In this work we analyze the effect of label noise in training and test data when performing classification experiments on chest radiographs (CXRs) with modern deep learning architectures. We use ChestXRay14, the largest publicly available CXR dataset. We simulate situs inversus by horizontal flipping of the CXRs, allowing us to precisely control the amount of label noise. We also perform experiments in classifying emphysema using the ChestXRay14 provided labels that are known to be noisy. Our situs inversus experiments confirm results from the computer vision literature that deep learning architectures are relatively robust but not completely insensitive to label noise in the training data: without or with very low noise, classification results are near perfect; 16% and 32% training label noise only lead to a 1.5% and 4.6% drop in accuracy. We investigate two metrics that could be used to identify test samples that have an incorrect label: model confidence and model uncertainty. We show, in an observer study with an experienced chest radiologist, that both measures are effective in identifying samples in ChestXRay14 that are erroneously labeled for the presence of emphysema.
A novel method for quality assessment in medical image registration is presented. It is evaluated on 24 follow-up
CT scan pairs of the lung. Based on a reference standard of manually matched landmarks we established
a pattern recognition approach for detection of local registration errors. To capture characteristics of these
misalignments a set of intensity, entropy and deformation related features was employed. Feature selection was
conducted and a kNN classifier was trained and evaluated on a subset of landmarks. Registration errors larger
than 2 mm were classified with a sensitivity of 88% and specificity of 94%.
This paper presents a new computer-aided detection scheme for lung nodules attached to the pleural or mediastinal
surface in low dose CT scans. First the lungs are automatically segmented and smoothed. Any connected
set of voxels attached to the wall - with each voxel above minus 500 HU and the total object within a specified
volume range - was considered a candidate finding. For each candidate, a refined segmentation was computed
using morphological operators to remove attached structures. For each candidate, 35 features were defined,
based on their position in the lung and relative to other structures, and the shape and density within and around
each candidate. In a training procedure an optimal set of 15 features was determined with a k-nearest-neighbor
classifier and sequential floating forward feature selection.
The algorithm was trained with a data set of 708 scans from a lung cancer screening study containing 224
pleural nodules and tested on an independent test set of 226 scans from the same program with 58 pleural
nodules. The algorithm achieved a sensitivity of 52% with an average of 0.76 false positives per scan. At 2.5
false positive marks per scan, the sensitivity increased to 80%.
The automated detection of lung nodules in CT scans is an important problem in computer-aided diagnosis. In this paper an approach to nodule candidate detection is presented which utilises the local image features of shape index and curvedness. False-positive candidates are removed by means of a two-step approach using kNN classification. The kNN classifiers are trained using features of the image intensity gradients and grey-values in addition to further measures of shape index and curvedness profiles in the candidate regions. The training set consisted of data from 698 scans while the independent test set comprised a further 142 images. At 84% sensitivity an average of 8.2 false-positive detections per scan were observed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.