Stereoscopic X-Ray Reconstruction Algorithms for Preventing Radiation Induced Cancer

Overview

Although Computerized Axial Tomography (CT) scans are critical for cancer discovery and treatment, they pose a threat as a source of immense radiation – a cause of cancer itself. However, the radiation risks of CT scans can be reduced by a factor of a hundred by replacing full scans with an accurate stereoscopic reconstruction algorithm based on automated computation. This project developed a stereoscopic reconstruction tool with an accurate depth measurement feature, a vital step in building a better, safer alternative to CT scans. Furthermore, multi-perspective reconstruction algorithms were applied to recreate models from X-ray image pairs to develop the technique and evaluate its feasibility in replacing CT’s. A multi-perspective stereo reconstruction algorithm was developed in MATLAB, refined to calculate the distance between two points requiring only a stereo image pair and camera calibration parameters, and then compared to the manually calculated actual distance. Using the developed algorithm, the difference between each point pair was computed over a set of data points. Both goals of radiation reduction and increasing accuracy of measurements were achieved. A stereoscopic reconstruction algorithm can be implemented in the medical field to eventually replace CT scans for some diagnoses and procedures, computing depth with the same accuracy while eliminating radiation hazards. Further research will involve optimizing the algorithm for even better accuracy. Constructing such an algorithm could save 14,500 lives that are lost due to CT scans annually.

Competition history

• AJAS 2017 Category not listed

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• All Biomedical Engineering projects
• AJAS Biomedical Engineering projects
• Every AJAS 2017 project