Synthetic CT generation from MRI using 3D transformer‐based denoising diffusion model

Medical Physics, Год журнала: 2023, Номер 51(4), С. 2538 - 2548

Опубликована: Ноя. 27, 2023

Abstract Background and purpose Magnetic resonance imaging (MRI)‐based synthetic computed tomography (sCT) simplifies radiation therapy treatment planning by eliminating the need for CT simulation error‐prone image registration, ultimately reducing patient dose setup uncertainty. In this work, we propose a MRI‐to‐CT transformer‐based improved denoising diffusion probabilistic model (MC‐IDDPM) to translate MRI into high‐quality sCT facilitate planning. Methods MC‐IDDPM implements processes with shifted‐window transformer network generate from MRI. The proposed consists of two processes: forward process, which involves adding Gaussian noise real scans create noisy images, reverse in V‐net (Swin‐Vnet) denoises conditioned on same produce noise‐free scans. With an optimally trained Swin‐Vnet, process was used matching anatomy. We evaluated method generating institutional brain dataset prostate dataset. Quantitative evaluations were conducted using several metrics, including Mean Absolute Error (MAE), Peak Signal‐to‐Noise Ratio (PSNR), Multi‐scale Structure Similarity Index (SSIM), Normalized Cross Correlation (NCC). Dosimetry analyses also performed, comparisons mean target coverages 95% 99%. Results generated sCTs state‐of‐the‐art quantitative results MAE 48.825 ± 21.491 HU, PSNR 26.491 2.814 dB, SSIM 0.947 0.032, NCC 0.976 0.019. For dataset: 55.124 9.414 28.708 2.112 0.878 0.040, 0.940 0.039. demonstrates statistically significant improvement (with p < 0.05) most metrics when compared competing networks, both CT. indicated that coverage differences within 0.34%. Conclusions have developed validated novel approach images routine MRIs DDPM. This effectively captures complex relationship between allowing robust be matter minutes. has potential greatly simplify additional scans, amount time patients spend planning, enhancing accuracy delivery.

Язык: Английский

---

CT synthesis from MR in the pelvic area using Residual Transformer Conditional GAN

Computerized Medical Imaging and Graphics, Год журнала: 2022, Номер 103, С. 102150 - 102150

Опубликована: Ноя. 29, 2022

Язык: Английский

---

MRI-LINAC: A transformative technology in radiation oncology

Frontiers in Oncology, Год журнала: 2023, Номер 13

Опубликована: Янв. 27, 2023

Advances in radiotherapy technologies have enabled more precise target guidance, improved treatment verification, and greater control versatility radiation delivery. Amongst the recent novel technologies, Magnetic Resonance Imaging (MRI) guided (MRgRT) may hold greatest potential to improve therapeutic gains of image-guided delivery dose. The ability MRI linear accelerator (LINAC) image tumors organs with on-table MRI, manage organ motion dose real-time, adapt plan on day while patient is table are major advances relative current conventional treatments. These advanced techniques demand efficient coordination communication between members team. MRgRT could fundamentally transform process within oncology centers through reorganization team workflow process. However, technology currently limited by accessibility due cost capital investment time personnel allocation needed for each fractional unclear clinical benefit compared platforms. As evolves becomes widely available, we present case that has become a utilized platform just as earlier disruptive therapy done.

Язык: Английский

---

Review and recommendations on deformable image registration uncertainties for radiotherapy applications

Physics in Medicine and Biology, Год журнала: 2023, Номер 68(24), С. 24TR01 - 24TR01

Опубликована: Ноя. 16, 2023

Deformable image registration (DIR) is a versatile tool used in many applications radiotherapy (RT). DIR algorithms have been implemented commercial treatment planning systems providing accessible and easy-to-use solutions. However, the geometric uncertainty of can be large difficult to quantify, resulting barriers clinical practice. Currently, there no agreement RT community on how quantify these uncertainties determine thresholds that distinguish good result from poor one. This review summarises current literature sources their impact applications. Recommendations are provided handle for patient-specific use, commissioning, research. also developers vendors help users understand make application safer more reliable.

Язык: Английский

---

Advances of Layered Double Hydroxide‐Based Materials for Tumor Imaging and Therapy

The Chemical Record, Год журнала: 2024, Номер 24(4)

Опубликована: Март 19, 2024

Abstract Layered double hydroxides (LDH) are a class of functional anionic clays that typically consist orthorhombic arrays metal with anions sandwiched between the layers. Due to their unique properties, including high chemical stability, good biocompatibility, controlled drug loading, and enhanced bioavailability, LDHs have many potential applications in medical field. Especially fields bioimaging tumor therapy. This paper reviews research progress nanocomposites field imaging First, structure advantages LDH discussed. Then, several commonly used methods for preparation presented, co‐precipitation, hydrothermal ion exchange methods. Subsequently, recent advances layered cancer therapy highlighted. Finally, based on current research, we summaries prospects challenges diagnosis

Язык: Английский

---

Challenges and opportunities in the development and clinical implementation of artificial intelligence based synthetic computed tomography for magnetic resonance only radiotherapy

Radiotherapy and Oncology, Год журнала: 2024, Номер 198, С. 110387 - 110387

Опубликована: Июнь 15, 2024

Язык: Английский

---

Balancing data consistency and diversity: Preprocessing and online data augmentation for multi-center deep learning-based MR-to-CT synthesis

Pattern Recognition Letters, Год журнала: 2025, Номер 189, С. 56 - 63

Опубликована: Янв. 10, 2025

Язык: Английский

---

Development of an MR-only radiotherapy treatment planning workflow using a commercial synthetic CT generator for brain and head & neck tumor patients

Zeitschrift für Medizinische Physik, Год журнала: 2025, Номер unknown

Опубликована: Фев. 1, 2025

In magnetic resonance (MR)-only radiotherapy (RT) workflows, synthetic computed tomography images (sCT) are needed as a surrogate for dose calculation. Commercial and certified sCT algorithms became recently available, but many have not been evaluated in clinical setting, especially the head neck tumor (HN) region. this study, an MRI-only workflow using commercial generator photon beam therapy brain HN body sites was terms of calculation accuracy, modelling immobilization devices, well usability autosegmentation. For 13 10 cancer patients, MR scans T1W mDIXON sequences were retrospectively collected. Four all patients scanned RT treatment position with devices. All MRIs converted to MRCAT algorithm (Philips, Eindhoven, The Netherlands). underwent standard planning CT (pCT) segmentation VMAT planning. rigidly registered pCT contours transferred sCT. dosimetric evaluation based calculation, plans recalculated on D1% Dmean compared structures between sCT, D95%, D98% targets only. MR-invisible device modelling, MR-visible markers placed into geometric robustness analysis performed same target dose-volume parameters. organs-at-risk (OARs) autosegmentation, both autosegmented clinically established CT-based autocontouring software. agreement analyzed by similar parameters dice similarity (DSC) Hausforff distance (HD). overall median deviation (± interquartile range) including model 1.1 ± 0.4% volumes, 1.3 1.2% OAR, 0.4 0.7% volumes 0.9% OAR. over autocontours resulted DSC = 0.82 OAR 0.79 MR-only software package feasible tumors, acceptable accuracy. devices could be modelled system autosegmentation sCTs tool feasible.

Язык: Английский

---

Classical and Learned MR to Pseudo-CT Mappings for Accurate Transcranial Ultrasound Simulation

IEEE Transactions on Ultrasonics Ferroelectrics and Frequency Control, Год журнала: 2022, Номер 69(10), С. 2896 - 2905

Опубликована: Авг. 19, 2022

Model-based treatment planning for transcranial ultrasound therapy typically involves mapping the acoustic properties of skull from an X-ray computed tomography (CT) image head. Here, three methods generating pseudo-CT (pCT) images magnetic resonance (MR) were compared as alternative to CT. A convolutional neural network (U-Net) was trained on paired MR-CT generate pCT T either T1-weighted or zero-echo time (ZTE) MR (denoted tCT and zCT, respectively). direct ZTE also implemented cCT). When comparing ground-truth CT test set, mean absolute error 133, 83, 145 Hounsfield units (HU) across whole head, 398, 222, 336 HU within tCT, cCT images, respectively. Ultrasound simulations performed using generated based An annular array transducer used targeting visual motor cortex. The differences in simulated focal pressure, position, volume 9.9%, 1.5 mm, 15.1% images; 5.7%, 0.6 5.7% zCT; 6.7%, 0.9 12.1% cCT. improved results mapped highlight advantage imaging sequences, which improves contrast bone. Overall, these demonstrate that can give comparable accuracy those

Язык: Английский

---

Artificial intelligence in radiotherapy

Seminars in Cancer Biology, Год журнала: 2022, Номер 86, С. 160 - 171

Опубликована: Авг. 20, 2022

Язык: Английский

---