Comparison of Shallow and Deep Learning Methods on Classifying the Regional Pattern of Diffuse Lung Disease

Abstract

This study aimed to compare shallow and deep learning of classifying the patterns of interstitial lung diseases (ILDs). Using high-resolution computed tomography images, two experienced radiologists marked 1200 regions of interest (ROIs), in which 600 ROIs were each acquired using a GE or Siemens scanner and each group of 600 ROIs consisted of 100 ROIs for subregions that included normal and five regional pulmonary disease patterns (ground-glass opacity, consolidation, reticular opacity, emphysema, and honeycombing). We employed the convolution neural network (CNN) with six learnable layers that consisted of four convolution layers and two fully connected layers. The classification results were compared with the results classified by a shallow learning of a support vector machine (SVM). The CNN classifier showed significantly better performance for accuracy compared with that of the SVM classifier by 6-9%. As the convolution layer increases, the classification accuracy of the CNN showed better performance from 81.27 to 95.12%. Especially in the cases showing pathological ambiguity such as between normal and emphysema cases or between honeycombing and reticular opacity cases, the increment of the convolution layer greatly drops the misclassification rate between each case. Conclusively, the CNN classifier showed significantly greater accuracy than the SVM classifier, and the results implied structural characteristics that are inherent to the specific ILD patterns.

Keywords: Convolution neural network; Deep architecture; Interscanner variation; Interstitial lung disease; Support vector machine.

Fig. 1

Images from high-resolution CT scans of the chest are shown. Each image shows the ROI that is a typical of each particular condition: a normal lung parenchyma, b ground-glass opacity, c consolidation, d reticular opacity, e emphysema, and f honeycombing

Fig. 2

Flow diagram of the automated classification system used in the CNN

Fig. 3

Overall architecture for training the CNN network

Fig. 4

Confusion matrix of subclasses in the case of training Siemens data and testing Siemens data

Fig. 5

Comparison of whole lung quantification data using the golden standard by a radiologist (two cases). Each pixel was coded by the classification result, which is indicated by a semi-transparent color (normal, green; ground-glass opacity, yellow; reticular opacity, cyan; honeycombing, blue; emphysema, red; and consolidation, pink). a, d Original scanned images. b, e CNN classifier results. c, f Golden standard obtained by a radiologist. The colored areas beyond the lung were removed using a separately prepared lung mask