Biophysics-informed deep learning framework for magnetic resonance imaging

基于生物物理学的磁共振成像深度学习框架

• 批准号: DP210101863
• 负责人: A/Prof Zhaolin Chen
• 金额: $ 36.43万
• 依托单位: Monash University
• 依托单位国家: 澳大利亚
• 项目类别: Discovery Projects
• 财政年份: 2022
• 资助国家: 澳大利亚
• 起止时间: 2022-01-01 至 2024-12-31
• 项目状态: 已结题
• 来源: https://dataportal.arc.gov.au/RGS/Web/Grants/DP210101863
• 关键词: Biophysics; informed; deep; learning; framework

This project aims to bring about a paradigm shift from the conventional non-quantitative magnetic resonance imaging to ultra-fast, quantitative, and artefact free imaging. This project integrates biophysics and artificial intelligence, and it is expected to bring new knowledge in both fields. The expected outcomes of this project include next generation magnetic resonance imaging methods with a fundamental shift in the approach to image artefacts and image quantification. This project is expected to advance both single subject and population level biomedical imaging with greater accuracy and cost-effectiveness. This project also promotes explainable and generalisable artificial intelligence in medical imaging.

该项目旨在实现从传统的非定量磁共振成像到超快速，定量和无伪影成像的范式转变。该项目整合了生物物理学和人工智能，预计将在这两个领域带来新的知识。 该项目的预期成果包括下一代磁共振成像方法，在图像伪影和图像量化方法方面有根本性的转变。预计该项目将以更高的准确性和成本效益推进单个受试者和人群水平的生物医学成像。该项目还促进了医学成像中可解释和可推广的人工智能。