Please use this identifier to cite or link to this item:
|Title: ||Texture analysis from 3D model and individual slice extraction for tuberculosis MDR detection, type classification and severity scoring|
|Authors: ||Ahmed, Md S.|
|Editors: ||Capellato, L.|
|Issue Date: ||2018|
|Citation: ||Md Sajib Ahmed, Md Obaidullah Sk, Mohan Jayatilake, Teresa Gonçalves, and Luı́s
Rato. Texture analysis from 3D model and individual slice extraction for tuberculosis
MDR detection, type classification and severity scoring. In Linda Cappellato, Nicola
Ferro, Jian-Yun Nie, and Laure Soulier, editors, Working Notes of CLEF 2018 - Conference and Labs of the Evaluation Forum, Avignon, France, September 10-14, 2018,
|Abstract: ||Tuberculosis (TB) is a dreaded bacterial infection that affects human lungs. It has been known to mankind since ancient ages.
ImageCLEF 2018 Tuberculosis task proposes three challenging subtasks based on Computed Tomography (CT) scan images of patients’ lungs: multi-drug resistance (MDR) detection, tuberculosis type (TBT) classification and severity scoring (SVR). In this work, two different approaches are presented: 3D Modeling and Slice Extraction. Several feature descriptors were calculated (mean and higher order moments, fractal dimension
and texture analysis based measures) from CT scans and different classifiers were tested. The 3D Modeling approach uses six features (Mean, Skewness, Kurtosis, Fractal Dimension, Homogeneity, and Energy) and Slice Extraction approach calculates a vector of 96 features (based on Mean, Correlation, Contrast, Homogeneity, Energy, and Entropy). In
accordance with the ranking given by the organizers, systems submitted were ranked 1 st for multi-drug resistance detection, 5 th for tuberculosis type classification and 3 rd tuberculosis severity scoring.|
|Appears in Collections:||INF - Publicações - Artigos em Revistas Internacionais Com Arbitragem Científica|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.