Article
Engineering, Biomedical
You Zhang, Hua-Chieh Shao, Tinsu Pan, Tielige Mengke
Summary: In this study, we propose a simultaneous spatial and temporal implicit neural representation (STINR) method for dynamic cone-beam CT (CBCT) reconstruction. The STINR method maps the unknown image and its motion into spatial and temporal multi-layer perceptrons (MLPs) to represent the dynamic CBCT series. Experimental results demonstrate that STINR achieves higher image reconstruction and motion tracking accuracy compared to traditional methods based on PCA and polynomial fitting-based neural representation.
PHYSICS IN MEDICINE AND BIOLOGY
(2023)
Article
Oncology
Abigail Bryce-Atkinson, Rianne De Jong, Tom Marchant, Gillian Whitfield, Marianne C. Aznar, Arjan Bel, Marcel van Herk
Summary: The study analyzed noise contributions in paediatric CBCT, recommending practical imaging protocols and dose thresholds. Anatomical noise was found to be the major factor affecting image quality in the abdominal/pelvic region, with appropriate soft tissue contrast and registration accuracy achievable at doses as low as 1 mGy. Increasing dose above 1 mGy does not provide any benefit in improving image quality or registration accuracy due to the presence of anatomical noise.
RADIOTHERAPY AND ONCOLOGY
(2021)
Article
Dentistry, Oral Surgery & Medicine
Joao Pedro de Lima, Jardel Francisco Mazzi-Chaves, Manoel Damiao de Sousa-Neto, Amanda Pelegrin Candemil
Summary: This study evaluated the accuracy of an optimized CBCT protocol for the detection of intraoperative endodontic complications. The results showed that there were no significant differences in accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve between the dose protocols, regardless of the presence of metallic materials.
JOURNAL OF ENDODONTICS
(2023)
Article
Computer Science, Artificial Intelligence
Feng Zhang, Jianjun Wang, Wendong Wang, Chen Xu
Summary: In this paper, a Modified-TPCP method is proposed which incorporates prior subspace information to recover the structure of data tensors under weaker incoherence assumptions. An efficient algorithm based on ADMM is designed for implementation, with promising performance demonstrated through simulations and real data applications.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2021)
Article
Computer Science, Artificial Intelligence
Weichao Kong, Feng Zhang, Wenjin Qin, Jianjun Wang
Summary: This article proposes a novel strategy to flexibly utilize accessible subspace prior information for optimizing low-rank tensor recovery. A multilayer subspace prior learning scheme is designed and applied to tensor completion and tensor robust component principal analysis. It is proven that the proposed approach achieves exact recovery of tensors under weaker incoherence assumption than previous conditions. Experimental results demonstrate that the proposed algorithms outperform other state-of-the-art algorithms in terms of both qualitative and quantitative metrics.
PATTERN RECOGNITION
(2023)
Article
Clinical Neurology
R. A. Helal, R. Jacob, M. A. Elshinnawy, A. Othman, I. M. Al-Dhamari, D. W. Paulus, T. T. Abdelaziz
Summary: In the comparison between cone-beam CT and multidetector CT for evaluating postoperative implant placement and anatomical structures in head and neck imaging, cone-beam CT showed higher sensitivity in visualizing electrode positions, cochlear outer walls, single electrode contacts, and measuring insertion angles.
AMERICAN JOURNAL OF NEURORADIOLOGY
(2021)
Article
Engineering, Electrical & Electronic
Phil Howlett, Anatoli Torokhti, Peter Pudney, Pablo Soto-Quiros
Summary: This paper introduces a multilinear filter for distributed signal processing and its associated techniques, including multilinear KLT-1 and multilinear KLT-2. These techniques help reduce the dimensionality of observed signals, improve accuracy, and decrease computational cost by breaking down the original problem into smaller matrix problems.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Zhaoyue Chen, Jianzhong Wang
Summary: A significant reduction in effective dose for CBCT-DCG in healthy volunteers is achievable by decreasing the scanning field of view (FOV) without compromising diagnostic capability. CBCT-DCG is a potential first-line imaging test for evaluating the lacrimal drainage system, allowing for functional assessment of lacrimal drainage.
Article
Engineering, Biomedical
Xiaoyu Hu, Yuncheng Zhong, Youfang Lai, Chenyang Shen, Kai Yang, Xun Jia
Summary: This study developed a PCD-based ME-CBCT to improve the accuracy of material differentiation and dose calculation. Compared to conventional FPD-based CBCT, PCD-based ME-CBCT showed lower errors in dose calculation for bone and soft tissue regions.
PHYSICS IN MEDICINE AND BIOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Kihwan Choi, Seung Hyoung Kim, Sungwon Kim
Summary: In this paper, a self-supervised learning method is proposed to reduce noise in projections acquired by ordinary CBCT scans. By partially blinding the input, a network is able to train the denoising model by mapping the partially blinded projections to the original projections. Noise-to-noise learning is also incorporated into the self-supervised learning by mapping adjacent projections to the original projections.
Review
Radiology, Nuclear Medicine & Medical Imaging
Abeer A. Almashraqi, Boshra A. Sayed, Lujain K. Mokli, Sarah A. Jaafari, Esam Halboub, Sameena Parveen, Mohammed Sultan Al-Ak'hali, Maged S. Alhammadi
Summary: This systematic review evaluated the reliability and comprehensiveness of imaging methods for three-dimensional assessment of the temporomandibular joint (TMJ), and proposed a standardized imaging method. The results showed that CBCT-based methods for assessing the positions and morphology of TMJ bony structures are imperfect and lack comprehensiveness. Therefore, criteria for a standardized assessment method of these TMJ structures are proposed for future research to improve validity and reliability.
EUROPEAN RADIOLOGY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Philipp Feldle, Jan-Peter Grunz, Andreas Steven Kunz, Theresa Sophie Patzer, Henner Huflage, Robin Hendel, Karsten Sebastian Luetkens, Suleyman Erguen, Thorsten Alexander Bley, Nora Conrads
Summary: This study aimed to evaluate the feasibility of weight-bearing cone-beam CT (CBCT) of the lumbar spine using a gantry-free scanner system and determine the most effective combination of scan parameters. The results showed that increasing the frame rate improved image quality and assessability of the posterior wall. However, tube voltage and dose level did not significantly affect reader assessment. In conclusion, optimized protocols for weight-bearing gantry-free CBCT of the lumbar spine allow for diagnostic imaging with reasonable radiation dose.
EUROPEAN JOURNAL OF RADIOLOGY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Benjamin K. F. Lau, Owen Dillon, Shalini K. K. Vinod, Ricky T. T. O'Brien, Tess Reynolds
Summary: This study investigates the impact of gantry velocity and angular separation between x-ray projections on image quality for fast low-dose 4DCBCT imaging. The results suggest that very fast gantry rotations have minimal effect on image quality and can be achieved with motion-compensated reconstruction.
Article
Materials Science, Characterization & Testing
Soomin Jeon, Seongeun Kim, Chang-Ock Lee
Summary: Metal artefact reduction is a challenging issue in X-ray CT applications due to contrast degradation and misinterpretation caused by these artefacts. The proposed methodology in this paper successfully reduces metal artefacts through a registration technique to find accurate segmentation regions. Experimentation and simulations confirm the effectiveness of the algorithm.
NONDESTRUCTIVE TESTING AND EVALUATION
(2021)
Article
Computer Science, Information Systems
Lisiqi Xie, Kangjian He, Jian Gong, Dan Xu
Summary: In this paper, we propose a novel multi-intensity optimization-based CT and cone beam CT registration method, which effectively solves the issues of position errors caused by breathing and heartbeat, as well as insufficient imaging clarity. The experimental results demonstrate that the proposed method meets the requirements of image-guided radiotherapy.
Article
Medicine, Research & Experimental
Nicholas A. George-Jones, Kai Wang, Jing Wang, Jacob B. Hunter
Summary: The study demonstrated excellent agreement, higher sensitivity, specificity, and AUC with the automated segmentation method compared to using the greatest linear dimension for growth detection. The convolutional neural network model outperformed traditional methods, showing potential for artificial intelligence applications in surveillance of vestibular schwannomas.
Article
Engineering, Biomedical
Shanzhou Niu, Hong Liu, Mengzhen Zhang, Min Wang, Jing Wang, Jianhua Ma
Summary: The study introduces a new statistical iterative reconstruction method, PWLS-PIDT, to improve the image quality of low-dose CPCT by utilizing a prior image and preserving important features in the target image. Experimental results show that the noise in CPCT images reconstructed using the PWLS-PIDT method is significantly reduced, without sacrificing structural details.
PHYSICS IN MEDICINE AND BIOLOGY
(2021)
Article
Engineering, Biomedical
Liyuan Chen, Xiao Liang, Chenyang Shen, Dan Nguyen, Steve Jiang, Jing Wang
Summary: Adaptive radiation therapy commonly uses CBCT for patient positioning, but its inaccuracy in Hounsfield units limits its application to dose calculation. Our proposed unsupervised style-transfer-based method successfully generates sCT with high CT image quality by combining CBCT and pCT.
PHYSICS IN MEDICINE AND BIOLOGY
(2021)
Article
Engineering, Biomedical
Rongfang Wang, Jinkun Guo, Zhiguo Zhou, Kai Wang, Shuiping Gou, Rongbin Xu, David Sher, Jing Wang
Summary: The study aims to develop an end-to-end multi-modality and multi-view feature extension method (MMFE) to predict LRR in H&N cancer. By utilizing a multi-modality convolutional neural network, the MMFE outperforms other methods in prediction accuracy, potentially identifying H&N cancer patients at high risk for LRR.
PHYSICS IN MEDICINE AND BIOLOGY
(2022)
Article
Engineering, Biomedical
Hua-Chieh Shao, Jing Wang, Ti Bai, Jaehee Chun, Justin C. Park, Steve Jiang, You Zhang
Summary: This study proposes a framework combining graph neural network-based deep learning and biomechanical modeling to track liver tumor in real-time from a single onboard x-ray projection, achieving high accuracy.
PHYSICS IN MEDICINE AND BIOLOGY
(2022)
Article
Engineering, Biomedical
Tao Peng, Caishan Wang, You Zhang, Jing Wang
Summary: This article introduces an automatic Hybrid Segmentation Network (H-SegNet) for lung segmentation on CXR images, which achieves superior segmentation results compared to several existing methods.
PHYSICS IN MEDICINE AND BIOLOGY
(2022)
Article
Oncology
Jason W. Chan, Nicole Hohenstein, Colin Carpenter, Adam J. Pattison, Olivier Morin, Gilmer Valdes, Cristina Tolentino Sanchez, Jennifer Perkins, Timothy D. Solberg, Sue S. Yom
Summary: The study aimed to develop a novel AI-guided clinical decision support system to predict radiation doses to subsites of the mandible using diagnostic CT scans before head and neck radiation therapy planning. Results showed that the AI predictions had a high accuracy in the test data set and strong correlation with physician predictions.
ADVANCES IN RADIATION ONCOLOGY
(2022)
Article
Engineering, Biomedical
Doran Wood, Sila Cetinkaya, Harsha Gangammanavar, Weigo Lu, Jing Wang
Summary: This study aims to develop optimization models and methods that adapt treatment decisions across multiple fractions by utilizing predictions of tumor evolution. By introducing a nonuniform allocation scheme, we demonstrate the superiority of this approach across multiple performance metrics.
PHYSICS IN MEDICINE AND BIOLOGY
(2022)
Article
Engineering, Biomedical
Hua-Chieh Shao, Tian Li, Michael J. Dohopolski, Jing Wang, Jing Cai, Jun Tan, Kai Wang, You Zhang
Summary: KS-RegNet is an end-to-end unsupervised network that utilizes deep learning and prior information to achieve real-time 3D MR imaging and provide high-quality images for motion tracking.
PHYSICS IN MEDICINE AND BIOLOGY
(2022)
Article
Biochemistry & Molecular Biology
Raquibul Hannan, Michael J. Dohopolski, Laurentiu M. Pop, Samantha Mannala, Lori Watumull, Dana Mathews, Ang Gao, Aurelie Garant, Yull E. Arriaga, Isaac Bowman, Jin-Sung Chung, Jing Wang, Kiyoshi Ariizumi, Chul Ahn, Robert Timmerman, Kevin Courtney
Summary: The addition of concurrent stereotactic ablative radiotherapy (SAbR) did not significantly increase the time to progression in patients with metastatic castrate-resistant prostate cancer (mCRPCA) treated with sipuleucel-T, but induced cellular and humoral immune responses.
Editorial Material
Radiology, Nuclear Medicine & Medical Imaging
Timothy D. Solberg
JOURNAL OF APPLIED CLINICAL MEDICAL PHYSICS
(2022)
Article
Engineering, Biomedical
Hua-Chieh Shao, Yunxiang Li, Jing Wang, Steve Jiang, You Zhang
Summary: A deep learning-based framework (Surf-X-Bio) was proposed to track the real-time 3D liver tumor motion using combined optical surface image and a single on-board x-ray projection. The results showed that Surf-X-Bio achieved higher accuracy and better robustness in tumor localization.
PHYSICS IN MEDICINE AND BIOLOGY
(2023)
Article
Engineering, Biomedical
Hengtao Guo, Hanqing Chao, Sheng Xu, Bradford J. Wood, Jing Wang, Pingkun Yan
Summary: This paper presents a novel deep learning approach, named deep contextual-contrastive network (DC2-Net), for sensorless ultrasound volume reconstruction from freehand 2D ultrasound scans without tracking information. The proposed method efficiently exploits content correspondence between ultrasound frames to estimate spatial movement and contrastive features and achieves superior performance compared with other methods, with a drift rate of 9.64% and a prostate Dice of 0.89. The promising results demonstrate the capability of deep neural networks for universal ultrasound volume reconstruction.
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING
(2023)
Article
Engineering, Biomedical
Zhiguo Zhou, Liyuan Chen, Michael Dohopolski, David Sher, Jing Wang
Summary: In this study, a new automated and reliable multi-objective learning model (ARMO) is proposed for identifying lymph node metastasis. The experimental results demonstrate that ARMO can achieve accurate and reliable prediction performance.
PHYSICS IN MEDICINE AND BIOLOGY
(2023)
Article
Oncology
David J. Sher, Dominic H. Moon, Dat Vo, Jing Wang, Liyuan Chen, Michael Dohopolski, Randall Hughes, Baran D. Sumer, Chul Ahn, Vladimir Avkshtol
Summary: This prospective phase II study indicates that elective neck irradiation is no longer mandatory for head and neck squamous cell carcinoma treatment, and radiation therapy can effectively reduce tumor recurrence and improve patients' quality of life.
CLINICAL CANCER RESEARCH
(2023)