4 - Role of AI techniques in enhancing multi-modality medical image fusion results

Abstract

The chapter signifies the benefits of using AI methods for medical image fusion of different modalities. The modality can be CT, MR-T1, MR-T2, PET, etc. depending on the suspected malignant region. The aim of fusion is to collaborate each modality's best information into a single image called as fused image. Depicting the decisive information on the resultant image (fused image) will help the oncologist in demarcation of the tumor volume. The image fusion process can be done upfront, i.e., carried on the whole image directly and at once. The other way is to decompose the images first using the decomposition methods like DWT, CWT, ST, HIS, CBF, etc. After the decomposition, fusion operation is carried out on each decomposed part. In the end, the decomposed fused parts form a single image using reconstruction. This chapter addresses the multimodality medical image fusion using Artificial Intelligence techniques like Fuzzy Logic and Adaptive Neuro-Fuzzy Inference System (ANFIS). For the evaluation of fusion results, researchers have used metrics: Conventional metrics like API, AG, ENTROPY, MI, etc. and objective metrics like QAB/F, LAB/F, and NAB/F. In this study, only conventional metrics are calculated. The study reveals that the AI techniques not only give better results, but their learning capabilities will likely make the future work self-driven.