Physica medica : PM : an international journal devoted to the applications of physics to medicine and biology : official journal of the Italian Association of Biomedical Physics (AIFB) · 2026
Fast MRI of bones in the knee - An AI-driven reconstruction approach for adiabatic inversion recovery prepared ultra-short echo time sequences.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Nunn Philipp Hans, Huflage Henner, Grunz Jan-Peter, Gruschwitz Philipp, Schad Oliver, Bley Thorsten Alexander, Tran-Gia Johannes, Wech Tobias
Аннотация
PURPOSE: Inversion recovery prepared ultra-short echo time (IR-UTE)-based MRI enables radiation-free visualization of osseous tissue. However, achieving a sufficient signal-to-noise ratio typically requires long acquisition times. We report on a feasibility study, which proposes a data-driven approach to reconstruct undersampled IR-UTE knee data, thereby accelerating MR-based 3D imaging of bones.
METHODS: Data were acquired with a 3D radial IR-UTE pulse sequence, implemented using the open-source framework Pulseq. A denoising convolutional neural network (DnCNN) was trained in a supervised fashion using data from eight healthy subjects. Conjugate gradient sensitivity encoding (CG-SENSE) reconstructions of different retrospectively undersampled subsets (corresponding to 2.5-min, 5-min, and 10-min acquisition times) were paired with the respective reference dataset reconstruction (30-min acquisition time). The DnCNN was then integrated into a FISTA-based reconstruction algorithm, enabling physics-based iterative reconstruction. Quantitative evaluation was performed on retrospectively undersampled datasets from four additional healthy subjects using scalar metric calculations and an expert reader study. Metrics were also assessed for one prospectively accelerated scan. A pathological case of a tibial plateau fracture was included for qualitative demonstration.
RESULTS: The trained DnCNN enabled effective noise suppression, and its application exhibited the most favorable quantitative results, whereas the iterative reconstruction scheme provided complementary mitigation of denoising-induced streaking artifacts, particularly for 5-min. Fracture features could be visualized in the patient case, albeit in less detail compared to photon-counting CT.
CONCLUSION: Utilizing a task-specific trained DnCNN shows potential to shorten scan times for hyperintense MR-based imaging of bone, thereby addressing a key hurdle to clinical implementation.
METHODS: Data were acquired with a 3D radial IR-UTE pulse sequence, implemented using the open-source framework Pulseq. A denoising convolutional neural network (DnCNN) was trained in a supervised fashion using data from eight healthy subjects. Conjugate gradient sensitivity encoding (CG-SENSE) reconstructions of different retrospectively undersampled subsets (corresponding to 2.5-min, 5-min, and 10-min acquisition times) were paired with the respective reference dataset reconstruction (30-min acquisition time). The DnCNN was then integrated into a FISTA-based reconstruction algorithm, enabling physics-based iterative reconstruction. Quantitative evaluation was performed on retrospectively undersampled datasets from four additional healthy subjects using scalar metric calculations and an expert reader study. Metrics were also assessed for one prospectively accelerated scan. A pathological case of a tibial plateau fracture was included for qualitative demonstration.
RESULTS: The trained DnCNN enabled effective noise suppression, and its application exhibited the most favorable quantitative results, whereas the iterative reconstruction scheme provided complementary mitigation of denoising-induced streaking artifacts, particularly for 5-min. Fracture features could be visualized in the patient case, albeit in less detail compared to photon-counting CT.
CONCLUSION: Utilizing a task-specific trained DnCNN shows potential to shorten scan times for hyperintense MR-based imaging of bone, thereby addressing a key hurdle to clinical implementation.