Enhanced X-ray imaging of spheroids: an O(n) algorithm for characterizing convex blobs

Area of Science:

• Computational geometry
• Medical imaging analysis
• Image processing

Background:

• Accurate circular fitting is crucial for analyzing round objects in images.
• Existing methods may lack efficiency or precision for specific applications.
• Radiographic analysis for early cancer detection requires robust image understanding techniques.

Purpose of the Study:

• To present a simple and efficient O(n) algorithm for least squares circular fitting of tabulated functions.
• To demonstrate the algorithm's practical application using digitized grapefruit radiographs.
• To propose the algorithm's utility in medical imaging for early cancer detection, specifically coin lesions.

Main Methods:

• Developed a linear time complexity O(n) algorithm for circular data fitting.
• Applied the algorithm to analyze digitized, band-filtered radiographic images of a grapefruit.
• Utilized least squares estimation for precise circular parameter determination.

Main Results:

• The algorithm successfully performed circular fitting on image data.
• Demonstrated the algorithm's capability to analyze features in radiographic images.
• The analysis of grapefruit images served as a proof of concept for the method's application.

Conclusions:

• The O(n) least squares circular fit algorithm is computationally efficient.
• The method shows promise for image understanding tasks in medical diagnostics.
• Proposed for early radiographic detection and analysis of cancerous coin lesions.