Radiomics Nomograms Based on Multi-Parametric MRI for Preoperative Differential Diagnosis of Malignant and Benign Sinonasal Tumors: A Two-Centre Study

ObjectivesTo investigate the efficacy of multi-parametric MRI-based radiomics nomograms for preoperative distinction between benign and malignant sinonasal tumors.MethodsData of 244 patients with sinonasal tumor (training set, n=192; test set, n=52) who had undergone pre-contrast MRI, and 101 patients who underwent post-contrast MRI (training set, n=74; test set, n=27) were retrospectively analyzed. Independent predictors of malignancy were identified and their performance were evaluated. Seven radiomics signatures (RSs) using maximum relevance minimum redundancy (mRMR), and the least absolute shrinkage selection operator (LASSO) algorithm were established. The radiomics nomograms, comprising the clinical model and the RS algorithms were built: one based on pre-contrast MRI (RNWOC); the other based on pre-contrast and post-contrast MRI (RNWC). The performances of the models were evaluated with area under the curve (AUC), calibration, and decision curve analysis (DCA) respectively.ResultsThe efficacy of the clinical model (AUC=0.81) of RNWC was higher than that of the model (AUC=0.76) of RNWOC in the test set. There was no significant difference in the AUC of radiomic algorithms in the test set. The RS-T1T2 (AUC=0.74) and RS-T1T2T1C (RSWC, AUC=0.81) achieved a good distinction efficacy in the test set. The RNWC and the RNWOC showed excellent distinction (AUC=0.89 and 0.82 respectively) in the test set. The DCA of the nomograms showed better clinical usefulness than the clinical models and radiomics signatures.ConclusionsThe radiomics nomograms combining the clinical model and RS can be accurately, safely and efficiently used to distinguish between benign and malignant sinonasal tumors.

Providing a framework for knowledge sharing in knowledge-based organizations according to social capital indicators

Purpose: This study aimed to provide a framework for examining knowledge sharing from the perspective of social capital in knowledge-based organizations. For this purpose, the social capital and knowledge sharing literature has been reviewed and appropriate dimensions and components have been selected for use in the proposed framework. Research methodology: The number of experts participating in this study was 12. Kendall’s W Ranks is a statistical index for assessing the degree of agreement in a group. This statistical index, which is related to non-parametric statistics, is used to assess the degree to which a number of evaluators agree on the ranking of several factors. Result: This research's proposed framework shows that the main bottleneck in the knowledge sharing process is the sender. Numerous factors affect the motivation, quantity and quality of knowledge shared by him. The proposed conceptual model shows the factors in the field of social capital well. This means that the organization's managers should pay special attention to people with knowledge to activate the process of knowledge sharing in the organization as a driving force. Limitations: This research only described knowledge-based organizations. Contribution: The proposed conceptual framework makes a good distinction between formal and informal knowledge-sharing channels in a knowledge-based organization. Formal channels are controlled and manageable channels. Although the main burden of knowledge sharing in organizations lies with informal channels, the more managers can shift the organization's knowledge to formal channels, the more they can manage the process. This is a part of the job that social capital, in particular, helps to achieve. The proposed framework discusses the acceptance of the use of formal frameworks. Keywords: Social capital, Knowledge management, Knowledge sharing, Expertise method

Disinhibition in Frontotemporal Dementia and Alzheimer’s Disease: A Neuropsychological and Behavioural Investigation

Objective:Cognitive tests of inhibitory control show variable results for the differential diagnosis between behavioural variant of Frontotemporal Dementia (bvFTD) and Alzheimer’s disease (AD). We compared the diagnostic accuracies of tests of inhibitory control and of a behavioural questionnaire, to distinguish bvFTD from AD.Methods:Three groups of participants were enrolled: 27 bvFTD patients, 25 AD patients, and 24 healthy controls. Groups were matched for gender, education, and socio-economic level. Participants underwent a comprehensive neuropsychological assessment of inhibitory control, including Hayling Test, Stroop, the Five Digits Test (FDT) and the Delay Discounting Task (DDT). Caregivers completed the Barratt Impulsiveness Scale 11th version (BIS-11).Results:bvFTD and AD groups showed no difference in the tasks of inhibitory control, while the caregiver questionnaire revealed that bvFTD patients were significantly more impulsive (BIS-11: bvFTD 76.1+9.5, AD 62.9+13, p < .001).Conclusions:Neuropsychological tests of inhibitory control failed to distinguish bvFTD from AD. On the contrary, impulsivity caregiver-completed questionnaire provided good distinction between bvFTD and AD. These results highlight the current limits of cognitive measures of inhibitory control for the differential diagnosis between bvFTD and AD, whereas questionnaire information appears more reliable and in line with clinical diagnostics.

EDGE-BASED LOCALLY AGGREGATED DESCRIPTORS FOR IMAGE CLUSTERING

The current global image descriptors are mostly obtained by using the local image features aggregation, which fail to take full account of the details of the image, resulting in the loss of the semantic content information. It cannot be well used to make a good distinction between the high similarity images. In this paper, a new method of image representation, which can express the whole semantics and detail features of the image, is proposed by combining the edge features of the image. It is used to make a global description of the images and then clustering. The experimental results show that the proposed method is capable of clustering of the similarity images with high accuracy and low error rate.

Is it appropriate to use fixed assay cut-offs for estimating seroprevalence?

SUMMARYPopulation seroprevalence can be estimated from serosurveys by classifying quantitative measurements into positives (past infection/vaccinated) or negatives (susceptible) according to a fixed assay cut-off. The choice of assay cut-offs has a direct impact on seroprevalence estimates. A time-resolved fluorescence immunoassay (TRFIA) was used to test exposure to human parvovirus 4 (HP4). Seroprevalence estimates were obtained after applying the diagnostic assay cut-off under different scenarios using simulations. Alternative methods for estimating assay cut-offs were proposed based on mixture modelling with component distributions for the past infection/vaccinated and susceptible populations. Seroprevalence estimates were compared to those obtained directly from the data using mixture models. Simulation results showed that when there was good distinction between the underlying populations all methods gave seroprevalence estimates close to the true one. For high overlap between the underlying components, the diagnostic assay cut-off generally gave the most biased estimates. However, the mixture model methods also gave biased estimates which were a result of poor model fit. In conclusion, fixed cut-offs often produce biased estimates but they also have advantages compared to other methods such as mixture models. The bias can be reduced by using assay cut-offs estimated specifically for seroprevalence studies.

Raman microspectroscopy as a diagnostic tool to study single living nasopharyngeal carcinoma cell lines

Raman spectroscopy can provide molecular-level fingerprint information about the biochemical composition and structure of cells and tissues with excellent spatial resolution. In this study, Raman spectroscopy of 3 different nasopharyngeal carcinoma cell lines C666-1, CNE1, and CNE2 and 1 nasopharyngeal normal cell line NP69 acquired on a piece of silica glass slide are presented to investigate the differences among them. The results show the ratio of I1657/I1449 (= 0.7) could provide good distinction between tumor and normal cell lines very easily, which coincides with existing reports about the study of different cell lines and bronchial tissue. In addition, several statistical analytical methods were used to classify these 4 different cell lines and then achieved an exciting result with great sensitivity and specificity of >90%, respectively. The findings of this work further support former work where cells' Raman spectra were acquired on a different substrate. All of these results indicate Raman spectroscopy has the potential to discriminate between normal and tumor cells and have potential use in early diagnosis of nasopharyngeal carcinoma.

Changes in blood vessels in fetuses 4 to 9 months intrauterine life old by postmortem angiography method

We monitored changes in caliber, position and branching of blood vessels in fetuses of 4 - 9 months of intrauterine life. By precise dissection we prepared starting parts of common carotid artery and internal jugular vein in 40 cadaver fetuses. The vessels were injected with Telebrix and subjected to postmortem angiography at the Institute of Radiology Clinics Center in Sarajevo. Thereafter, arteries obtained were compared and analyzed. In preparations of few months old fetal material we observed arteries of fairly straight course, low caliber and with no observable ramification. When preparations of more mature stillborn infants were examined, we detected arteries of undulating course, more expressed ramification and higher caliber. In stillborn babies, all three arteries are of high caliber with rich branching. Considering that in this phase of brain development sulcuses are relatively wide, we can say that course of arteries is partially tortuous. Analysis of venous vessels shows good distinction of venous sinuses and sub-arachnoidal cisterns. We can conclude with great certainty that the changes occur in position, caliber and relationship among vessels in fetus during the period of brain sulci and gyri formation.

KERNEL WHITENING FOR ONE-CLASS CLASSIFICATION

In one-class classification one tries to describe a class of target data and to distinguish it from all other possible outlier objects. Obvious applications are areas where outliers are very diverse or very difficult or expensive to measure, such as in machine diagnostics or in medical applications. In order to have a good distinction between the target objects and the outliers, good representation of the data is essential. The performance of many one-class classifiers critically depends on the scaling of the data and is often harmed by data distributions in (nonlinear) subspaces. This paper presents a simple preprocessing method which actively tries to map the data to a spherical symmetric cluster and is almost insensitive to data distributed in subspaces. It uses techniques from Kernel PCA to rescale the data in a kernel feature space to unit variance. This transformed data can now be described very well by the Support Vector Data Description, which basically fits a hypersphere around the data. The paper presents the methods and some preliminary experimental results.