- AI
- Molecular Imaging
- CT
- X-Ray
- Ultrasound
- MRI
- Facility Management
- Mammography
The Reading Room: NYU Langone and Facebook AI Partner for Faster MRIs
In this episode, Dr. Michael Recht, chair of the radiology department at New York University Langone Health, discusses the partnership between NYU and Facebook AI that has created an accelerated MRI that is four times as fast as a standard MRI.
Through a partnership with Facebook AI, researchers at NYU Langone Health have trained a neural network that can create MRI images in a fraction of the time needed for standard scans.
In a study published in the American Journal of Roentgenology, the team revealed the radiologists reading the resulting images detected the same abnormalities with the accelerated scans as they did with traditional images.
For more coverage based on industry expert insights and research, subscribe to the Diagnostic Imaging e-Newsletter here.
To reach this determination, they examined two sets of knee MRI scans from 108 patients and created two sets of images -- one using the standard approach and one using the accelerated MRI artificial intelligence model. Based on their analysis, they determined that their 3T fastMRI protocol used four times less data and could produce an image quicker, requiring the patient to spend less time in the scanner.
The goal, Recht said, is to one day apply accelerated MRI to other organs, including the brain.
New Study Examines Short-Term Consistency of Large Language Models in Radiology
November 22nd 2024While GPT-4 demonstrated higher overall accuracy than other large language models in answering ACR Diagnostic in Training Exam multiple-choice questions, researchers noted an eight percent decrease in GPT-4’s accuracy rate from the first month to the third month of the study.
