3D Computed Tomography Images Research Articles

Evaluation of the accuracy and interobserver agreement of the reverse shoulder angle in the preoperative planning of reverse total shoulder arthroplasties.

Differences in implant positioning between anatomical and reverse shoulder arthroplasties have raised concerns about the adequacy of assessing the global glenoid inclination (GGI) using the method described by Maurer to define the position of the metallic base in reverse shoulder arthroplasty. The reverse shoulder angle (RSA) has been proposed to measure the inclination of the lower half of the glenoid. This study aims to evaluate the interobserver agreement of manual measurements of the RSA using two-dimensional (2D) computed tomography (CT) images and its relationship with the automated measurement of the GGI. This cross-sectional study evaluated 2D CT images of 38 CT scans of patients with degenerative shoulder diseases. Manual measurements of the RSA were conducted by five independent shoulder surgeons. GGI measured by automated software was determined. The interclass correlation coefficient was 0.72 for RSA. Mean RSA was 25.7° ± 7.1°, significantly higher than automated GGI measurements (11.2 ± 9.0; p < .0001). On average, RSA was 14.6 ± 6.3 greater than GGI, irrespective, diagnosis, Favard's glenoid subtype, and version angle. Our study demonstrated that the RSA angle can be reproducibly used to assess glenoid inclination on 2D CT images presenting high interobserver agreement. RSA differs significantly from the GGI, indicating that measuring glenoid inclination for reverse or anatomical arthroplasty requires distinct methodologies that account for the inherent differences in these angles.

Determination of the Surgical Landmarks for the Anterior and Middle Cranial Fossa in Dry Skulls With the Photography System, Cadavers and 3-Dimensional Computed Tomography

The present paper was designed to analyze the dimensions of such important bony structures and surgical landmarks, which are used by many clinicians in many surgical interventions, in dry skull, cadaver, and healthy subjects on computed tomography (CT) images, and to determine whether there is a significant difference between these methods, and to obtain reference values from 3 different methods. Eight cadavers and 16 dry skulls and 100 three-dimensional (3D) CT images were studied. Necessary permissions for the study were obtained from Ethics Comittee. The 16 parameters were measured with an electronic digital caliper accurate 0.01 mm (LCD Digital Vernier Dial Microcaliper (INCA, DCLA-0605, 0.6–150 mm, USA). Also, the images obtained were transferred to the 3D Slicer (version 5.6.2) software program. Eight cadavers and 16 dry skulls of Turkish adults were unknown age and sex, whereas the mean age of females and males on CT images were 31.63±11.23 and 33.70±13.34 years, respectively. All values of the surgical landmarks for the anterior and middle cranial fossa obtained from cadavers, dry skulls, and 3D CT subjects (except length of lesser wing, anterior clinoid lengths for 2 sides, and width for right side) were statistically significant between 3 groups (P&lt;0.05). This paper was conducted for the morphometric analysis of the specific regions of the anterior cranial fossa (ACF), and middle cranial fossa (MCF), which are used in neurosurgical procedures This detailed anatomic and radiologic reference values will be an extremely important source in the planning of both clinical and surgical approaches for neurosurgeon, anatomist, and radiologists.

Advancements in 3D Transoesophageal Echocardiography (TOE) and Computed Tomography (CT) for Stroke Prevention in Left Atrial Appendage Occlusion Interventions.

Left atrial appendage occlusion (LAAO) has emerged as a highly effective alternative to oral anticoagulation for stroke prevention in patients with non-valvular atrial fibrillation. Precise pre-procedural planning and meticulous post-procedural follow-up are essential for achieving successful LAAO outcomes. This review explores the latest advancements in three-dimensional (3D) transoesophageal echocardiography (TOE) and computed tomography (CT) imaging modalities, which have considerably improved the planning, intra-procedural guidance, and follow-up processes for LAAO interventions. Innovations in 3D TOE and CT imaging have transformed the approach to LAAO by providing a more detailed and accurate assessment of the left atrial appendage, enabling clinicians to acquire comprehensive anatomical and morphological information, crucial for optimising device selection and positioning, thus reducing the risk of complications and enhancing the overall safety and efficacy of the procedure. Post-procedurally, CT and TOE imaging are invaluable in the monitoring of patients, ensuring that the device is correctly positioned and functioning as intended. Early detection of any complications (e.g., device-related thrombus and peri-device leaks) can help to risk-stratify patient at increased risk of stroke and initiate timely interventions, thereby improving long-term outcomes for patients.

CSSANet: A channel shuffle slice-aware network for pulmonary nodule detection

Lung cancer stands as the leading cause of cancer related mortality worldwide. Precise and automated identification of lung nodules through 3D Computed Tomography (CT) scans is an essential part of screening for lung cancer effectively. Due to the small size of pulmonary nodules and the close correlation between neighboring slices of 3D CT images, most of the existing methods only consider the characteristics of a single slice, thus easily lead to insufficient detection accuracy of pulmonary nodules. To solve this problem, this paper proposes a Channel Shuffle Slice-Aware Network (CSSANet), which aims to fully exploit the spatial correlation between slices and effectively utilize the intra-slice features and inter-slice contextual information to achieve accurate detection of lung nodules. Specifically, we design a Group Shuffle Attention module (GSA module) to fuse the inter-slice feature in order to enhance the discrimination and extraction of corresponding shape information of distinct nodules in the same group of slices. Experiments and ablation study on a publicly available LUNA16 dataset demonstrate that the proposed method can enhance the detection sensitivity effectively. The Competition Performance Metric (CPM) score of 89.8 % is superior over other representative detection models.

Topological analysis of the three-dimensional radiodensity distribution of fish otoliths: Point sampling effects on dimensionality reduction

Otoliths are calcified structures found in the inner ears of teleost fish, pivotal in marine biology for studies on metabolism, age, growth, and the identification of fish stocks, potentially leading to sustainable management practices. An important feature of this structure is its density, as it corresponds to modifications in the crystalline form of calcium carbonate during the fish's lifetime, resulting in variations in its final shape. The internal and external 3D radiodensity of otoliths from different species was obtained utilizing micro-computed tomography, however, an appropriate methodology for describing and conducting comparative studies on these data appears to be absent in the current body of literature. Therefore, we study otolith density variations from 3D computed tomography images, employing the Ball Mapper technique of Topological Data Analysis. We focus on reducing the computational cost of this analysis by applying probabilistic sampling and assessing its effects on the density variations provided by the Ball Mapper graph. To determine the sample size, we used the topology to establish what we term "Topological Sample Validation", which provided the minimum resolution with the same density information as raw data. Sample representativeness was validated through non-parametric statistical tests on the density variable. Based on the network's structural characteristics, network properties allowed for evaluating similarity between graphs. Besides the small sample size, remarkable correlations were obtained between age and network variables. Additionally, the Ball Mapper technique proved effective as a preprocessing algorithm for tomographic images, enabling the segmentation of undesired features in the object of interest.

Motion and anatomy dual aware lung ventilation imaging by integrating Jacobian map and average CT image using dual path fusion network.

Deep learning-based computed tomography (CT) ventilation imaging (CTVI) is a promising technique for guiding functional lung avoidance radiotherapy (FLART). However, conventional approaches, which rely on anatomical CT data, may overlook important ventilation features due to the lack of motion data integration. This study aims to develop a novel dual-aware CTVI method that integrates both anatomical information from CT images and motional information from Jacobian maps to generate more accurate ventilation images for FLART. A dataset of 66 patients with four-dimensional CT (4DCT) images and reference ventilation images (RefVI) was utilized to develop the dual-path fusion network (DPFN) for synthesizing ventilation images (CTVIDual). The DPFN model was specifically designed to integrate motion data from 4DCT-generated Jacobian maps with anatomical data from average 4DCT images. The DPFN utilized two specialized feature extraction pathways, along with encoders and decoders, designed to handle both 3D average CT images and Jacobian map data. This dual-processing approach enabled the comprehensive extraction of lung ventilation-related features. The performance of DPFN was assessed by comparing CTVIDual to RefVI using various metrics, including Spearman's correlation coefficients (R), Dice similarity coefficients of high-functional region (DSCh), and low-functional region (DSCl). Additionally, CTVIDual was benchmarked against other CTVI methods, including a dual-phase CT-based deep learning method (CTVIDLCT), a radiomics-based method (CTVIFM), a super voxel-based method (CTVISVD), a Unet-based method (CTVIUnet), and two deformable registration-based methods (CTVIJac and CTVIHU). In the test group, the mean R between CTVIDual and RefVI was 0.70, significantly outperforming CTVIDLCT (0.68), CTVIFM (0.58), CTVISVD (0.62), and CTVIUnet (0.66), with p<0.05. Furthermore, the DSCh and DSCl values of CTVIDual were 0.64 and 0.80, respectively, outperforming CTVISVD (0.63; 0.73) and CTVIUnet (0.62; 0.77). The performance of CTVIDual was also significantly better than that of CTVIJac and CTVIHU. A novel dual-aware CTVI model that integrates anatomical and motion information was developed to synthesize lung ventilation images. It was shown that the accuracy of lung ventilation estimation could be significantly enhanced by incorporating motional information, particularly in patients with tumor-induced blockages. This approach has the potential to improve the accuracy of CTVI, enabling more effective FLART.

Fast reconstruction 3D computed tomography image of stacked cell under faster scanning by dual-branch cross-fusion flat bottom network

Abstract Cone beam computed tomography (CBCT) fast scanning and reconstruction is a key step to achieve rapid detection of internal defects in batteries. In this work, we have achieved a faster CT scanning just in 5 seconds by reducing the X-ray exposure time in sparse view CT. However, the CT data is extremely incomplete by faster scanning; the existing reconstruction methods are difficult to reconstruct a high quality three-dimensional (3D) CT image of stacked cells. To address this issue, we propose a 3D CT image reconstruction network, which can reconstruct higher quality CT images from low quality 3D volume data. The input data of the reconstruction network is not 2D projection data, but 3D volume data. In this network, a high and low resolution dual-branch cross-fusion flat bottom structure is designed. The high resolution flat bottom branch aims to preserve detailed information, while the low resolution flat bottom branch focuses on capturing more semantic information. Cross-fusion between these branches mitigates the loss of semantic details. Additionally, the auxiliary loss function, the main loss function, and the 3D attention module are designed to enhance semantic accuracy and the learning performance of the network. The 3D training data is collected under a fast scanning strategy spanning 5-60 seconds. During the training phase, we use clipping block technology to cut the 3D volume data, enabling direct training on the 3D volume data. Our experimental results demonstrate that our 3D reconstruction network outperforms mainstream algorithms under this faster scanning strategy, which is able to reconstruct higher quality 3D CT images just in 15 seconds. Ablation experiments confirm the positive impact of the dual-branch cross-fusion flat bottom structure, attention module, and loss functions on improving the quality of 3D CT images.

Accelerated muscle mass estimation from CT images through transfer learning

BackgroundThe cost of labeling to collect training data sets using deep learning is especially high in medical applications compared to other fields. Furthermore, due to variances in images depending on the computed tomography (CT) devices, a deep learning based segmentation model trained with a certain device often does not work with images from a different device.MethodsIn this study, we propose an efficient learning strategy for deep learning models in medical image segmentation. We aim to overcome the difficulties of segmentation in CT images by training a VNet segmentation model which enables rapid labeling of organs in CT images with the model obtained by transfer learning using a small number of manually labeled images, called SEED images. We established a process for generating SEED images and conducting transfer learning a model. We evaluate the performance of various segmentation models such as vanilla UNet, UNETR, Swin-UNETR and VNet. Furthermore, assuming a scenario that a model is repeatedly trained with CT images collected from multiple devices, in which is catastrophic forgetting often occurs, we examine if the performance of our model degrades.ResultsWe show that transfer learning can train a model that does a good job of segmenting muscles with a small number of images. In addition, it was confirmed that VNet shows better performance when comparing the performance of existing semi-automated segmentation tools and other deep learning networks to muscle and liver segmentation tasks. Additionally, we confirmed that VNet is the most robust model to deal with catastrophic forgetting problems.ConclusionIn the 2D CT image segmentation task, we confirmed that the CNN-based network shows better performance than the existing semi-automatic segmentation tool or latest transformer-based networks.

Sex estimation from skull measurements of a contemporary Japanese population using three-dimensional computed tomography images.

In this study, we assessed the sexual dimorphism of the contemporary Japanese skull and established sex discriminant function equations based on cranial measurements using three-dimensional (3D) computed tomography (CT) images. The CT images of 263 corpses (142 males, 121 females) that underwent postmortem CT scanning and subsequent forensic autopsy were evaluated. Twenty-one cranial measurements were obtained from 3D CT reconstructed images, which extracted only bone data. We performed descriptive statistics and discriminant function analyses for the measurements. Nineteen measurements were significantly larger in males, suggesting sexual dimorphism of the Japanese skulls. Univariate discriminant function analyses using these measurements showed a sex classification accuracy of 57.8-88.2%, and bizygomatic breadth provided the highest correct prediction rate. Multivariate discriminant function analyses offered the most accurate model using seven variables with an estimation rate of 93.9%. Our results suggest that cranial measurements based on 3D CT images may help in the sex estimation of unidentified bodies in a contemporary Japanese population.

Accuracy of Orthodontic Anchor Screw Placement Using a 3D-Printed Surgical Guide.

Background Although radiographs and computed tomography (CT) images are reviewed before temporary anchorage device (TAD) implantation, implantation of TADs exactly as planned is difficult. This study aimed to evaluate the accuracy of TAD implantation using an original surgical guide fabricated using cone-beam CT data and computer-aided design software. Methodology The studyparticipants included six experienced orthodontists who had implanted ≥20 TADs, and six inexperienced orthodontists who had never implanted a TAD. Maxillary dental typodont models with radiopaque tooth crowns and roots were used. A total of four TADs were implanted on the buccal sides: between the second bicuspid and first molars and between the first and second molars bilaterally. The accuracy of TAD implantation was examined in two groups: in 12 dental typodont models, TAD implantation was performed using a surgical guide (guide group), and in 12 dental typodont models, TAD implantation was performed without a surgical guide (freehand group). All dental typodont models implanted a total of 96 TADs.The TAD position was evaluated using the CT coordinate system and 3D image measurement software. Using the long axis of the TAD as a reference, the distance between the coronal and apical ends of the implanted TAD and those of the planned TAD, i.e., the ideal implantation position, was measured in both groups along the x, y, and z axes. The medians of the values were compared between the groups. Additionally, the presence of root contact was compared between the experienced and inexperienced orthodontists. Results On the x-axis, the linear deviations (median) of the coronal and apical ends of the TAD in the freehand group were 1.06 mm and 1.36 mm, respectively. In contrast, in the guide group, the deviations were 0.65 mm and 0.90 mm, respectively, and the difference was statistically significant (p = 0.002 and p = 0.005, respectively). On the y-axis, the deviations in the freehand group were 1.13 mm and 1.08 mm, respectively. In contrast, the deviations in the guide group were 0.71 mm and 0.79 mm, respectively, and only the coronal deviations were significantly different between the groups (p = 0.006). On the z-axis, the deviations in the freehand group were 1.44 mm and 1.86 mm, respectively. In contrast, the deviations in the guide group were 0.75 mm and 1.16 mm, respectively, and the difference was statistically significant (p = 0.006 and p = 0.002, respectively). Conclusions The use of a surgical guide allowed for more accurate TAD implantation. Additionally, TAD implantation using a guide prevented root damage.

Three-Dimensional Computed Tomography Scanning Study of the Superior Labial Artery in Chinese Individuals for Assessing Filler Injection Safety.

Lip filler injection is one of the most common minimally invasive cosmetic procedures involving the face; however, vascular complications are not uncommon. The aim of this study was to investigate the anatomy of the superior labial artery (SLA) and provide precise topographic information for dermal filler injection into the lips. Computed tomography (CT) scans of 52 cadaveric heads injected with lead oxide were obtained. We then used Mimics software to construct 3D images of the SLA described by a coordinate system based on the bilateral external auditory canal and the left orbit. This study aimed to classify the SLA in the Han Chinese population, measure its diameter at specific points, and determine the thickness of the lip at those points. Ultimately, we utilized a thermal imaging technique to illustrate the course and depth of the SLA within the lip. The objective of this study was to provide safe guidance for clinical injections. In this study, the SLA was successfully identified in all cadavers. The mean overall diameter of the superior labial arteries was 1.36±0.28 mm. The superior labial artery showed a general course from deep to shallow with an average depth of 5.68±1.68 mm from the oral commissure to the midline. There are anatomical differences in the superior labial arteries among Chinese people. Furthermore, 3D CT images can digitally elucidate the exact positions of the superior labial artery via a coordinate system, improving the safety of upper lip filler injections in clinical settings. This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Comparison of the quadrant method measuring four points and bernard method in femoral tunnel position evaluation on 3-dimensional reconstructed computed tomography after anatomical single-bundle anterior cruciate ligament reconstruction

PurposeThis prospective study aimed to compare the postoperative evaluation of the quadrant method measuring four points and Bernard method in femoral tunnel position evaluation on 3-Dimensional (3D) reconstructed computed tomography (CT) following the arthroscopic single-bundle anterior cruciate ligament (ACL) reconstruction.MethodsThirty-eight patients with ACL tears that were reconstructed using single-bundle ACL reconstruction between May 2021 and March 2023 were included in this study. Postoperative 3D CT images were obtained after the operation. The femoral tunnel position was measured by use of the quadrant method measuring four points and Bernard method.ResultsAverage mean position of the femoral tunnel insertion center on the 3D CT image was at 26.16 ± 6.27% in the x-coordinate and at 24.36 ± 5.52% in the y-coordinate according to the Bernard method. Meanwhile, the position of the femoral insertion of the ACL measured by the quadrant method measuring four points was 24.2% ± 6.86% in the x-coordinate and 21.16% ± 5.14% in the y-coordinate.ConclusionsBoth the quadrant method measuring four points and Bernard method were effective in femoral tunnel position evaluation on 3D reconstructed CT. Application of the quadrant method measuring four points on 3D CT showed the advantage that measurement can be taken regardless of the shape of the bone tunnel.

HLFSRNN-MIL: A Hybrid Multi-Instance Learning Model for 3D CT Image Classification

At present, many diseases are diagnosed by computer tomography (CT) image technology, which affects the health of the lives of millions of people. In the process of disease confrontation, it is very important for patients to detect diseases in the early stage by deep learning of 3D CT images. The paper offers a hybrid multi-instance learning model (HLFSRNN-MIL), which hybridizes high-low frequency feature fusion (HLFFF) with sequential recurrent neural network (SRNN) for CT image classification tasks. Firstly, the hybrid model uses Resnet-50 as the deep feature. The main feature of the HLFSRNN-MIL lies in its ability to make full use of the advantages of the HLFFF and SRNN methods to make up for their own weakness; i.e., the HLFFF can extract more targeted feature information to avoid the problem of excessive gradient fluctuation during training, and the SRNN is used to process the time-related sequences before classification. The experimental study of the HLFSRNN-MIL model is on two public CT datasets, namely, the Cancer Imaging Archive (TCIA) dataset on lung cancer and the China Consortium of Chest CT Image Investigation (CC-CCII) dataset on pneumonia. The experimental results show that the model exhibits better performance and accuracy. On the TCIA dataset, HLFSRNN-MIL with Residual Network (ResNet) as the feature extractor achieves an accuracy (ACC) of 0.992 and an area under curve (AUC) of 0.997. On the CC-CCII dataset, HLFSRNN-MIL achieves an ACC of 0.994 and an AUC of 0.997. Finally, compared with the existing methods, HLFSRNN-MIL has obvious advantages in all aspects. These experimental results demonstrate that HLFSRNN-MIL can effectively solve the disease problem in the field of 3D CT images.

Enhancing trabecular CT scans based on deep learning with multi-strategy fusion

Trabecular bone analysis plays a crucial role in understanding bone health and disease, with applications like osteoporosis diagnosis. This paper presents a comprehensive study on 3D trabecular computed tomography (CT) image restoration, addressing significant challenges in this domain. The research introduces a backbone model, Cascade-SwinUNETR, for single-view 3D CT image restoration. This model leverages deep layer aggregation with supervision and capabilities of Swin-Transformer to excel in feature extraction. Additionally, this study also brings DVSR3D, a dual-view restoration model, achieving good performance through deep feature fusion with attention mechanisms and Autoencoders. Furthermore, an Unsupervised Domain Adaptation (UDA) method is introduced, allowing models to adapt to input data distributions without additional labels, holding significant potential for real-world medical applications, and eliminating the need for invasive data collection procedures. The study also includes the curation of a new dual-view dataset for CT image restoration, addressing the scarcity of real human bone data in Micro-CT. Finally, the dual-view approach is validated through downstream medical bone microstructure measurements. Our contributions open several paths for trabecular bone analysis, promising improved clinical outcomes in bone health assessment and diagnosis.

No-Reference-Based and Noise Level Evaluations of Cinematic Rendering in Bone Computed Tomography.

Cinematic rendering (CR) is a new 3D post-processing technology widely used to produce bone computed tomography (CT) images. This study aimed to evaluate the performance quality of CR in bone CT images using blind quality and noise level evaluations. Bone CT images of the face, shoulder, lumbar spine, and wrist were acquired. Volume rendering (VR), which is widely used in the field of diagnostic medical imaging, was additionally set along with CR. A no-reference-based blind/referenceless image spatial quality evaluator (BRISQUE) and coefficient of variation (COV) were used to evaluate the overall quality of the acquired images. The average BRISQUE values derived from the four areas were 39.87 and 46.44 in CR and VR, respectively. The difference between the two values was approximately 1.16, and the difference between the resulting values increased, particularly in the bone CT image, where metal artifacts were observed. In addition, we confirmed that the COV value improved by 2.20 times on average when using CR compared to VR. This study proved that CR is useful in reconstructing bone CT 3D images and that various applications in the diagnostic medical field will be possible.

Measurement of the Acetabular Cup Orientation After Total Hip Arthroplasty Based on 3-Dimensional Reconstruction From a Single X-Ray Image Using Generative Adversarial Networks

BackgroundThe purpose of this study was to reconstruct 3-dimensional (3D) computed tomography (CT) images from single anteroposterior (AP) postoperative total hip arthroplasty (THA) X-ray images using a deep learning algorithm known as generative adversarial networks (GANs) and to validate the accuracy of cup angle measurement on GAN-generated CT. MethodsWe used 2 GAN-based models, CycleGAN and X2CT-GAN, to generate 3D CT images from X-ray images of 386 patients who underwent primary THAs using a cementless cup. The training dataset consisted of 522 CT images and 2,282 X-ray images. The image quality was validated using the peak signal-to-noise ratio and the structural similarity index measure. The cup anteversion and inclination measurements on the GAN-generated CT images were compared with the actual CT measurements. Statistical analyses of absolute measurement errors were performed using Mann–Whitney U tests and nonlinear regression analyses. ResultsThe study successfully achieved 3D reconstruction from single AP postoperative THA X-ray images using GANs, exhibiting excellent peak signal-to-noise ratio (37.40) and structural similarity index measure (0.74). The median absolute difference in radiographic anteversion was 3.45° and the median absolute difference in radiographic inclination was 3.25°, respectively. Absolute measurement errors tended to be larger in cases with cup malposition than in those with optimal cup orientation. ConclusionsThis study demonstrates the potential of GANs for 3D reconstruction from single AP postoperative THA X-ray images to evaluate cup orientation. Further investigation and refinement of this model are required to improve its performance.

LIT-Former: Linking In-Plane and Through-Plane Transformers for Simultaneous CT Image Denoising and Deblurring.

This paper studies 3D low-dose computed tomography (CT) imaging. Although various deep learning methods were developed in this context, typically they focus on 2D images and perform denoising due to low-dose and deblurring for super-resolution separately. Up to date, little work was done for simultaneous in-plane denoising and through-plane deblurring, which is important to obtain high-quality 3D CT images with lower radiation and faster imaging speed. For this task, a straightforward method is to directly train an end-to-end 3D network. However, it demands much more training data and expensive computational costs. Here, we propose to link in-plane and through-plane transformers for simultaneous in-plane denoising and through-plane deblurring, termed as LIT-Former, which can efficiently synergize in-plane and through-plane sub-tasks for 3D CT imaging and enjoy the advantages of both convolution and transformer networks. LIT-Former has two novel designs: efficient multi-head self-attention modules (eMSM) and efficient convolutional feed-forward networks (eCFN). First, eMSM integrates in-plane 2D self-attention and through-plane 1D self-attention to efficiently capture global interactions of 3D self-attention, the core unit of transformer networks. Second, eCFN integrates 2D convolution and 1D convolution to extract local information of 3D convolution in the same fashion. As a result, the proposed LIT-Former synergizes these two sub-tasks, significantly reducing the computational complexity as compared to 3D counterparts and enabling rapid convergence. Extensive experimental results on simulated and clinical datasets demonstrate superior performance over state-of-the-art models. The source code is made available at https://github.com/hao1635/LIT-Former.

Comparison of deep inspiration breath hold and free breathing intensity modulated proton therapy of locally advanced lung cancer

Background and purposeFor locally advanced non-small cell lung cancer (LA-NSCLC), intensity-modulated proton therapy (IMPT) can reduce organ at risk (OAR) doses compared to intensity-modulated radiotherapy (IMRT). Deep inspiration breath hold (DIBH) reduces OAR doses compared to free breathing (FB) in IMRT. In IMPT, differences in dose distributions and robustness between DIBH and FB are unclear. In this study, we compare DIBH to FB in IMPT, and IMPT to IMRT. Materials and methodsFortyone LA-NSCLC patients were prospectively included. 4D computed tomography images (4DCTs) and DIBH CTs were acquired for treatment planning and during weeks 1 and 3 of treatment. A new system for automated robust planning was developed and used to generate a FB and a DIBH IMPT plan for each patient. Plans were compared in terms of dose-volume parameters and normal tissue complication probabilities (NTCPs). Dose recalculations on repeat CTs were used to compare inter-fraction plan robustness. ResultsIn IMPT, DIBH reduced median lungs Dmean from 9.3 Gy(RBE) to 8.0 Gy(RBE) compared to FB, and radiation pneumonitis NTCP from 10.9 % to 9.4 % (p < 0.001). Inter-fraction plan robustness for DIBH and FB was similar. Median NTCPs for radiation pneumonitis and mortality were around 9 percentage points lower with IMPT than IMRT (p < 0.001). These differences were much larger than between FB and DIBH within each modality. ConclusionDIBH IMPT resulted in reduced lung dose and radiation pneumonitis NTCP compared to FB IMPT. Inter-fraction robustness was comparable. OAR doses were far lower in IMPT than IMRT.

Split Bregman quantum noise removal algorithm for 3D reconstruction of neutron computed tomography image

The low intensity of the neutron source for neutron computed tomography (CT) results in a long acquisition time for a single projection, which causes the neutron projection data to contain a large amount of quantum noise. Quantum noise will degrade the quality of neutron CT reconstruction images. Therefore, an efficient quantum noise removal algorithm must be used in CT reconstruction. In this paper, an efficient quantum noise removal algorithm for neutron CT 3D image reconstruction is proposed by analysing classical image processing algorithms and quantum image processing algorithms, which employs the maximum likelihood expectation maximization to reconstruct the image and split Bregman algorithm to solve for the total variation (MLEM-SBTV). Experimental results show that MLEM-SBTV performs well in removing quantum noise and reconstructing the detailed structure of images.

Evaluation of glenoid morphology and bony Bankart lesion in shoulders with traumatic anterior instability using zero echo time magnetic resonance imaging

BackgroundPreoperative computed tomography (CT) evaluation of bone morphometry aids in determining treatment strategies for shoulder instability. The use of zero echo time (ZTE) sequence in magnetic resonance imaging (MRI), a new bone cortex imaging technique, may help reduce radiation exposure and medical costs. Therefore, this study aimed to evaluate the glenoid morphology and detect the presence of bony Bankart lesion using ZTE MRI in shoulders with anterior instability and compare its diagnostic accuracy with that of CT. MethodsThirty-six patients (36 shoulders) with anterior instability who underwent preoperative CT and MRI examinations between April 2019 and October 2021 were retrospectively analyzed. The percentages of glenoid bone defects on three-dimensional (3D) CT and ZTE images were determined, and the correlation between these percentages was evaluated. The number of cases with bony Bankart lesion on CT and two types of ZTE (3D and CT-like) images was determined, and the diagnostic accuracy of ZTE for detecting bony Bankart lesion was assessed, with CT as the gold standard. Patients with bony Bankart lesion on CT images were divided into two groups based on whether the lesion was detectable on 3D ZTE or CT-like images. The longer diameters of bony Bankart lesion were compared between the groups. ResultsThe median percentage of glenoid bone loss was 12.1% (range, 1.3%–45.9%) and 12.3% (range, 0%–46.6%) on 3D CT and 3D ZTE images, respectively. The Spearman’s rank correlation coefficient was 0.89. Bony Bankart lesion was detected in 18, 13, and eight shoulders of the 36 patients on CT, 3D ZTE, and CT-like images, respectively. The overall diagnostic accuracy of the CT-like and 3D ZTE images for detecting bony Bankart lesion was 86.1% and 72.2%, respectively. A significant difference was observed between the groups with and without bony Bankart lesion on CT-like images in terms of the long diameter of the bone fragments on CT (p < 0.01). ConclusionsZTE MRI demonstrated high reproducibility for the evaluation of glenoid bone defect in shoulders with anterior instability. Although no significant difference in the measurement was observed compared with that on CT, the ability of ZTE MRI to delineate bone Bankart lesion remains limited.