Abstract:Rapid and accurate diagnosis of pneumothorax, utilizing chest X-ray and computed tomography (CT), is crucial for assisted diagnosis. Chest X-ray is commonly used for initial localization of pneumothorax, while CT ensures accurate quantification. However, CT scans involve high radiation doses and can be costly. To achieve precise quantitative diagnosis while minimizing radiation exposure, we proposed X-Recon, a CT ultra-sparse reconstruction network based on ortho-lateral chest X-ray images. X-Recon integrates generative adversarial networks (GANs), including a generator with a multi-scale fusion rendering module and a discriminator enhanced by 3D coordinate convolutional layers, designed to facilitate CT reconstruction. To improve precision, a projective spatial transformer is utilized to incorporate multi-angle projection loss. Additionally, we proposed PTX-Seg, a zero-shot pneumothorax segmentation algorithm, combining image processing techniques with deep-learning models for the segmentation of air-accumulated regions and lung structures. Experiments on a large-scale dataset demonstrate its superiority over existing approaches. X-Recon achieved a significantly higher reconstruction resolution with a higher average spatial resolution and a lower average slice thickness. The reconstruction metrics achieved state-of-the-art performance in terms of several metrics including peak signal-to-noise ratio. The zero-shot segmentation algorithm, PTX-Seg, also demonstrated high segmentation precision for the air-accumulated region, the left lung, and the right lung. Moreover, the consistency analysis for the pneumothorax chest occupancy ratio between reconstructed CT and original CT obtained a high correlation coefficient. Code will be available at: <a class="link-external link-https" href="https://github.com/wangyunpengbio/X-Recon" rel="external noopener nofollow">this https URL</a>

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: how to use orthogonal X - ray images (i.e., anteroposterior and lateral X - ray films) to generate high - resolution CT images with a low radiation dose, so as to achieve accurate diagnosis of pneumothorax. Specifically, the paper aims to: 1. **Reduce radiation exposure**: Traditional CT scans involve a relatively high radiation dose and high cost, while using orthogonal X - ray images can significantly reduce the patient's radiation absorption. 2. **Improve reconstruction accuracy**: Through deep - learning techniques, especially generative adversarial networks (GANs), reconstruct high - resolution 3D CT images from X - ray images of only two perspectives. 3. **Achieve zero - shot segmentation**: Propose a zero - shot pneumothorax segmentation algorithm named PTX - Seg, which combines traditional image - processing techniques and deep - learning models to segment air - accumulation areas and lung structures. ### Specific problem description - **High radiation dose problem**: Although CT scans can provide detailed three - dimensional images, their radiation dose is relatively high, increasing the patient's risk. The paper hopes to replace some CT scans with low - dose X - ray images to reduce radiation exposure. - **Information loss problem**: Projecting complex 3D geometric structures into 2D images will lead to information loss. Therefore, reconstructing 3D CT images from 2D X - ray images is a very challenging problem. - **Limitations of existing methods**: Existing sparse - sampling CT reconstruction methods still require images from multiple perspectives and cannot significantly reduce the radiation dose. The X - Recon method proposed in the paper attempts to achieve high - quality CT reconstruction using only two orthogonal perspectives. ### Solution The paper proposes X - Recon, a CT reconstruction network based on generative adversarial networks (GANs), which specifically includes the following aspects: 1. **Multi - scale fusion rendering module (MFusionRen)**: Used to integrate features at different scales and improve reconstruction quality. 2. **3D coordinate convolution layer**: Enhance the performance of the discriminator so that it can better capture position information. 3. **Multi - angle projection loss**: By introducing multi - angle projection loss, ensure that the projections of the reconstructed CT images at different perspectives are consistent with the real CT images. 4. **Zero - shot segmentation algorithm (PTX - Seg)**: Combine traditional image processing and deep - learning models to segment pneumothorax areas and evaluate the quality of reconstructed CT images. Through these innovations, X - Recon not only improves the resolution and accuracy of CT reconstruction but also provides a reliable tool for clinical applications.

X-Recon: Learning-based Patient-specific High-Resolution CT Reconstruction from Orthogonal X-Ray Images

X2CT-GAN: Reconstructing CT From Biplanar X-Rays With Generative Adversarial Networks

TRCT-GAN: CT reconstruction from biplane X-rays using transformer and generative adversarial networks

Automated segmentation and diagnosis of pneumothorax on chest X-rays with fully convolutional multi-scale ScSE-DenseNet: a retrospective study

XTransCT: Ultra-Fast Volumetric CT Reconstruction using Two Orthogonal X-Ray Projections for Image-guided Radiation Therapy via a Transformer Network

XctNet: Reconstruction network of volumetric images from a single X-ray image

SdCT-GAN: Reconstructing CT from Biplanar X-Rays with Self-driven Generative Adversarial Networks

DuoLift-GAN:Reconstructing CT from Single-view and Biplanar X-Rays with Generative Adversarial Networks

XLSor: A Robust and Accurate Lung Segmentor on Chest X-Rays Using Criss-Cross Attention and Customized Radiorealistic Abnormalities Generation

Encoding CT Anatomy Knowledge for Unpaired Chest X-ray Image Decomposition

A Deep Learning Reconstruction Method for Sparsely Spaced Multidetector Helical CT Using Cascaded Dual-Domain CNN

Prior information-based high-resolution tomography image reconstruction from a single digitally reconstructed radiograph

Clinical ultra high resolution CT scans enabled by using a generative adversarial network

X-ray2CTPA: Generating 3D CTPA scans from 2D X-ray conditioning

X-CTRSNet: 3D cervical vertebra CT reconstruction and segmentation directly from 2D X-ray images

Super-Resolution Reconstruction of CT Images Based on Multi-scale Information Fused Generative Adversarial Networks

Segmentation of Pneumothorax on Chest CTs Using Deep Learning Based on Unet-Resnet-50 Convolutional Neural Network Structure

Deep multi‐instance transfer learning for pneumothorax classification in chest X‐ray images

CT Reconstruction from Few Planar X-rays with Application towards Low-resource Radiotherapy

Task-Specific Data Preparation for Deep Learning to Reconstruct Structures of Interest from Severely Truncated CBCT Data

DP-GAN+b: A lightweight generative adversarial network based on depthwise separable convolutions for generating CT volumes