Lung Cancer Detection in Chest X-Ray Images Empowered by 3D Computed Tomography Deep Convolutional Radiomics (CXRClear)
Cancer is treatable if it is discovered at an early stage, and lung cancer screening is a critical component in a preventive care protocol. Although CT imaging affords higher spatial resolution and 3D density information than digital chest X-rays, there are still limitations to having it as a cheap and fast method for rural areas outreach. These limitations are outlined in cost, limited access, and portability issues. Chest X-rays modality has been developed recently with the preface of high-resolution digital X-rays however, the smallest observable nodule size is limited to 1-2 cm. To detect such small-size cancer at an early stage, CT is the ideal solution.
The overall objectives achieved in the project are as follows:
- Develop a practical lung cancer detection system that resolves the issue related to detecting small and medium-sized tumors in Chest X-rays.
- Identify the stratification features that define small, medium, and large-size cancer from CT.
- Develop a projection of those features to CXR data so we can perform screening and detection of different-size cancer areas from CXR standalone.
- Unified Dataset has been created.
- Assistive tools such as rib suppression have been developed to assist with X-ray scan reading.
Kareem Elgohary, Samar Ibrahim, Sahar Selim and Mustafa Elattar (2023). A CAD system for lung cancer detection using Chest X-ray: A Review. In: Fournier-Viger, P., Hassan, A., Bellatreche, L. (eds) Model and Data Engineering. MEDI 2022. Communications in Computer and Information Science. Springer, Cham.
Ibrahim, S., Elgohary, K., Higazy, M., Mohannad, T., Selim, S., Elattar, M. (2022). Lung Segmentation Using ResUnet++ Powered by Variational Auto Encoder-Based Enhancement in Chest X-ray Images. In: Yang, G., Aviles-Rivero, A., Roberts, M., Schönlieb, CB. (eds) Medical Image Understanding and Analysis. MIUA 2022. Lecture Notes in Computer Science, vol 13413. Springer, Cham.