Deep learning-enhanced nuclear medicine SPECT imaging applied to cardiac studies

Area of Science:

• Nuclear Medicine
• Artificial Intelligence
• Cardiovascular Imaging

Background:

• Deep learning (DL) is increasingly used in medical imaging for tasks like classification and denoising.
• Nuclear medicine, including Positron Emission Tomography (PET) and Single-Photon Emission Computed Tomography (SPECT), benefits from AI advancements.
• Applications of DL in cardiovascular disease using SPECT imaging are less explored compared to other medical domains.

Purpose of the Study:

• To review recent deep learning (DL) approaches applied to cardiac SPECT imaging.
• To identify and discuss key application domains and dominant techniques in DL for cardiac SPECT.
• To outline current limitations and future research directions for DL in this field.

Main Methods:

• A comprehensive literature search was conducted to identify relevant studies.
• Fifty-five studies focusing on DL in cardiac SPECT imaging were analyzed.
• Studies were categorized by application domain: diagnosis, attenuation correction, denoising, full-count estimation, and reconstruction.

Main Results:

• DL has diverse applications in cardiac SPECT, including disease diagnosis, attenuation correction, image denoising, and reconstruction.
• Key findings and dominant DL techniques employed across these applications were identified.
• The review synthesized information from 55 relevant research papers.

Conclusions:

• Deep learning offers significant potential for advancing cardiac SPECT imaging analysis and quality.
• Further research is needed to address current limitations and fully leverage DL for cardiovascular applications.
• The review provides a roadmap for future DL research in cardiac SPECT imaging.