Lung Region Segmentation and 3D Visualization from CT Scans: A Reproducible Python Implementation

Current segmentation and precise RECIST practises are carried out either manually or semi-automatically, aside from possible human error in the loop, this takes up valuable time and resources. Hence, this paper showcases a fully automated methodology for the detection and 3D segmentation of non-small cell lung cancer (NSCLC). The used method is from a nature communications paper that presents this method along with actual figures, institutions’ datasets validations and a clinical trial setting. This is a Python step by step tutorial and explanation of the aforementioned automated method for the user to implement into their projects. This paper presents the code, its corresponding output along with the process explanation. Additionally, there is also a 3D interactive model created after all the code implementation for a better look at the provided lungs’ images.