the two groups by the chi-square test revealed that the dead group had a higher incidence of males; higher cardiac function class; a previous history of hypertension, diabetes, coronary artery disease (CAD), or cerebrovascular disease; a history of smoking; the need for intra-aortic balloon pump (IABP) support; and more patients taking aspirin, clopidogrel, ticagrelor, and -blockers. The results were analyzed by a multifactorial COX regression model, and after adjusting for confounders, age, cardiac function class, history of CAD, use of aspirin and -blockers, and the modified GRACE scoring system were found to be associated with all-cause mortality (ACM) in patients with AMI. The ROC curve was used to compare the predictive performance of the conventional GRACE scoring system with that of the modified GRACE scoring system, and it was found that the modified GRACE scoring system (Area Under Curve (AUC) = 0.809, p 0.001, 95% (Confidence Interval) CI (0.789-0.829)) was significantly better than the traditional GRACE scoring system (AUC = 0.786, p 0.001, 95% CI (0.764-0.808)), the comparison between the two scores was statistically significant (p 0.001). The change in the C statistic after 10-fold crossover internal validation of the modified GRACE score was not significant, and the integrated discrimination improvement (IDI) between the old and new models was calculated with IDI = 0.019 0, suggesting that the modified GRACE score has a positive improvement on the traditional GRACE score. The modified GRACE scoring system, established by combining B-type natriuretic peptide precursor (BNP) and the traditional GRACE scoring system, was independently associated with ACM in patients with AMI, with a larger AUC and higher predictive value than the traditional GRACE scoring system. NCT02737956.

Comparative study of logistic regression and artificial neural networks on predicting breast cancer cytology

<p>Currently, breast cancer is one of the most common cancers and a main reason of women death worldwide particularly in<strong> </strong>developing countries such as Iraq. our work aims to predict the type of tumor whether benign or malignant through models that were built using logistic regression and neural networks and we hope it will help doctors in detecting the type of breast tumor. Four models were set using binary logistic regression and two different types of artificial neural networks namely multilayer perceptron MLP and radial basis function RBF. Evaluation of validated and trained models was done using several performance metrics like accuracy, sensitivity, specificity, and AUC (area under receiver operating characteristic ROC). Dataset was downloaded from UCI ml repository; it is composed of 9 attributes and 699 samples. The findings are clearly showing that the RBF NN classifier is the best in prediction of the type of breast tumors since it had recorded the highest performance in terms of correct classification rate (accuracy), sensitivity, specificity, and AUC (area under Receiver Operating Characteristic ROC) among all other models.</p>

Multi-Platform Characterization of Cerebrospinal Fluid and Serum Metabolome of Patients Affected by Relapsing-Remitting and Primary Progressive Multiple Sclerosis.

Background: Multiple sclerosis (MS) is a chronic immunemediated disease of the central nervous system with a highly variable clinical presentation and disease progression. In this study, we investigate the metabolomics profile of patients affected by relapsing–remitting MS (RRMS)and primary progressive MS (PPMS), in order to find potential biomarkers to distinguish between the two forms. Methods: Cerebrospinal Fluid CSF and blood samples of 34 patients (RRMS n = 22, PPMS n = 12) were collected. Nuclear magnetic resonance (1H-NMR) and mass spectrometry (coupled with a gas chromatography and liquid chromatography) were used as analytical techniques. Subsequently, a multivariate statistical analysis was performed; the resulting significant variables underwent U-Mann–Whitney test and correction for multiple comparisons. Receiver Operating Characteristic ROC curves were built and the pathways analysis was conducted. Results: The analysis of the serum and the CSF of the two classes, allowed the identification of several altered metabolites (lipids, biogenic amines, and amino acids). The pathways analysis indicated the following pathways were affected: Glutathione metabolism, nitrogen metabolism, glutamine–glutamate metabolism, arginine–ornithine metabolism, phenylalanine, tyrosine and tryptophan biosynthesis etc. Conclusion: The analysis allowed the identification of a set of metabolites able to classify RRMS and PPMS patients, each of whom express different patterns of metabolites in the two biofluids.

D Dimer as a marker of severity and prognosis in acute pancreatitis

Introduction: Acute Pancreatitis (AP) has a substantial burden on health System. AP scores needs battery of tests, applicable only after 48 hours. CT severity index (CTSI) is not helpful at an early stage. It would be of great help, with a simple single test that is widely available, cost effective which can predict Severe AP (SAP), clinical progression & outcome. Method: Prospective observational study. Diagnosis of AP and severity grading was made as per Revised Atlanta Criteria. Patients were categorised into Mild AP and Complicated AP (CAP) which includes Moderate and severe AP. D-dimer done within 3 days of disease onset. Chronic/Recurrent Pancreatitis, coagulopathy, Pregnancy, Trauma, Liver disease, on anticoagulants and organ failure excluded. Modified Marshal Scoring system used for organ failure. APACHE II score, RANSONS, CTSI used. D-Dimer, CRP, routine tests done. Mean, Median, Mann whitney U test, Chi square test, Receiver Operator Characteristic ROC curve analysis done. Result: 30 patients. 50% mild AP, 43.3% moderately severe AP, 6.7% SAP. 93.3% elevated D-Dimer. Median D-Dimer in Mild AP 567 ng/L, CAP 2732 ng/L (p<0.001). D-Dimer in normal CT 363 ng/L vs. positive CTSI 1916 ng/L (p=0.005), increases as CTSI increases (p=0.04). D-Dimer in organ failure 4776 ng/L vs 776.5ng/L in absent organ failure (p=0.001). D Dimer increases as severity of Organ failure increases (p=0.04). D-dimer in ICU patients 2732ng/L (p=0.021) D-dimer and APACHE II score correlates well with increase in predictive mortality rate (p=0.01). On ROC, D-dimer>933.5ng/L suggestive of CAP, D-Dimer > 827.5ng/L is highly associated with positive CTSI sensitivity 90.9%, specificity 83.3%, D-dimer > 1060.5ng/L suggestive of Organ failure sensitivity 85.7%, specificity 72.7%. Conclusion: Coagulopathy & microthrombi plays a significant role in early pathogenesis. D-Dimer test acts at the level of this core pathogenesis, even before the complications sets in, unlike scores/CTSI which are not helpful in the initial days of AP.

Secretary leukocyte peptide inhibitor (SLPI) in pleural fluid is useful as a diagnostic marker of malignant pleural mesothelioma.

e23098 Background: There is no established diagnostic marker for malignant pleural mesothelioma (MPM). In particular, the differentiation from benign asbestos pleural effusion (BAPE) is challenging. Secretary leukocyte peptide inhibitor (SLPI) is an enzyme encoded by SLPI gene. SLPI gene is reported to be overexpressed in MPM cells. Methods: SLPI in pleural fluid was determined using Quantikine ELISA Human SLPI kit (R&amp;D Systems). Results: Exploratory research revealed that SLPI value in pleural fluid of patients with MPM (n = 52) were significantly higher than those in lung cancer (LC) (n = 69) and BAPE (n = 50) ( P= 0.000). Prospective validation study included 12 pts of MPM, 24 pts of LC, 26 pts of BAPE. Median values of SLPI in MPM, LC, and BAPE were 159 .6 ng/ml, 90.5 ng/ml, and 43.2 ng/ml, respectively. SLPI value in patients with MPM were significantly higher than those in other groups ( P= 0.000). Receiver operating characteristics ROC) analysis was performed to examine the usefulness of differentiation of MPM and other diseases, and demonstrated that area under the curve (AUC) value was 0.758. With the cut of value of 88.6 ng/ml, the sensitivity was 83.3% and the specificity was 59.8%. Concerning the differentiation between MPM and BAPE, AUC value was 0.904 and with the cut of value of 81.8 ng/ml, the sensitivity was 83.3% and the specificity was 92.3%. Conclusions: Pleural fluid SLPI is a useful biomarker for the diagnosis of MPM, in particular, for the differentiation from BAPE.

Automatic plant disease diagnosis using mobile capture devices, applied on a wheat use case

Disease diagnosis based on the detection of early symptoms is a usual threshold taken into account for integrated pest management strategies. Early phytosanitary treatment minimizes yield losses and increases the efficacy and efficiency of the treatments. However, the appearance of new diseases associated to new resistant crop variants complicates their early identification delaying the application of the appropriate corrective actions. The use of image based automated identification systems can leverage early detection of diseases among farmers and technicians but they perform poorly under real field conditions using mobile devices. A novel image processing algorithm based on candidate hot-spot detection in combination with statistical inference methods is proposed to tackle disease identification in wild conditions. This work analyses the performance of early identification of three European endemic wheat diseases – septoria, rust and tan spot. The analysis was done using 7 mobile devices and more than 3500 images captured in two pilot sites in Spain and Germany during 2014, 2015 and 2016. Obtained results reveal AuC (Area under the Receiver Operating Characteristic–ROC–Curve) metrics higher than 0.80 for all the analyzed diseases on the pilot tests under real conditions.

MP62-02 VARIATION COEFFICIENT OF STONE DENSITY: A NOVEL PREDICTOR OF TREATMENT OUTCOME FOLLOWING EXTRACORPOREAL SHOCKWAVE LITHOTRIPSY.

You have accessJournal of UrologyStone Disease: Shock Wave Lithotripsy1 Apr 2017MP62-02 VARIATION COEFFICIENT OF STONE DENSITY: A NOVEL PREDICTOR OF TREATMENT OUTCOME FOLLOWING EXTRACORPOREAL SHOCKWAVE LITHOTRIPSY. Shimpei Yamashita, Yasuo Kohjimoto, Takashi Iguchi, Akinori Iba, and Isao Hara Shimpei YamashitaShimpei Yamashita More articles by this author , Yasuo KohjimotoYasuo Kohjimoto More articles by this author , Takashi IguchiTakashi Iguchi More articles by this author , Akinori IbaAkinori Iba More articles by this author , and Isao HaraIsao Hara More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2017.02.1933AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES Although previous studies have indicated that stone heterogeneity can affect shock wave lithotripsy (SWL) outcomes, there is no established measurement of stone heterogeneity on non-contrast computed tomography (NCCT) imagery. We investigated whether variation coefficient of stone density (VCSD) can predict shock wave success. METHODS We conducted a retrospective review of 245 patients with urinary calculi who had undergone SWL for upper urinary tract calculi. We compared discriminative powers for treatment success using receiver operating characteristics ROC analyses among three indices based on CT attenuation of the stone, i.e., mean stone density (MSD), standard deviation of stone density (SDSD) and VCSD. We also performed logistic regression analysis to identify factors contributing to treatment success. Treatment success was determined as absence or residual fragments <4mm using NCCT within three months after a first session of SWL. RESULTS Treatment success rate was 47.8% (117/245 cases). From ROC analyses for treatment success, AUC of VCSD (0.71) was larger than that of MSD (0.64, p=0.09) and SDSD (0.54, p<0.01) (Figure). Multivariate analysis revealed that MDS (p=0.028) and VCSD (p<0.001) independently predicted the outcome (Table). Categorized by stone location, VCSD was the independent significant predictor for SWL outcomes in both kidney (p=0.047) and ureteral calculi (p<0.001). CONCLUSIONS We found that variation coefficient of stone density can be a novel predictor of SWL success. The development of nomograms or scoring systems including VCSD can assist in the decision process for patients and minimize unnecessary delay in treatment of urolithiasis. © 2017FiguresReferencesRelatedDetails Volume 197Issue 4SApril 2017Page: e827-e828 Advertisement Copyright & Permissions© 2017MetricsAuthor Information Shimpei Yamashita More articles by this author Yasuo Kohjimoto More articles by this author Takashi Iguchi More articles by this author Akinori Iba More articles by this author Isao Hara More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...

An underestimated blood parameter before myeloproliferative neoplasm diagnosis: the case of thrombocytosis and the delayed essential thrombocytemia diagnosis

Background: Thrombocytosis, i.e. platelet count >400.000/μL in peripheral blood, is also due to Essential Thrombocythemia (ET), and delaying the diagnosis of thrombocytosis might coincide with a delayed ET diagnosis. The purpose of our study was to investigate the factors virtually leading to a delay in ET diagnosis. Methods: We analyzed the interval between the first evidence of thrombocytosis and ET diagnosis, and defined it as ETunderestimation- period (ET-uP). We analyzed 280 ET patients followed at our Institution. To verify possible correlations with ET-uP, we investigated patients’ ages, gender, history of vascular complications, thrombocytosis, splenomegaly, and marrow fibrosis degree. We also considered their General Practitioners (GP) characteristics: age, personal computer use, number of patient visits in the two years before diagnosis. Univariate logistic analysis for each parameter was performed using the Spearman’s rank correlation coefficient to assess correlation with ET-uP. Multivariate logistic regression analysis was performed on the parameters that resulted significant at the previous analyses. The logistic regression analysis was performed dichotomizing the ET-uP variable using its median value. Results: Median ET-uP was 16 months (range 1-240). The analysis showed that a shorter ET-uP correlated with thrombocytosis (p<0.001) and previous cardiovascular events (p=0.04). At multivariate analysis, the Area. Under the Curve (AUC) of the (Receiver Operating Characteristic) ROC curve was 0.72. No correlation was found analyzing GP data. Conclusion: Our results underline a deficient medical practice, and prompt for corrective actions (use of more restrictive normal range for platelet count–with lower upper limit and warning flags for pathological findings) to timely commence the diagnostic workup of this condition.

Detection of hypochromic anemia in blood sample microscopic images using adaptive network-based fuzzy inference system

Volume of red blood cell is an important factor in distinguishing its abnormalities. Mean corpuscular volume MCV of red blood cells contributes much to differentiation of several blood diseases like iron deficiency and other types of anemia. This paper proposes an automated system to classify blood samples using cell microscopic images instead of pathology test results. Adaptive local thresholding is first used to segment cell images. The volumes of red cells are then estimated by assuming torus geometry for cells. Finally, an adaptive network-based fuzzy inference system ANFIS is used to classify blood samples to normal and abnormal. Accuracy of the proposed system and area under Receiver Operating Characteristics ROC curve are 100% and 1 respectively.

Hybrids of support vector machine wrapper and filter based framework for malware detection

Malware replicates itself and produces offspring with the same characteristics but different signatures by using code obfuscation techniques. Current generation Anti-Virus (AV) engines employ a signature-template type detection approach where malware can easily evade existing signatures in the database. This reduces the capability of current AV engines in detecting malware. In this paper we propose a hybrid framework for malware detection by using the hybrids of Support Vector Machines Wrapper, Maximum-Relevance–Minimum-Redundancy Filter heuristics where Application Program Interface (API) call statistics are used as a malware features. The novelty of our hybrid framework is that it injects the filter’s ranking score in the wrapper selection process and combines the properties of both wrapper and filters and API call statistics which can detect malware based on the nature of infectious actions instead of signature. To the best of our knowledge, this kind of hybrid approach has not been explored yet in the literature in the context of feature selection and malware detection. Knowledge about the intrinsic characteristics of malicious activities is determined by the API call statistics which is injected as a filter score into the wrapper’s backward elimination process in order to find the most significant APIs. While using the most significant APIs in the wrapper classification on both obfuscated and benign types malware datasets, the results show that the proposed hybrid framework clearly surpasses the existing models including the independent filters and wrappers using only a very compact set of significant APIs. The performances of the proposed and existing models have further been compared using binary logistic regression. Various goodness of fit comparison criteria such as Chi Square, Akaike’s Information Criterion (AIC) and Receiver Operating Characteristic Curve ROC are deployed to identify the best performing models. Experimental outcomes based on the above criteria also show that the proposed hybrid framework outperforms other existing models of signature types including independent wrapper and filter approaches to identify malware.

Prospective evaluation of acoustic radiation force impulse technology in the differentiation of thyroid nodules: accuracy and interobserver variability assessment.

The aim of this study is to assess the diagnostic efficacy and interobserver agreement of acoustic radiation force impulse (ARFI) elasticity imaging in differentiating thyroid nodules. In our study, 74 consecutive patients (52 females, 22 males; age range 27-77years, mean: 41years) with 82 thyroid nodules (60 benign nodules, and 22 malignant) were examined by two radiologists with different experience. Patients underwent either cytology using fine needle aspiration cytology or thyroid surgery. The diagnostic performance of the two operators at ARFI with sensitivity, specificity, positive predictive and negative predictive value, and ROC curves was estimated. Inter-reader variability between the two operators was defined using Cohen's k. According to receiver operating characteristics ROC curves (AUROC=0.86 for observer 1; 0.81 for observer 2) sensitivity, specificity, PPV and NPV of reader 1 and 2 were respectively: 90, 75, 90.91 and 96.55%; (cut-off value of shear wave: 2.455m/s); 90, 72, 90 and 96.90% (cut-off value shear wave: 2.365m/s). Concordance between the two operators was good (k=0.755). This work is a feasibility study evaluating ARFI imaging. Its results suggest that ARFI imaging is a reproducible method which can be utilized with good diagnostic performance in the thyroid for discriminating benign and malignant nodules using the cut-off value of 2.455m/s. However, larger studies are needed to validate this method.

A unified framework for the performance analyses of diversity energy detectors over fading channels

This paper revisits the problem of energy detection of an unknown deterministic signal over a myriad of fading environments. Specifically, a new analytical approach based on six alternative series representations of the generalised Marcum Q-function of real order in conjunction with moments or derivatives of the Moment Generating Function of Signal-to-Noise Ratio SNR is proposed to analyse the performance of Maximal-Ratio Combining MRC and Square-Law Combining SLC energy detectors with independent but non-identically distributed i.n.d fading statistics. Our analytical framework is sufficiently general, and capable of handling half-odd integer values for the time-bandwidth product and non-integer fading severity indices for the Nakagami-m fading environment, as well as i.n.d Rice and mixed fading scenarios. Many of these cases were either intractable with the classical probability density function/contour integral approaches, or had resisted a simple/computationally efficient solution. Selected numerical results are also provided for the Receiver Operating Characteristic ROC of MRC and SLC diversity-based energy detectors over Rice and Nakagami-m channels.

Pre-ejection mitral annular motion velocity responses to dobutamine infusion: A quantitative approach for assessment of myocardial viability

BackgroundDobutamine stress echocardiography (DSE) is widely used for detection of myocardial viability. The main limitation of DSE is its subjective interpretation. Assessment of mitral annular motion velocities with tissue Doppler imaging is a simple and quantitative measurement. ObjectiveTo determine the relationship between myocardial viability and regional systolic mitral annular motion tissue Doppler velocities responses to dobutamine stress. MethodsOur study group included 42 patients with previous myocardial infarction referred for coronary angiography and revascularization. We did dobutamine stress tissue Doppler echocardiography (DSTDE) measuring velocities of pre-ejection wave (pre-Ej) and peak ejection wave (Ej) at rest and during low-dose dobutamine infusion. We did follow up echocardiography after 1month. ResultsAfter exclusion of the normokinetic walls, we analyzed 196 walls. Using receiver operator characteristic ROC curves, the optimal cut-off value for viability assessment was an increase of 1.75cm/s in pre-ejection velocity during DSTDE (area under the curve 0.70, p<0.001). On the other hand, the optimal cut-off value for viability assessment was an increase of 1.75cm/s in ejection velocity during DSTDE (area under the curve 0.613, p=0.01). The sensitivity, specificity, and total accuracy of the DSTSE (pre-Ej) versus the gold standard for detection of myocardial viability were 66.15%, 67.94%, and 67.35%, respectively. The sensitivity, specificity, and total accuracy of the DTSE (Ej) were 56.92%, 64.12%, and 61.43%, respectively. There was a good correlation between the pre-Ej at 5ug/kg/min dobutamine infusion and the pre-Ej after revascularization (r=0.64, p=0.01) while the correlation with the Ej was moderate (r=0.50, p=0.01). ConclusionViable left ventricular myocardium could be identified easily and quantitatively with pre-ejection mitral annular velocity during dobutamine infusion. The pre-ejection wave during DSTDE showed greater sensitivity and specificity for the prediction of myocardial viability than the ejection wave.

A Weighted Threshold for Detection of Cancerous miRNA Expressions

MicroRNAs miRNA are one kind of non-coding RNA which play many important roles in eukaryotic cell. Investigations on miRNAs show that miRNAs are involved in cancer development in animal body. In this article, a threshold based method to check the condition normal or cancer of miRNAs of a given sample/patient, using weighted average distance between the normal and cancer miRNA expressions, is proposed. For each miRNA, the city block distance between two representatives, corresponding to scaled normal and cancer expressions, is obtained. The average of all such distances for different miRNAs is weighted by a factor, to generate the threshold. The weight factor, which is cancer dependent, is determined through an exhaustive search by maximizing the F score during training. In a part of the investigation, a ranking algorithm for cancer specific miRNAs is also discussed. The performance of the proposed method is evaluated in terms of Matthews Correlation Coefficient MCC and by plotting points 1 --Specificity vs: Sensitivity in Receiver Operating Characteristic ROC space, besides the F score. Its efficiency is demonstrated on breast, colorectal, melanoma lung, prostate and renal cancer data sets and it is observed to be superior to some of the existing classifiers in terms of the said indices.

The Use of Receiver Operating Characteristic (Roc) Analysis in the Evaluation of the Performance of Two Binary Diagnostic Tests of Gestational Diabetes Mellitus

Objective: To compare the accuracy measures of the random glucose test and the 50-g glucose challenge test as screening tests for gestational diabetes mellitus (GDM). Research Design and Methods: In this prospective cohort study, pregnant women without preexisting diabetes in two perinatal centers in the Ebonyi State underwent a random glucose test and a 50-g glucose challenge test between 24 and 28 weeks of gestation. If one of the screening tests exceeded predefined threshold values, the 75-g oral glucose tolerance test (OGTT) was performed within 1 week. Furthermore, the OGTT was performed in a random sample of women in whom both screening tests were normal. GDM was considered present when the OGTT (reference test) exceeded predefined threshold values. Receiver operating characteristic (ROC) analysis was used to evaluate the performance of the two screening tests. The results were corrected for verification bias. Results: We included 1,301 women. The OGTT was performed in 322 women. After correction for verification bias, the random glucose test showed an area under the ROC curve of 0.69 (95% CI 0.61– 0.78), whereas the glucose challenge test had an area under the curve of 0.88 (0.83– 0.93). There was a significant difference in area under the curve of the two tests of 0.19 (0.11– 0.27) in favor of the 50-g glucose challenge test. Conclusions: In screening for GDM, the 50-g glucose challenge test is more useful than the random glucose test.

Comparison of regression tree data mining methods for prediction of mortality in head injury

With this research, we sought to examine the performance of six different regression tree data mining methods to predict mortality in head injury. Using a data set consisting of 1603 head injury cases, we assessed the performance of: the Classification and Regression Trees (CART) method; the Chi-squared Automatic Interaction Detector (CHAID) method; the Exhaustive CHAID (E-CHAID) method; the Quick, Unbiased, Efficient Statistical Tree (QUEST) method; the Random Forest Regression and Classification (RFRC) method; and the Boosted Tree Classifiers and Regression (BTCR) method, in each case based on sensitivity, specificity, positive/negative predictive, and accuracy rates. Next, we compared their areas under the (Receiver Operating Characteristic) ROC curves. Finally, we examined whether they could be grouped in meaningful clusters with hierarchical cluster analysis. Areas under the ROC curves of regression tree data mining methods ranged from 0.801 to 0.954 ( p < 0.001 for all). In predicting mortality in head injury under the ROC curve, the BTCR method achieved both the highest area (0.954) and accuracy rate (93.0%), while the CART method achieved both the lowest area (0.801) and accuracy rate (91.1%). All of the regression tree data mining methods were clustered in the same grouping, but the BTCR method was at the origin of the cluster while the CART and QUEST methods produced results that were least like the others. The BTCR, demonstrating a 93.0% accuracy rate and showing statistically significantly differences from the others, may be a helpful tool in medical decision-making for predicting mortality in head injury.

Finding a benchmark for monitoring hospital cleanliness

This study evaluated three methods for monitoring hospital cleanliness. The aim was to find a benchmark that could indicate risk to patients from a contaminated environment. We performed visual monitoring, ATP bioluminescence and microbiological screening of five clinical surfaces before and after detergent-based cleaning on two wards over a four-week period. Five additional sites that were not featured in the routine domestic specification were also sampled. Measurements from all three methods were integrated and compared in order to choose appropriate levels for routine monitoring. We found that visual assessment did not reflect ATP values nor environmental contamination with microbial flora including Staphylococcus aureus and meticillin-resistant S. aureus (MRSA). There was a relationship between microbial growth categories and the proportion of ATP values exceeding a chosen benchmark but neither reliably predicted the presence of S. aureus or MRSA. ATP values were occasionally diverse. Detergent-based cleaning reduced levels of organic soil by 32% (95% confidence interval: 16-44%; P<0.001) but did not necessarily eliminate indicator staphylococci, some of which survived the cleaning process. An ATP benchmark value of 100 relative light units offered the closest correlation with microbial growth levels <2.5 cfu/cm(2) (receiver operating characteristic ROC curve sensitivity: 57%; specificity: 57%). In conclusion, microbiological and ATP monitoring confirmed environmental contamination, persistence of hospital pathogens and measured the effect on the environment from current cleaning practices. This study has provided provisional benchmarks to assist with future assessment of hospital cleanliness. Further work is required to refine practical sampling strategy and choice of benchmarks.

A CROC stronger than ROC: measuring, visualizing and optimizing early retrieval

The performance of classifiers is often assessed using Receiver Operating Characteristic ROC [or (AC) accumulation curve or enrichment curve] curves and the corresponding areas under the curves (AUCs). However, in many fundamental problems ranging from information retrieval to drug discovery, only the very top of the ranked list of predictions is of any interest and ROCs and AUCs are not very useful. New metrics, visualizations and optimization tools are needed to address this 'early retrieval' problem. To address the early retrieval problem, we develop the general concentrated ROC (CROC) framework. In this framework, any relevant portion of the ROC (or AC) curve is magnified smoothly by an appropriate continuous transformation of the coordinates with a corresponding magnification factor. Appropriate families of magnification functions confined to the unit square are derived and their properties are analyzed together with the resulting CROC curves. The area under the CROC curve (AUC[CROC]) can be used to assess early retrieval. The general framework is demonstrated on a drug discovery problem and used to discriminate more accurately the early retrieval performance of five different predictors. From this framework, we propose a novel metric and visualization-the CROC(exp), an exponential transform of the ROC curve-as an alternative to other methods. The CROC(exp) provides a principled, flexible and effective way for measuring and visualizing early retrieval performance with excellent statistical power. Corresponding methods for optimizing early retrieval are also described in the Appendix. Datasets are publicly available. Python code and command-line utilities implementing CROC curves and metrics are available at http://pypi.python.org/pypi/CROC/ CONTACT: pfbaldi@ics.uci.edu

Correlationship of plasminogen activator inhibitor-1 (PAI-1) and mullerian inhibiting substance (MIS) in follicular fluid with endometrial receptivity and pregnancy outcome in ART cycles

OBJECTIVE: Plasminogen activator inhibitor -1 (PAI-1)and Mullerian inhibiting substance (MIS) also known as antimullerian hormone (AMH) in follicular fluid were estimated to evaluate their relationships with endometrial receptivity and pregnancy outcome in ART cycles. DESIGN: Prospective study of 124 IVF cycles. MATERIALS AND METHODS: Controlled ovarian hyperstimulation in all women (Mean age: 30.8 ± 0.4 yrs, BMI: 24.66 ± 0.4) was done with gonadotropin (225 IU / day). Oocyte retrieval was done 34 hours after hCG administration. PAI-1 and MIS concentrations per mg of protein by ELISA and serum E2 and P by RIA method in FF were estimated. Conventional luteal phase support was given in all cycles. Seven days after embryo transfer (ET), serum P, 17OHP and E2 along with transvaginal sonography for endometrial response were measured. Measurement of β hCG on day 14 of ET indicated pregnancy outcome. By the end of sixth week, presence of gestational sac with cardiac activity confirmed clinical pregnancy. Statistical analysis, viz.t test, correlation coefficient and receiver operator characteristics ROC AUC of PAI-1 to determine the cutoff value for achieving pregnancy, was done using Graph pad Prism version 5.0. RESULTS: Out of 124 IVF cycles, six cycles were abandoned due to no response and in eight cycles no ETs were done. A mean of 2.5 ± 0.1 embryos was transferred in 110 IVF cycles. Out of 46 pregnancies (PR: 41.8 %) that resulted, 6 turned out as biochemical pregnancies. The cycles were divided into pregnant and non-pregnant groups. Both FF PAI-1 and FF MIS levels in pregnant group were significantly low (P value: <0.0001) as compared to non-pregnant group. FF MIS strongly correlated with FF PAI-1 and D7 OHP in pregnant group (Pearson r: 0.4515 and 0.3321 respectively). Serum17- OHP levels and endometrial thickness on D7 in non-pregnant group showed inverse correlation with FF PAI-1 (Pearson r = - o.3324). Though sr. P in both groups remained almost unchanged, 17- OHP and E2 levels were found to be significantly high (0.05 and 0.005 respectively) in pregnant group as compared to non-pregnant group on D7 of ET. CONCLUSIONS: MIS seemed to regulate the activity of PAI-1 in ART cycles. Lower PAI-1 activity in follicular fluid seemed to denote the promotion or triggering factor for increased proteolytic activity at the endometrial level to prepare a favorable bed for invasion of trophoblasts of the embryo, which facilitated its implantation.