Search results for "Word error rate"
showing 6 items of 26 documents
A classification approach to prostate cancer localization in 3T Multi-Parametric MRI
2016
International audience; Multiparametric-magnetic resonance imaging (mp-MRI) has demonstrated, in many studies, its potential in prostate cancer detection and analysis. We propose a supervised classification approach based on mp-MRI data base of 20 patients, in order to localize prostate cancer and to achieve a cartographic representation of the prostate voxels based on classification results. Proposed method provides a computer aided detection (CAD) software for prostatic cancer. For that, we have extracted varied features providing functional, anatomical and metabolic information helping the classifier to distinguish between three different classes ("Healthy", "Benign" and "Pathologic"). W…
Palmprint and face score level fusion: hardware implementation of a contactless small sample biometric system
2011
Including multiple sources of information in personal identity recognition and verification gives the opportunity to greatly improve performance. We propose a contactless biometric system that combines two modalities: palmprint and face. Hardware implementations are proposed on the Texas Instrument Digital Signal Processor and Xilinx Field-Programmable Gate Array (FPGA) platforms. The algorithmic chain consists of a preprocessing (which includes palm extraction from hand images), Gabor feature extraction, comparison by Hamming distance, and score fusion. Fusion possibilities are discussed and tested first using a bimodal database of 130 subjects that we designed (uB database), and then two …
An Innovative Similarity Measure for Sentence Plagiarism Detection
2016
We propose and experimentally assess Semantic Word Error Rate (SWER), an innovative similarity measure for sentence plagiarism detection. SWER introduces a complex approach based on latent semantic analysis, which is capable of outperforming the accuracy of competitor methods in plagiarism detection. We provide principles and functionalities of SWER, and we complement our analytical contribution by means of a significant preliminary experimental analysis. Derived results are promising, and confirm to use the goodness of our proposal.
How Quickly Can We Predict Users’ Ratings on Aesthetic Evaluations of Websites? Employing Machine Learning on Eye-Tracking Data
2020
This study examines how quickly we can predict users’ ratings on visual aesthetics in terms of simplicity, diversity, colorfulness, craftsmanship. To predict users’ ratings, first we capture gaze behavior while looking at high, neutral, and low visually appealing websites, followed by a survey regarding user perceptions on visual aesthetics towards the same websites. We conduct an experiment with 23 experienced users in online shopping, capture gaze behavior and through employing machine learning we examine how fast we can accurately predict their ratings. The findings show that after 25 s we can predict ratings with an error rate ranging from 9% to 11% depending on which facet of visual ae…
Individualized Monitoring of Muscle Recovery in Elite Badminton
2019
\(\bf Purpose:\) Individualized reference ranges for serum creatine kinase (CK) and urea are a promising tool for the assessment of recovery status in high-level endurance athletes. In this study, we investigated the application of this approach in racket sports, specifically for the monitoring of elite badminton players during the preparation for their world championships. \(\bf Methods:\) Seventeen elite badminton players were enrolled of which 15 could be included in the final analysis. Repeated measurements of CK and urea at recovered (R) and non-recovered (NR) time points were used for the stepwise individualization of group-based, prior reference ranges as well as for the evaluation o…
American College of Cardiology/ European Society of Cardiology international study of angiographic data compression phase III. Measurement of image q…
2000
Objectives We sought to investigate up to which level of Joint Photographic Experts Group (JPEG) data compression the perceived image quality and the detection of diagnostic features remain equivalent to the quality and detectability found in uncompressed coronary angiograms. Background Digital coronary angiograms represent an enormous amount of data and therefore require costly computerized communication and archiving systems. Earlier studies on the viability of medical image compression were not fully conclusive. Methods Twenty-one raters evaluated sets of 91 cine runs. Uncompressed and compressed versions of the images were presented side by side on one monitor, and image quality differe…