Lung Nodule Sizes Are Encoded When Scaling CT Image for CNN's.

Dmitry Cherezov,Lawrence O. Hall,Robert J. Gillies,Nikolai Fetisov,Rahul Paul,Dmitry B. Goldgof,Matthew B. Schabath

doi:10.18383/j.tom.2019.00024

Dmitry Cherezov, Lawrence O. Hall + Show 5 more

Open Access

https://doi.org/10.18383/j.tom.2019.00024

Copy DOI

Abstract

Noninvasive diagnosis of lung cancer in early stages is one task where radiomics helps. Clinical practice shows that the size of a nodule has high predictive power for malignancy. In the literature, convolutional neural networks (CNNs) have become widely used in medical image analysis. We study the ability of a CNN to capture nodule size in computed tomography images after images are resized for CNN input. For our experiments, we used the National Lung Screening Trial data set. Nodules were labeled into 2 categories (small/large) based on the original size of a nodule. After all extracted patches were re-sampled into 100-by-100-pixel images, a CNN was able to successfully classify test nodules into small- and large-size groups with high accuracy. To show the generality of our discovery, we repeated size classification experiments using Common Objects in Context (COCO) data set. From the data set, we selected 3 categories of images, namely, bears, cats, and dogs. For all 3 categories a 5- × 2-fold cross-validation was performed to put them into small and large classes. The average area under receiver operating curve is 0.954, 0.952, and 0.979 for the bear, cat, and dog categories, respectively. Thus, camera image rescaling also enables a CNN to discover the size of an object. The source code for experiments with the COCO data set is publicly available in Github (https://github.com/VisionAI-USF/COCO_Size_Decoding/).

Highlights

In radiomics studies, convolutional neural networks (CNNs) are applied to address different medical questions including diagnosing [1,2,3,4], treatment response [5,6,7], and patient survival time prediction [8,9,10]
In our previous work [12], we presented a CNN model that was trained to predict whether a benign lung nodule will become a malignant tumor in 2 years using low-dose computed tomography (CT) images
The CNN model showed 76% accuracy on the National Lung Screening Trial (NLST) data set

Summary

Introduction

Zech et al [11] presented a CNN model for pneumonia detection in chest xray images and showed that the resulting model could identify hospitals, departments, and imaging device because patients with different risk scores of pneumonia were scanned using different imaging protocols. Hospital, department and scanner information are predictive by themselves and was learned by the CNN. The warping method extracts a patch with a minimum bounding box, which is enough to include the region of interest (ROI). The rectangle was located on an image such that it enclosed the ROI. Voxels/pixels within the rectangle were extracted as a patch. The patch is resampled to the size required for the CNN input. Cropping extracts an ROI patch with size equal to the CNN input image, resampling is not used.

Objectives

Methods

Results

Conclusion