Lung Segmentation in Chest Radiographs Using Anatomical Atlases With Nonrigid Registration

Abstract

The National Library of Medicine (NLM) is developing a digital chest X-ray (CXR) screening system for deployment in resource constrained communities and developing countries worldwide with a focus on early detection of tuberculosis. A critical component in the computer-aided diagnosis of digital CXRs is the automatic detection of the lung regions. In this paper, we present a nonrigid registration-driven robust lung segmentation method using image retrieval-based patient specific adaptive lung models that detects lung boundaries, surpassing state-of-the-art performance. The method consists of three main stages: 1) a content-based image retrieval approach for identifying training images (with masks) most similar to the patient CXR using a partial Radon transform and Bhattacharyya shape similarity measure, 2) creating the initial patient-specific anatomical model of lung shape using SIFT-flow for deformable registration of training masks to the patient CXR, and 3) extracting refined lung boundaries using a graph cuts optimization approach with a customized energy function. Our average accuracy of 95.4% on the public JSRT database is the highest among published results. A similar degree of accuracy of 94.1% and 91.7% on two new CXR datasets from Montgomery County, MD, USA, and India, respectively, demonstrates the robustness of our lung segmentation approach.

Automatic Extraction of Lung Regions Using Curve Analysis for the Detection of Lung Cancer in CT Scans

“…In order to compare the proposed system with the segmentation techniques in the literature, the performance metric Jaccard similarity coefficient (overlap measure (Ω)) [22] is used.…”

Automatic Lung Field Segmentation in X-ray Radiographs Using Statistical Shape and Appearance Models

“…37 In Refs. [31,33,36], the resolution of images is 256 × 256, whereas in our work the resolution of images is 1024 × 1024.…”

“…Since the resolutions of the images are different, it is unreasonable to compare the performance directly. Candemir 37 gave the evaluation result on both 256 × 256 and 1024 × 1024. According to Ref.…”

“…Wu et al 36 developed a new objective function with constraints of distance and edge to push the vertices of active shape model to the real lung edge and pull the vertices out of the stomach gas regions. For the third one, Candemir et al 37 proposed a hybrid method that combined a shape-prior model with graph cuts. The SIFT-flow based deformable model was employed to select initial "object" and "background" for the graph cuts.…”

Multiatlas Segmentation Using Robust Feature-Based Registration

“…the STAPLE algorithm [34] and different types of machine-learning approaches [27]. The fused segmentation proposal can be further refined into a final segmentation by using graph cut [3,22] or random forest-based methods [9]. For a comprehensive survey of multiatlas segmentation methods and their applications, see [13].…”