Machine Learning in Quantitative PET Imaging

TL;DR

This paper reviews recent machine learning methods applied to quantitative PET imaging, focusing on attenuation correction and low-count reconstruction, comparing various approaches, performances, and discussing challenges and future directions.

Contribution

It provides a comprehensive summary and comparison of recent machine learning techniques in PET imaging, highlighting key developments and challenges.

Findings

Machine learning improves PET attenuation correction accuracy.

Deep learning enhances low-count PET image quality.

Current methods show promising performance but face challenges in clinical translation.

Abstract

This paper reviewed the machine learning-based studies for quantitative positron emission tomography (PET). Specifically, we summarized the recent developments of machine learning-based methods in PET attenuation correction and low-count PET reconstruction by listing and comparing the proposed methods, study designs and reported performances of the current published studies with brief discussion on representative studies. The contributions and challenges among the reviewed studies were summarized and highlighted in the discussion part followed by.