Automatic estimation of heart boundaries and cardiothoracic ratio from chest x-ray images

Dallal, Ahmed; Agarwal, Chirag; Arbabshirani, Mohammad R.; Patel, Aalpen A.; Moore, Gregory J.

doi:10.1117/12.2254136

Cited by 17 publications

(15 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared to the previous studies using segmentationbased solutions for cardiomegaly classification [7], [13], The histogram of the CTR differences between the reference standard and the seg-method and second reader respectively. (e) and (f): The box plot of the differences between the maximal horizontal cardiac and thoracic diameters between the reference standard and the seg-method and the second reader in mm.…”

Section: Discussionmentioning

confidence: 99%

“…The U-net architecture [32] is a state-of-the-art segmentation network, which has achieved promising results on a variety of medical image segmentation tasks [9], [13]. It consists of contracting and expanding paths, where the contracting path is composed of convolution operations decreasing the spatial resolution and the expanding path consists of transposed convolutions increasing the resolution.…”

Section: B Segmentation-based Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Cardiomegaly Detection on Chest Radiographs: Segmentation Versus Classification

et al. 2020

View full text Add to dashboard Cite

In this study, we investigate the detection of cardiomegaly on frontal chest radiographs through two alternative deep-learning approaches -via anatomical segmentation and via image-level classification. We used the publicly available ChestX-ray14 dataset, and obtained heart and lung segmentation annotations for 778 chest radiographs for the development of the segmentation-based approach. The classification-based method was trained with 65k standard chest radiographs with image-level labels. For both approaches, the best models were found through hyperparameter searches where architectural, learning, and regularization related parameters were optimized systematically. The resulting models were tested on a set of 367 held-out images for which cardiomegaly annotations were hand-labeled by two independent expert radiologists. Sensitivity, specificity, positive predictive value, negative predictive value, and area under the receiver operating characteristic curve (AUC) were calculated. The performance of the segmentation-based system with an AUC of 0.977 is significantly better for classifying cardiomegaly than the classificationbased model which achieved an AUC of 0.941. Only the segmentation-based model achieved comparable performance to an independent expert reader (AUC of 0.978). We conclude that the segmentation-based model requires 100 times fewer annotated chest radiographs to achieve a substantially better performance, while also producing more interpretable results.INDEX TERMS Deep learning, chest radiographs, anatomy segmentations, cardiomegaly.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: B Segmentation-based Methodsmentioning

confidence: 99%

Cardiomegaly Detection on Chest Radiographs: Segmentation Versus Classification

et al. 2020

View full text Add to dashboard Cite

show abstract

“…U-Net architecture [50] is one of the state-of-the-art architecture used for segmentation with promising results es- pecially on medical imaging [22], [51]. The architecture comprises of contraction (encoder) and expansion (decoder) paths as shown in figure 5.…”

Section: ) Segmentationmentioning

confidence: 99%

“…Healthy individuals have clear lung boundaries and it is relatively easier for a segmentation algorithm to identify the lung from the entire CXR. Whereas, the patients infected with pulmonary diseases can lead to consolidation, cloudiness, cavities, and mass on lungs making it reasonably more difficult to identify and segment the lung region [22]. Figure 1 highlights the difference between the lung boundary of a healthy individual against the one infected with the disease.…”

Section: Introductionmentioning

confidence: 99%

Lung Segmentation-Based Pulmonary Disease Classification Using Deep Neural Networks

et al. 2021

View full text Add to dashboard Cite

Interpreting chest x-ray (CXR) to find anomalies in the thoracic region is a tedious job and can consume an ample amount of radiologist's time when there are thousands of them to process. In such scenarios, the Computer-Aided Diagnostic (CAD) systems can help radiologists by doing the trivial processing and presenting the information in a meaningful way so that, the radiologist can make more accurate decisions by spending less amount of time and energy. This research study intends to propose a better, accurate, and efficient CNN based pulmonary disease diagnosis system using CXR images. In the proposed system, the capabilities of deep neural network architecture are exploited by proposing a custom CNN architecture with additional layers and modified hyperparameters to meet the required results. The input CXR is examined for healthy or infected at the surface level and the infected images are further processed for class level label classification. The lung region is segmented from the entire input CXR image to reduce the amount of noise and increase the processing efficiency by processing less overall information. The proposed model is evaluated on the benchmark split of the NIH chest x-ray dataset and achieves better segmentation and classification results when compared to the state of the art approaches.

show abstract

“…Provided the estimated CTR, cardiomegaly can be predicted under different thresholds for different age groups. Following [2], the threshold, T , is chosen to be 0.5.…”

Section: Estimation Of Ctrmentioning

confidence: 99%

Unsupervised Domain Adaptation for Automatic Estimation of Cardiothoracic Ratio

Dong

Kampffmeyer

Liang

et al. 2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

The cardiothoracic ratio (CTR), a clinical metric of heart size in chest X-rays (CXRs), is a key indicator of cardiomegaly. Manual measurement of CTR is time-consuming and can be affected by human subjectivity, making it desirable to design computer-aided systems that assist clinicians in the diagnosis process. Automatic CTR estimation through chest organ segmentation, however, requires large amounts of pixel-level annotated data, which is often unavailable. To alleviate this problem, we propose an unsupervised domain adaptation framework based on adversarial networks. The framework learns domain invariant feature representations from openly available data sources to produce accurate chest organ segmentation for unlabeled datasets. Specifically, we propose a model that enforces our intuition that prediction masks should be domain independent. Hence, we introduce a discriminator that distinguishes segmentation predictions from ground truth masks. We evaluate our system's prediction based on the assessment of radiologists and demonstrate the clinical practicability for the diagnosis of cardiomegaly. We finally illustrate on the JSRT dataset that the semi-supervised performance of our model is also very promising.

show abstract

Automatic estimation of heart boundaries and cardiothoracic ratio from chest x-ray images

Cited by 17 publications

References 25 publications

Cardiomegaly Detection on Chest Radiographs: Segmentation Versus Classification

Cardiomegaly Detection on Chest Radiographs: Segmentation Versus Classification

Lung Segmentation-Based Pulmonary Disease Classification Using Deep Neural Networks

Unsupervised Domain Adaptation for Automatic Estimation of Cardiothoracic Ratio

Contact Info

Product

Resources

About