Article
Computer Science, Artificial Intelligence
Bo Zhou, Zachary Augenfeld, Julius Chapiro, S. Kevin Zhou, Chi Liu, James S. Duncan
Summary: Multimodal image registration plays a crucial role in diagnostic medical imaging and image-guided interventions, potentially improving therapeutic outcomes. However, challenges such as suboptimal image quality in intra-procedural CBCT and lack of standard intensity-based registration methods call for new solutions, including leveraging deep learning anatomy extractors and robust point matching in multimodal registration frameworks.
MEDICAL IMAGE ANALYSIS
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Lukas Birklein, Stefan Niebler, Elmar Schoemer, Robert Brylka, Ulrich Schwanecke, Ralf Schulze
Summary: We propose a motion estimation and motion correction approach for oral and maxillofacial cone beam CT scans based solely on 2D projection images. This algorithm improves the visualization quality of motion impaired scans, eliminating the need for re-scans and reducing radiation dosage for patients.
Article
Radiology, Nuclear Medicine & Medical Imaging
Laurens Hermie, Elisabeth Dhondt, Peter Vanlangenhove, Jan De Waele, Helena Degroote, Luc Defreyne
Summary: The study evaluated the clinical effect and safety of CBCT-guided empirical embolization for acute LGIB patients with positive CTA but negative DSA. Results showed that empirical CBCT-guided embolization significantly reduced rebleeding rates in patients with negative DSA, improving clinical success.
EUROPEAN RADIOLOGY
(2021)
Review
Radiology, Nuclear Medicine & Medical Imaging
David A. Jaffray
Summary: The development of cone-beam CT guided radiotherapy has greatly impacted radiation oncology in the past 20 years. It has changed daily clinical practice, encouraged multidisciplinary collaboration, and facilitated new treatment methods. Furthermore, the integration of quantitative CT in robotic platforms can help address the global shortage of high-quality radiotherapy services by utilizing machine learning approaches and automating expertise.
Article
Radiology, Nuclear Medicine & Medical Imaging
D. Amiras, T. J. Hurkxkens, D. Figueroa, P. J. Pratt, B. Pitrola, C. Watura, S. Rostampour, G. J. Shimshon, M. Hamady
Summary: The study demonstrates the feasibility of simulating a CT-guided procedure using augmented reality and suggests that it could be an effective training tool. Most users found the application to be accurate and realistic enough for training purposes, and felt more confident in their CT biopsy skills after the training session.
EUROPEAN RADIOLOGY
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Wonju Hong, Soon Ho Yoon, Jin Mo Goo, Chang Min Park
Summary: This study retrospectively reviewed 336 cone-beam CT-guided percutaneous transthoracic needle biopsy procedures in 326 patients and found that the target lesion size was the only significant predictor of diagnostic failure, while complications occurred at a reasonable rate.
KOREAN JOURNAL OF RADIOLOGY
(2021)
Article
Medicine, General & Internal
Mohammed G. Sghaireen, Kiran Kumar Ganji, Kumar Chandan Srivastava, Mohammad Khursheed Alam, Shadi Nashwan, Fayeq Hasan Migdadi, Ahmad Al-Qerem, Yousef Khader
Summary: This clinical study evaluated the correlation between Vitamin D (Vit D), cholesterol levels, and T- and Z-scores of dual-energy X-ray absorptiometry (DXA) scans with cone beam computed tomography values in the maxillary and mandibular jaws. A total of 187 patients aged between 45 and 65 were included in the study, divided into control (non-osteoporosis) and case (osteoporosis) groups. The results showed a significant inverse relationship between Vit D and cholesterol in the case group, and a positive relationship between Vit D and cone beam computed tomography values in all regions of the jaws except the mandibular posterior region. Pearson correlation analysis revealed an insignificant negative association between Vit D, cholesterol levels, and cone beam computed tomography values in all regions of the jaws, while Z-values were highly correlated with the cone beam computed tomography values in all regions of the jaws. Vit D, cholesterol levels, and Z-values were found to be related to the cone beam computed tomography values of the jaws in women and men aged 45-65.
JOURNAL OF CLINICAL MEDICINE
(2023)
Article
Engineering, Biomedical
Chulong Zhang, Wenfeng He, Lin Liu, Jingjing Dai, Isah Salim Ahmad, Yaoqin Xie, Xiaokun Liang
Summary: In this paper, a medical image registration method based on volumetric feature points integration and bio-structure guidance is proposed. By using surface-registered point pairs and voxel feature point pairs to guide the training process, higher registration accuracy is achieved. The method has been validated on paired CT-CBCT datasets and shows a 6% improvement in precision compared to other deep learning methods, reaching a state-of-the-art status.
PHYSICS IN MEDICINE AND BIOLOGY
(2023)
Article
Dentistry, Oral Surgery & Medicine
Chun Yi, Sha Li, Aonan Wen, Yong Wang, Yijiao Zhao, Yu Zhang
Summary: This study compared the ability of conventional CBCT and digital registration to evaluate the accuracy of implant positioning in guided surgery. The results showed good agreement between the two methods.
Article
Oncology
Hongfei Sun, Rongbo Fan, Chunying Li, Zhengda Lu, Kai Xie, Xinye Ni, Jianhua Yang
Summary: An improved 3D CycleGAN method was used to generate pseudo-CT images, with the imaging quality improved by adding a gradient loss function. The model was validated through fivefold cross-validation, demonstrating the accuracy of the pseudo-CT images in terms of electronic density and anatomical structure.
FRONTIERS IN ONCOLOGY
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Haytham Derbel, Mahdi Krichen, Julia Chalaye, Laetitia Saccenti, William van der Sterren, Anne-Hilde Muris, Lionel Lerman, Athena Galletto, Youssef Zaarour, Alain Luciani, Hicham Kobeiter, Vania Tacher
Summary: This study evaluated the anatomical and volumetric predictability of CBCT-based VPP software for SPECT/CT imaging results in HCC patients undergoing TARE. The results showed that the CBCT-based VPP software accurately and reliably predicted the anatomical and volumetric results of Tc-99m-MAA SPECT/CT in HCC patients during TARE.
EUROPEAN RADIOLOGY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Yue Ma, Aidi Liu, Avice M. O'Connell, Yueqiang Zhu, Haijie Li, Peng Han, Lu Yin, Hong Lu, Zhaoxiang Ye
Summary: CE-CBBCT features are associated with IHC receptors and molecular subtypes of breast cancer, which can help predict IHC receptor status and differentiate molecular subtypes. Specific features such as lesion number, tumor size, and internal enhancement pattern are related to different subtypes.
EUROPEAN RADIOLOGY
(2021)
Article
Engineering, Electrical & Electronic
Berk Iskender, Yoram Bresler
Summary: Scatter caused by the interaction between photons and the imaged object is a fundamental problem in X-ray CT, as it leads to various artifacts during reconstruction. Hardware-based methods require modification in the hardware or increased scan time or dose, while software-based methods have gained great interest. In this work, two novel physics-inspired deep learning methods, PhILSCAT and OV-PhILSCAT, are proposed to estimate and correct for scatter in acquired projection measurements.
IEEE TRANSACTIONS ON COMPUTATIONAL IMAGING
(2022)
Review
Radiology, Nuclear Medicine & Medical Imaging
Yueqiang Zhu, Avice M. O'Connell, Yue Ma, Aidi Liu, Haijie Li, Yuwei Zhang, Xiaohua Zhang, Zhaoxiang Ye
Summary: Dedicated breast CT is increasingly used for breast imaging, providing images with no compression, removal of tissue overlap, rapid acquisition, and simultaneous assessment of microcalcifications and contrast enhancement. The current focus is on optimizing practice for lesion detection and characterization, therapy monitoring, density assessment, intervention, and implant evaluation, as well as exploring the potential for future breast CT screening.
EUROPEAN RADIOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Yueqiang Zhu, Avice M. O'Connell, Yue Ma, Aidi Liu, Haijie Li, Yuwei Zhang, Xiaohua Zhang, Zhaoxiang Ye
Summary: Dedicated breast CT is an emerging 3D isotropic imaging technology that overcomes the limitations of traditional breast imaging methods. Despite the technical challenges such as scan protocol, radiation dose, breast coverage, patient comfort, and image artifact, proposed methods have been discussed to address these issues and improve the technology further.
EUROPEAN RADIOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
R. Han, C. K. Jones, J. Lee, P. Wu, P. Vagdargi, A. Uneri, P. A. Helm, M. Luciano, W. S. Anderson, J. H. Siewerdsen
Summary: This study introduces a deep learning-based dual-channel registration framework to address brain tissue deformation in minimally invasive, intracranial neurosurgery. Through end-to-end training of synthesis and registration subnetworks, the proposed method achieves superior geometric accuracy and runtime compared to state-of-the-art baseline methods and other variations.
MEDICAL IMAGE ANALYSIS
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Kai Mei, Michael Geagan, Leonid Roshkovan, Harold Litt, Grace J. Gang, Nadav Shapira, J. Webster Stayman, Peter B. Noel
Summary: Purpose phantoms play a vital role in assessing and verifying performance in CT research and clinical practice, with patient-based realistic lung phantoms being crucial for developing and evaluating novel CT hardware and software. This study introduces PixelPrint as a solution for creating patient-based lung phantoms with accurate attenuation profiles and textures. Results show a high level of linear correlation between filament line widths and Hounsfield units for calibration phantoms, with patient-based phantoms closely resembling original CT images in texture and contrast levels. The study demonstrates the feasibility of 3D-printed patient-based lung phantoms with accurate organ geometry, image texture, and attenuation profiles.
Article
Radiology, Nuclear Medicine & Medical Imaging
Niral Milan Sheth, Ali Uneri, Patrick A. Helm, Wojciech Zbijewski, Jeffrey H. Siewerdsen
Summary: This study investigates the imaging performance of a new indium gallium zinc oxide (IGZO) TFT-based detector in 2D fluoroscopy and 3D cone-beam CT (CBCT). The results show that the IGZO-based detector exhibits improvements in electronic noise, image lag, and NEQ compared to a conventional detector based on a-Si:H TFTs.
Article
Engineering, Biomedical
R. Han, C. K. Jones, J. Lee, X. Zhang, P. Wu, P. Vagdargi, A. Uneri, P. A. Helm, M. Luciano, W. S. Anderson, J. H. Siewerdsen
Summary: A deep learning-based deformable registration method was proposed to address deep brain deformations in minimally invasive neurosurgery, achieving superior performance in deep brain structure registration with short runtime. The method was trained on a simulated dataset and refined on real clinical images to support translation into high-precision neurosurgical studies.
PHYSICS IN MEDICINE AND BIOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Xiaoxuan Zhang, Ali Uneri, Yixuan Huang, Craig K. Jones, Timothy F. Witham, Patrick A. Helm, Jeffrey H. Siewerdsen
Summary: This study reports a deformable 3D-2D registration method that accurately evaluates spinal device placement and alignment in the operating room. The method utilizes multi-scale masking and spline fitting techniques to achieve accurate patient and device registration, demonstrating the accuracy and effectiveness of the approach.
Article
Computer Science, Interdisciplinary Applications
Matthew Tivnan, Wenying Wang, Grace Gang, J. Webster Stayman
Summary: Spectral CT has the potential for high-sensitivity quantitative imaging and material decomposition. In this study, a new device called a spatial-spectral filter (SSF) is introduced to modulate the spectral shape of the x-ray beam. A one-step direct model-based material decomposition (MBMD) is used to reconstruct basis material density images from the sparse SSF CT data. A Fisher-information-based separability index is defined to evaluate different SSF designs. Simulation-based design optimization is conducted to find optimized combinations of filter materials, thicknesses, widths, and source settings. The results demonstrate the ability to reconstruct basis material density images and show the benefits of the optimized designs for material discrimination tasks.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2022)
Article
Surgery
I. Butz, M. Fernandez, A. Uneri, N. Theodore, W. S. Anderson, J. H. Siewerdsen
Summary: A system for performance assessment and quality assurance of surgical trackers is reported based on principles of geometric accuracy and statistical process control. The system uses a simple test phantom and open-source software to measure accuracy and stability of the trackers.
COMPUTER ASSISTED SURGERY
(2023)
Article
Engineering, Biomedical
Prasad Vagdargi, Ali Uneri, Craig K. Jones, Pengwei Wu, Runze Han, Mark G. Luciano, William S. Anderson, Patrick A. Helm, Gregory D. Hager, Jeffrey H. Siewerdsen
Summary: A robot-assisted ventriculoscopy (RAV) system was implemented to reconstruct, register, and augment the neuroendoscopic scene with intraoperative imaging. It can provide guidance and visualization even in the presence of tissue deformation.
IEEE TRANSACTIONS ON MEDICAL ROBOTICS AND BIONICS
(2022)
Proceedings Paper
Engineering, Biomedical
Y. Huang, C. K. Jones, X. Zhang, A. Johnston, N. Aygun, T. Witham, P. A. Helm, J. H. Siewerdsen, A. Uneri
Summary: This study introduces a custom, multi-perspective, region-based convolutional neural network (R-CNN) for labeling vertebrae in Long-Film images and evaluates two methods for incorporating contextual information. The results show that the multi-perspective R-CNN with LSTM module significantly improves labeling accuracy compared to the base model, while sequence sorting has limited effectiveness. The proposed LSTM module achieves better performance in terms of labeling accuracy and flexibility through end-to-end training.
MEDICAL IMAGING 2022: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING
(2022)
Proceedings Paper
Engineering, Biomedical
P. Vagdargi, A. Uneri, C. K. Jones, P. Wu, R. Han, M. Luciano, W. S. Anderson, P. A. Helm, G. D. Hager, J. H. Siewerdsen
Summary: This study developed a neuroendoscopic navigation system based on SLAM 3D point-cloud reconstruction, providing a promising platform for the development of robot-assisted endoscopic neurosurgery. The effectiveness of feature detectors and descriptors was evaluated in anthropomorphic phantom studies, and optimal feature detection parameters were identified. Future work aims to improve the SLAM framework, assess the geometric accuracy of reconstruction, and translate the methods to clinical studies.
MEDICAL IMAGING 2022: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING
(2022)
Proceedings Paper
Engineering, Biomedical
Rohan C. Vijayan, Niral Sheth, Lina Mekki, Alexander Lu, Ali Uneri, Alejandro Sisniega, Jessica Maggaragia, Gerhard Kleinszig, Sebastian Vogt, Jeffrey Thiboutot, Hans Lee, Lonny Yarmus, Jeffrey H. Siewerdsen
Summary: This study presents a method to address deformable motion issues in fluoroscopy-guided pulmonary interventions by using locally rigid/globally deformable 3D-2D registration. The results showed that soft-tissue thresholding and contrast enhancement can improve target registration accuracy, and the proposed method has the potential to enhance guidance accuracy in pulmonary interventions.
MEDICAL IMAGING 2022: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING
(2022)
Proceedings Paper
Engineering, Biomedical
R. Han, C. K. Jones, P. Wu, P. Vagdargi, X. Zhang, A. Uneri, J. Lee, M. Luciano, W. S. Anderson, P. Helm, J. H. Siewerdsen
Summary: This study presents a deep learning-based method, called JSR, for deformable MR-to-CBCT registration in neuro-endoscopic surgery. The JSR algorithm achieves accurate and near real-time registration results through a joint synthesis and registration network.
MEDICAL IMAGING 2022: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING
(2022)
Proceedings Paper
Engineering, Biomedical
Justin D. Opfermann, Benjamin D. Killeen, Christopher Bailey, Majid Khan, Ali Uneri, Kensei Suzuki, Mehran Armand, Ferdinand Hui, Axel Krieger, Mathias Unberath
Summary: This study introduces a novel cannula mounted vertebral augmentation robot, which demonstrates feasibility and effectiveness in a simulated X-ray environment. The design and evaluation of the robot show potential for successful autonomous vertebral augmentation procedures.
2021 IEEE 21ST INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOENGINEERING (IEEE BIBE 2021)
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Michael D. Ketcha, Michael Marrama, Andre Souza, Ali Uneri, Pengwei Wu, Xiaoxuan Zhang, Patrick A. Helm, Jeffrey H. Siewerdsen
Summary: This study presents a dual convolutional neural network approach to address metal artifacts in intraoperative CBCT imaging. Trained on cadaver scans with simulated metal hardware, the method showed superior performance on images, outperforming a single image domain operation method.
JOURNAL OF MEDICAL IMAGING
(2021)
Proceedings Paper
Engineering, Biomedical
X. Zhang, A. Uneri, P. Wu, M. D. Ketcha, S. A. Doerr, C. K. Jones, P. A. Helm, J. H. Siewerdsen
Summary: The EV imaging technique allows for long length visualization of the spine and long surgical constructs, providing improved visualization and accurate measurement for intraoperative assessment of spinal deformity correction.
MEDICAL IMAGING 2020: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING
(2021)