Improvement of image quality at CT and MRI using deep learning

Area of Science:

• Computer Science
• Medical Imaging
• Radiology

Background:

• Deep learning, a subset of artificial intelligence, has advanced computational capabilities.
• Diagnostic imaging modalities like computed tomography (CT) and magnetic resonance imaging (MRI) are crucial for clinical diagnosis.
• Image quality in CT and MRI directly impacts diagnostic accuracy.

Purpose of the Study:

• To review and categorize deep learning techniques for enhancing medical image quality.
• To highlight the application of deep learning in noise and artifact reduction, super-resolution, and image acquisition/reconstruction for CT and MRI.

Main Methods:

• Categorization of deep learning techniques based on their application in image quality enhancement.
• Review of existing studies and their methodologies in applying deep learning to medical imaging.
• Outline of the features and contributions of selected studies within each category.

Main Results:

• Deep learning effectively reduces noise and artifacts in CT and MRI scans.
• Super-resolution techniques using deep learning significantly improve image detail and clarity.
• Advanced deep learning methods enhance image acquisition and reconstruction processes, leading to better diagnostic information.

Conclusions:

• Deep learning offers powerful tools for improving diagnostic CT and MRI image quality.
• These advancements hold significant potential for more accurate and efficient medical diagnoses.
• Further research and application of deep learning are expected to revolutionize medical imaging.