Search

While the thermoelectric (TE) materials have attracted significant attention in recent years, the design and discovery of new TE materials with optimal carrier concentration and band gap remains a great challenge. Herein, we report the development of machine learning (ML) methods to predict TE materials with introducing physically meaningful simple descriptors. Specifically, we use the number of electrons, Pauling electronegativity and relative atomic mass as the basic physical variables and compute 242 descriptors in 64 categories to characterize the molecular information of a TE material. Multiple stepwise regression is employed to reduce the dimensionality in the developed ML models, and 5 and 4 important features for the band gap and carrier concentration is selected, respectively. The important features are used as input of a total number of 19 ML methods to select the optimal ML models for the prediction of band gap and carrier concentration, respectively. It is shown that the least square support vector machines method is the best model for the prediction of the band gap, while the back propagating artificial neutral network model exhibits the best performance in predicting the carrier concentration values. This work provides novel theoretical guidance for the rapid prediction properties of TE materials. The simple descriptors we defined can accurately predict the band gap and carrier concentration of quaternary TE materials. [ABSTRACT FROM AUTHOR]

In this article, we propose a novel content-based image-retrieval (CBIR) scheme using compressible encrypted images called “encryption-then-compression (EtC) images.” The proposed scheme allows us not only to directly retrieve images from visually protected images but to also apply EtC images that can be compressed by using the JPEG standard for the first time. In addition, the sensitive management of secret keys is not required in our framework. The proposed retrieval scheme is carried out on the basis of weighted searching images with MPEG-7-powered localized descriptors (weighted SIMPLE descriptors) combining scalable color descriptor (SCD) or color and edge directivity descriptor (CEDD). Weighted SIMPLE descriptors are extended, and CEDD is also modified to avoid the influence of image encryption. In an experiment, the proposed scheme is demonstrated to have almost no degradation in retrieval performance compared with conventional content-based retrieval methods with plain images under the use of two datasets. In addition, the proposed scheme is shown to outperform conventional privacy-preserving CBIR schemes including state-of-the-art ones in terms of mean average precision (mAP) scores.

Objectives: To perform an external validation of the diagnostic performance of the three-step strategy proposed by the International Ovarian Tumor Analysis (IOTA) group for classifying adnexal masses as benign or malignant, when ultrasound is performed by non-expert sonographers in the first two steps. The second objective was to assess the diagnostic performance of an alternative strategy using simple-rules risk (SRR), instead of simple rules (SR), in the second step.Methods: This was a prospective observational study conducted at two university hospitals, from September 2015 to August 2017, of consecutive patients diagnosed with an adnexal mass. All women were evaluated by ultrasound using the IOTA three-step strategy. Non-expert sonographers performed the first step (use of simple descriptors to classify the masses) and the second step (use of SR if the mass could not be classified in the first step); masses that could not be classified in the first two steps were categorized by an expert sonographer based on their subjective assessment (third step). The reference standard was histological diagnosis in patients who underwent surgery or at least 12 months of follow-up in cases managed expectantly. The sensitivity, specificity, positive (LR+) and negative (LR-) likelihood ratios and overall accuracy of the IOTA three-step strategy were estimated. Furthermore, we evaluated retrospectively an alternative two-step strategy using SRR in the second step to categorize the masses not classifiable with simple descriptors, classifying the lesions as being of low, intermediate or high risk for malignancy. The diagnostic performance of this strategy was estimated by calculating its sensitivity and specificity, assuming surgical intervention for intermediate- or high-risk lesions.Results: The study included 283 patients (median age, 48 (range, 18-90) years), of whom 165 (58.3%) were premenopausal and 118 (41.7%) postmenopausal. Two hundred and sixteen (76.3%) women underwent surgery (154 benign and 62 malignant masses) and 67 (23.7%) were managed expectantly with serial ultrasound follow-up for at least 12 months. All expectantly managed masses were considered benign because no sonographic changes suggestive of malignancy were observed during follow-up. Simple descriptors could be applied in 126 (44.5%) masses. Of the remaining 157 lesions, 112 (39.6%) could be characterized using SR. Therefore, 238 (84.1%) masses could be classified by non-expert sonographers in the first two steps. Of the remaining 45 (15.9%) masses, all could be classified by an expert sonographer. Overall sensitivity, specificity, LR+ and LR- of the IOTA three-step strategy were 95.2%, 97.7%, 42.1 and 0.05, respectively. The diagnostic accuracy was 97.2%. Following the two-step strategy using SRR in the second step, of the 157 lesions not classified with simple descriptors, 42, 38 and 77 presented low, intermediate or high risk for malignancy, respectively. Based on this method, 210 women would have undergone surgical treatment. The sensitivity and specificity of this two-step strategy were 98.4% and 63.8%, respectively.Conclusions: The IOTA three-step strategy shows high accuracy for discriminating between benign and malignant adnexal lesions when used by non-expert sonographers. An alternative strategy using the SRR calculator in the second step might improve on this diagnostic performance by decreasing the number of surgical interventions and increasing sensitivity. Copyright © 2018 ISUOG. Published by John Wiley & Sons Ltd. [ABSTRACT FROM AUTHOR]

Objective: To assess the diagnostic performance of a three-step strategy proposed by the International Ovarian Tumor Analysis (IOTA) Group for discriminating between benign and malignant adnexal masses.Methods: This was a prospective observational study, performed at two tertiary-care university hospitals, of women diagnosed with an adnexal mass on transvaginal or transabdominal ultrasound between December 2012 and December 2014. Women were scheduled for an ultrasound evaluation, which was initially performed by non-expert examiners. The examiner had to classify the mass using 'simple descriptors' (first step) and, if not possible, using 'simple rules' (second step). For inconclusive masses, an expert examiner classified the mass according to their subjective impression (third step). Masses were managed expectantly, with serial follow-up examinations, or surgically, according to ultrasound findings and clinical symptoms. Histology was used as the reference standard. Masses that were managed expectantly with at least 1 year of follow-up were considered as benign for analytical purposes. Women with less than 1 year of follow-up were not included in the study.Results: Six hundred and sixty-six women were included (median age, 41 (range, 18-81) years) of whom 514 were premenopausal and 152 were postmenopausal. Based on the three-step strategy, 362 women had surgical removal of the mass (53 malignant and 309 benign), 71 masses resolved spontaneously and 233 persisted. Four hundred and forty-eight (67.3%) of 666 masses could be classified using simple descriptors and, of the 218 that could not, 147 (67.4%) were classified using simple rules. Of the remaining 71 masses, the expert examiner classified 45 as benign, 12 as malignant and 14 as uncertain. Overall sensitivity, specificity, positive likelihood ratio and negative likelihood ratio of the three-step strategy were 94.3%, 94.9%, 18.6 and 0.06, respectively.Conclusion: The IOTA three-step strategy, based on the sequential use of simple descriptors, simple rules and expert evaluation, performs well for classifying adnexal masses as benign or malignant. Copyright © 2016 ISUOG. Published by John Wiley & Sons Ltd. [ABSTRACT FROM AUTHOR]

In this paper, we explore, extend and simplify the localization of the description ability of the well-established MPEG-7 (Scalable Colour Descriptor (SCD), Colour Layout Descriptor (CLD) and Edge Histogram Descriptor (EHD)) and MPEG-7-like (Color and Edge Directivity Descriptor (CEDD)) global descriptors, which we call the SIMPLE family of descriptors. Sixteen novel descriptors are introduced that utilize four different sampling strategies for the extraction of image patches to be used as points of interest. Designing with focused attention for content-based image retrieval tasks, we investigate, analyse and propose the preferred process for the definition of the parameters involved (point detection, description, codebook sizes and descriptors' weighting strategies). The experimental results conducted on four different image collections reveal an astonishing boost in the retrieval performance of the proposed descriptors compared to their performance in their original global form. Furthermore, they manage to outperform common SIFT- and SURF-based approaches while they perform comparably, if not better, against recent state-of-the-art methods that base their success on much more complex data manipulation. [ABSTRACT FROM AUTHOR]