3D IMAGE RECONSTRUCTION USING ACO-BASED TVR-DART

Abstract

The ability to clearly see inside structures is crucial for doctors to make accurate diagnoses and provide effective treatments. A powerful
3D brain imaging was essential for the radiologist in both the diagnosis and subsequent procedures for several brain tumours. Brain
tumour detection via MRI and subsequent image restoration is a computationally intensive and unpredictable process. Due to the
limitations of 2D imaging, 3D tumour reconstruction is necessary for studies and therapeutic planning. Due to the tumour’s intricacy
and variety, MRI imaging is often unsuccessful. Particularly in the realm of biomedical imaging, 3D image reconstruction has emerged
as one of the most promising paths for the processing of digital inputs. The research resulted in a methodical and effective strategy for
3D restoration. It involves combining many processes, including picture pre-processing, image segmentation, 3D model advancement,
and tumour reconstruction. In this work, we introduce the total-variation regularised discrete algebraic reconstruction technique (TVR
DART) algorithm, which uses ant colony optimization (ACO) to perform reconstruction, and the modified fuzzy c means segmentation
clustering (MFCM) method, which uses a supervised learning method.