Keywords: Machine Learning/Artificial Intelligence, AI/ML Image Reconstruction
Motivation: FLAIR is valuable for detecting brain abnormalities but often has low SNR, resulting in longer scan-times. A novel approach is needed to enhance image quality without extending scan-time, particularly for detecting white-matter hyperintensities(WMH).
Goal(s): This study assesses super-resolution deep-learning reconstruction(SR-DLR) to improve FLAIR image sharpness, SNR, CNR, and WMH segmentation accuracy, comparing it to DLR and conventional methods.
Approach: Fourteen-subjects underwent FLAIR, reconstructed using conventional, DLR, and SR-DLR. We assessed image-noise, SNR, CNR, sharpness, and WMH segmentation in a case with moderate hyperintensity.
Results: SR-DLR outperformed conventional and DLR methods in SNR, CNR, and sharpness, improving WMH segmentation precision and overall diagnostic utility.
Impact: The SR-DLR method could transform FLAIR MRI, enabling improved detection and quantification of white matter lesions without extended scan times. This advancement may influence imaging standards for brain pathology and drive new insights into neuro-degenerative disease diagnosis.
How to access this content:
For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.
After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.
After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.
Keywords