Nodule detection in 3D chest CT images using 2nd order autocorrelation features

CNN models discriminating between pulmonary micro-nodules and non-nodules from CT images

Background

Early and automatic detection of pulmonary nodules from CT lung screening is the prerequisite for precise management of lung cancer. However, a large number of false positives appear in order to increase the sensitivity, especially for detecting micro-nodules (diameter < 3 mm), which increases the radiologists’ workload and causes unnecessary anxiety for the patients. To decrease the false positive rate, we propose to use CNN models to discriminate between pulmonary micro-nodules and non-nodules from CT image patches.

Methods

A total of 13,179 micro-nodules and 21,315 non-nodules marked by radiologists are extracted with three different patch sizes (16 × 16, 32 × 32 and 64 × 64) from LIDC/IDRI database and used in the experiments. Three CNN models with different depths (1, 2 or 4 convolutional layers) are designed; their performances are evaluated by the fivefold cross-validation in term of the accuracy, area under the curve (AUC), F-score and sensitivity. The network parameters are also optimized.

Results

It is found that the performance of the CNN models is greatly dependent on the patches size and the number of convolutional layers. The CNN model with two convolutional layers presented the best performance in case of 32 × 32 patches size, achieving an accuracy of 88.28%, an AUC of 0.87, a F-score of 83.45% and a sensitivity of 83.82%.

Conclusions

The CNN models with appropriate depth and size of image patches can effectively discriminate between pulmonary micro-nodules and non-nodules, and reduce the false positives and help manage lung cancer precisely.

Toward precise pulmonary nodule descriptors for nodule type classification

A framework for nodule feature-based extraction is presented to classify lung nodules in low-dose CT slices (LDCT) into four categories: juxta, well-circumscribed, vascularized and pleural-tail, based on the extracted information. The Scale Invariant Feature Transform (SIFT) and an adaptation to Daugman's Iris Recognition algorithm are used for analysis. The SIFT descriptor results are projected to lower-dimensional subspaces using PCA and LDA. Complex Gabor wavelet nodule response obtained from an adopted Daugman Iris Recognition algorithm revealed improvements from the original Daugman binary iris code. This showed that binarized nodule responses (codes) are inadequate for classification since nodules lack texture concentration as seen in the iris, while the SIFT algorithm projected using PCA showed robustness and precision in classification.