Why is image quality so important in Cardiac CT scans?

Cardiac computed tomography (CT) scans are a powerful tool, but the images can be affected by motion artifacts. These artifacts arise from the heart's constant movement, which blurs the images and makes it harder for doctors to see details. The presence of artifacts can make it difficult to assess the extent of plaque buildup or narrowing of the arteries (stenosis). This can lead to inaccurate diagnoses, impacting patient care.

How is Artificial Intelligence being used to improve Cardiac CT imaging?

Artificial intelligence (AI) is being developed to improve the quality of Cardiac CT images by assessing the images and identifying potential problems. The AI system analyzes the images and predicts assessability indices, or image quality scores. This helps doctors identify scans where the image quality is good enough for accurate diagnosis. The use of AI can lead to more objective and consistent interpretations, potentially reducing errors and improving patient outcomes.

What is the role of a cardiac phantom in this research?

The study used a physical, dynamic cardiac phantom to mimic the motion of a human heart and create calcified plaque images with varying degrees of motion artifacts. The phantom allowed researchers to generate controlled conditions to test the AI's ability to assess image quality. A 64-channel CT scanner was used to capture images at different heart rates, cardiac phases, and plaque locations. The AI then predicted assessability indices based on image features, and the AI-predicted indices were compared to the assessment of expert radiologists in identifying images with diagnostic calcium scores.

What is the significance of the 'assessability index' in this context?

The 'assessability index' is essentially an image quality score. The images are assigned a score from 1 (excellent) to 5 (very poor), reflecting the quality of the images and their suitability for diagnosis. This score helps doctors determine if the images are good enough to rely on for diagnosis. The AI-predicted assessability indices were compared to the expert radiologist's assessments to validate the AI's accuracy. The AI system generated these quality assessments much faster than human observers.

What is the overall significance of this AI research in cardiac imaging?

AI's potential to enhance Cardiac CT imaging lies in its ability to objectively assess image quality and identify potentially problematic scans. By providing a more efficient and consistent method for evaluating images, AI can assist doctors in making more accurate diagnoses, even when motion artifacts are present. The AI system offers the potential to improve the accuracy of diagnoses, which is essential for patients with heart conditions. This approach can lead to better patient outcomes, reduce healthcare costs, and improve the overall efficiency of cardiac imaging.

AI improving clarity of heart scan image.

Cardiac CT Scans: Can AI Help Doctors See More Clearly?

Theo Raines in Health & Wellness December 2025 • 5 min read.

"New research explores how artificial intelligence can assess image quality in cardiac CT scans, potentially improving the accuracy of diagnoses and reducing the impact of motion artifacts."

Cardiac computed tomography (CT) is a powerful tool for visualizing the heart and detecting problems like coronary artery disease. However, these scans are susceptible to motion artifacts, caused by the heart's continuous movement, which can significantly compromise image quality. These artifacts can obscure important details, making it difficult for doctors to accurately assess the extent of plaque buildup or stenosis (narrowing) in the arteries.

Currently, doctors rely on their expertise to evaluate the quality of cardiac CT images, deciding whether the artifacts are minor enough to still allow for a reliable diagnosis. This subjective assessment can be time-consuming and may vary between physicians, leading to potential inconsistencies in interpretation. Recognizing this challenge, researchers are exploring the use of computerized methods, particularly artificial intelligence (AI), to provide a more objective and efficient way to assess cardiac CT image quality.

This article delves into a recent study investigating the potential of AI to evaluate the diagnostic image quality of calcified plaque images in cardiac CT. The study focuses on a novel approach using an artificial neural network (ANN) to predict 'assessability indices' – essentially, image quality scores – and validates this method using a physical, dynamic cardiac phantom. We'll explore how this AI-powered assessment can help identify images with diagnostic calcium scores, offering the potential for more accurate and reliable cardiac diagnoses.

AI to the Rescue: How Computerized Assessment Works

AI improving clarity of heart scan image.

The study, conducted by researchers at the University of Chicago and Oregon Health and Science University, sought to validate an AI-based method for assessing the quality of cardiac CT images. The core idea is that better image quality leads to more accurate calcium scores, a key indicator of coronary artery disease. The researchers used a physical dynamic cardiac phantom – a device that mimics the motion of a human heart – to generate a range of calcified plaque images with varying degrees of motion artifacts.

Here’s a breakdown of the AI assessment process:

Image Acquisition: A 64-channel CT scanner was used to scan the dynamic cardiac phantom, capturing images at different heart rates, cardiac phases, and plaque locations.
Expert Evaluation: Two experienced radiologists independently assessed the quality of each image, assigning an 'assessability index' on a scale from 1 (excellent) to 5 (very poor).
AI Analysis: An artificial neural network (ANN) was trained to predict the assessability indices based on various image features, including morphological characteristics, intensity-based metrics, and dynamic features related to plaque motion.
Performance Comparison: The performance of the AI-predicted assessability indices was compared to that of the expert radiologists in identifying images with diagnostic calcium scores.

The results of the study were promising. The AI-predicted assessability indices performed similarly to those assigned by the expert radiologists in identifying images with diagnostic calcium scores. This suggests that AI can effectively learn to recognize the characteristics of high-quality cardiac CT images, potentially assisting doctors in making more informed diagnostic decisions. Moreover, the AI system generated these quality assessments much faster than human observers.

The Future of Cardiac Imaging: AI-Powered Clarity

This study provides compelling evidence that AI has the potential to play a significant role in improving the quality and reliability of cardiac CT imaging. By objectively assessing image quality and flagging potentially problematic scans, AI can assist doctors in making more accurate diagnoses, especially in cases where motion artifacts are present.

While this research focused on a physical phantom, the next step is to validate these findings in clinical settings using real patient data. If successful, AI-powered image quality assessment could become a standard tool in cardiac CT imaging, leading to better patient outcomes.

Beyond calcium scoring, AI could also be used to assess the quality of other cardiac CT parameters, such as the degree of stenosis or the characteristics of vulnerable plaques. Ultimately, the goal is to provide physicians with a comprehensive and reliable assessment of cardiac health, empowering them to make the best possible treatment decisions.