Author : Bondfale Namrata 1
Date of Publication :22nd June 2017
Abstract: Automated tissue characterization is standout amongst the most essential parts of computer aided diagnosis (CAD) system for interstitial lung diseases (ILDs).Deep learning strategies provides impressive results in variety of computer vision problems such as medicinal picture investigation. In this paper, we plan and evaluate a convolutional neural network (CNN) for the classification of ILD patterns. The proposed system comprises of 5 convolutional layers, trailed by two fully connected layers. The last thick layer has different outputs, equivalent to the classes like ground glass opacity (GGO), micronodules, consolidation, reticulation, honeycombing etc. Future work incorporates extending CNN to three-dimensional data (information) gave by CT volume scans and integrating the proposed method into CAD framework that intends differential diagnosis for ILDs as a strong device for radiologists.
Reference :
-
- K. R. Heitmann et al., “Automatic detection of ground glass opacities on lung HRCT using multiple neural networks,” Eur. Radiol., vol. 7, no. 9, pp. 1463–1472.
- R. Uppaluri et al., “Computer recognition of regional lung disease patterns,” Am. J. Respir. Crit. Care Med., vol. 160, no. 2, pp. 648–654
- A. Krizhevsky, I. Sutskever, and G. Hinton, “ImageNet classification with deep convolutional neural networks,” Adv. Neural Inf. Process.Syst., p. 9, 2012.
- G. Van Tulder and M. de Bruijne, “Learning features for tissue classification with the classification restricted Boltzmann machine,” Med. Comput. Vis., Algorithms for Big Data, pp. 47–58, 2014.
- Q. Li et al., “Medical image classification with convolutional neural network,” in Proc. 13th Int. Conf. Control Automat. Robot. Vis., Dec. 2014, vol. 2014, pp. 844–848.