Pathological lung segmentation based on random forest combined with deep model and multi-scale superpixels

Abstract

Accurate segmentation of lungs in pathological thoracic computed tomography (CT) scans plays an important role in pulmonary disease diagnosis. However, it is still a challenging task due to the variability of pathological lung appearances and shapes. In this paper, we proposed a novel segmentation algorithm based on random forest (RF), deep convolutional network, and multi-scale superpixels for segmenting pathological lungs from thoracic CT images accurately. A pathological thoracic CT image is first segmented based on multi-scale superpixels, and deep features, texture, and intensity features extracted from superpixels are taken as inputs of a group of RF classifiers. With the fusion of classification results of RFs by a fractional-order gray correlation approach, we capture an initial segmentation of pathological lungs. We finally utilize a divide-and-conquer strategy to deal with segmentation refinement combining contour correction of left lungs and region repairing of right lungs. Our algorithm is tested on a group of thoracic CT images affected with interstitial lung diseases. Experiments show that our algorithm can achieve a high segmentation accuracy with an average DSC of 96.45% and PPV of 95.07%. Compared with several existing lung segmentation methods, our algorithm exhibits a robust performance on pathological lung segmentation. Our algorithm can be employed reliably for lung field segmentation of pathologic thoracic CT images with a high accuracy, which is helpful to assist radiologists to detect the presence of pulmonary diseases and quantify its shape and size in regular clinical practices.

Lung Segmentation on High-Resolution Computerized Tomography Images Using Deep Learning: A Preliminary Step for Radiomics Studies

“…In this way, the number of interactions performed by the network decreased, and a DSC of 96.60% was obtained with a segmentation time of 11 s (GPU hardware) for each CT study with an unknown number of slices but certainly much less than 600 slices of an HRCT study. In [ 41 ], an algorithm based on random forest, deep convolutional network, and multi-scale super-pixels was proposed for segmenting lungs with interstitial lung disease (IDL) using the ILDs database [ 42 ] with an average DSC of 96.45%. Khanna et al [ 43 ] implemented the residual U-Net with a false-positive removal algorithm using a training set of 173 images from three publicly available benchmark datasets, namely LUNA, VESSEL12, and HUG-ILD.…”

Construction of Pulmonary Nodule CT Radiomics Random Forest Model Based on Artificial Intelligence Software for STAS Evaluation of Stage IA Lung Adenocarcinoma

“…(2) The Gini coefficient is used to calculate the optimal separation method of each feature for each node in the classification tree. The more the Gini value decreases, the more important the feature becomes [ 13 ]. Finally, the diagnostic efficiency of the random forest model is calculated by confusion matrix.…”

Towards Machine Learning-Aided Lung Cancer Clinical Routines: Approaches and Open Challenges

“…More meaningful features can also be extracted by aggregating an auxiliary classification branch, enriching the information used for backpropagation [ 160 , 170 ]. Liu et al [ 171 ] integrated different feature extraction branches by combining deep, textured, and intensity features, to be classified as part of the lung mask or background.…”