Multimodality imaging and state-of-art GPU technology in discriminating benign from malignant breast lesions on real time decision support system
Department of Biomedical Engineering, Technological Educational Institute of Athens, Greece; School of Engineering and Design, Brunel University West London, Uxbridge, Middlesex, UB8 3PH, UK; Delta Digital Diagnostic Center, Semitelou 6, Athens, 11528, Greece
J. Phys.: Conf. Ser. 490 012137
@inproceedings{kostopoulos2014multimodality,
title={Multimodality imaging and state-of-art GPU technology in discriminating benign from malignant breast lesions on real time decision support system},
author={Kostopoulos, S and Sidiropoulos, K and Glotsos, D and Dimitropoulos, N and Kalatzis, I and Asvestas, P and Cavouras, D},
booktitle={Journal of Physics: Conference Series},
volume={490},
number={1},
pages={012137},
year={2014},
organization={IOP Publishing}
}
The aim of this study was to design a pattern recognition system for assisting the diagnosis of breast lesions, using image information from Ultrasound (US) and Digital Mammography (DM) imaging modalities. State-of-art computer technology was employed based on commercial Graphics Processing Unit (GPU) cards and parallel programming. An experienced radiologist outlined breast lesions on both US and DM images from 59 patients employing a custom designed computer software application. Textural features were extracted from each lesion and were used to design the pattern recognition system. Several classifiers were tested for highest performance in discriminating benign from malignant lesions. Classifiers were also combined into ensemble schemes for further improvement of the system’s classification accuracy. Following the pattern recognition system optimization, the final system was designed employing the Probabilistic Neural Network classifier (PNN) on the GPU card (GeForce 580GTX) using CUDA programming framework and C++ programming language. The use of such state-of-art technology renders the system capable of redesigning itself on site once additional verified US and DM data are collected. Mixture of US and DM features optimized performance with over 90% accuracy in correctly classifying the lesions.
March 19, 2014 by hgpu
