Machine Learning in Cardiovascular Imaging: A Scoping Review of Published Literature

Area of Science:

• Cardiovascular Imaging
• Machine Learning
• Artificial Intelligence

Background:

• Machine learning (ML), a subset of artificial intelligence (AI), empowers computers to learn from data, mimicking human decision-making.
• The healthcare sector is increasingly adopting ML due to its innovative applications.
• Cardiovascular imaging (CVI) is a dynamic field within medical imaging ripe for technological integration, particularly ML.

Purpose of the Study:

• To conduct a scoping review of the literature on machine learning (ML) applications in cardiovascular imaging (CVI).
• To understand the current landscape of ML investigations within CVI.
• To identify trends and provide recommendations for future research.

Main Methods:

• A comprehensive literature search was performed to identify studies investigating ML in CVI.
• Quantitative analysis of study characteristics, data handling, model development, and performance evaluation.
• Qualitative synthesis to identify common themes and inform future directions.

Main Results:

• Numerous studies utilize or develop ML models for segmentation, classification, object detection, generation, and regression in CVI.
• Analysis revealed diverse approaches in study design, data management, and model evaluation.
• Identified common themes and research gaps in the current literature.

Conclusions:

• ML is a rapidly growing area in cardiovascular imaging research.
• Further research is needed to standardize methodologies and enhance performance evaluation.
• Recommendations are provided to guide future ML investigations in CVI.