Multi-head Self-Attention and Window-Attention Transformer Model for Multi-label Abdominal Trauma Classification Using 3D CT Scans

ABSTRACT T‐spherical fuzzy sets (TSFSs) provide a robust and flexible framework for modeling uncertainty by aggregating membership, nonmembership, and neutrality degrees using t‐norms. This study introduces a new family of set‐theoretic weighted similarity and distance measures specifically designed for TSFSs, which should pique curiosity about their innovative approach. This study demonstrates that numerous traditional measurements fail to satisfy fundamental axiomatic properties. Conversely, our suggested measurements demonstrate unwavering reliability and efficacy over a wide range of scenarios. To validate the practical utility of these measures, we integrate them into a deep learning classification framework. Finally, we apply the proposed similarity measurement to deep learning classification of 3D CT scans. It can improve classification accuracy by effectively handling uncertain features in medical imaging data.

This study aimed to evaluate the dosimetric impact of respiratory motion after breathing training on Hybrid and IMRT plans in hypofractionated breast radiotherapy, using 4D-CT and deformable dose accumulation. Twenty-two postoperative breast cancer patients treated under free breathing underwent both 3D-CT and 4D-CT scanning after abdominal breathing training. Original Hybrid (O_HY) and IMRT (O_IM) plans were generated on 3D-CT. Each plan was mapped onto the ten 4D-CT phases and recalculated. Deformable image registration was applied to accumulate phase-specific doses onto the original 3D-CT, generating motion-inclusive deformed Hybrid (D_HY) and IMRT (D_IM) plans. Respiratory motion amplitudes of the chest wall and liver were quantified. Dosimetric parameters of targets and OARs were evaluated and compared. After breathing training, respiratory motion amplitudes were 13.2 ± 4.6 mm for the liver and 1.54 ± 0.85 mm for the chest wall. Both techniques achieved good target coverage in the original plans, with comparable hotspot, homogeneity, and conformity. Hybrid plans provided superior OAR sparing, particularly in the low-dose regions. After deformable dose accumulation, target coverage exhibited only minor decreases: PTV_V100% decreased by 2.5% for Hybrid and 2.1% for IMRT. PTV_D2% remained largely unchanged, and HI/CI deteriorated only slightly. OAR dose variations were minimal, with no significant differences except for a lower Breast_Dmean in the IMRT plan. Respiratory motion showed no meaningful correlations with most dosimetric parameters, except for weak correlations between liver motion and PTV_V95% and a significant correlation with Breast_Dmean in the IMRT plan. In this cohort, chest wall motion amplitude after breathing training remained within 3.1 mm. Under these conditions, Hybrid and IMRT plans demonstrate comparable target dose robustness, while Hybrid achieves superior OAR sparing, particularly in the low-dose region. Based on this comparison, Hybrid planning may be considered a favorable option relative to IMRT for hypofractionated whole-breast irradiation under free breathing.

Radiation pneumonitis occurs in approximately 10-30\% of lung cancer patients treated with radiation therapy, posing a significant dose-limiting factor. The assessment of regional ventilation changes from functional ventilation data can provide essential information regarding treatment response. However, this task can be challenging since ventilation maps contain noisy measurements and artifacts.
Methods: We introduce a framework that estimates physiological changes from a set of longitudinal ventilation scans. Our method identifies changes as more plausible if they follow a monotonic trend while attributing smaller confidence in regions where large fluctuations are observed. Our algorithm outputs the estimated volumes of significant function increase and decline. The proposed framework was calibrated and validated using synthetic datasets. We also applied our model to a dataset comprising 11 lung cancer patients for whom multiple 4DCT scans were obtained during the course of radiotherapy treatment. CT-derived ventilation maps were generated and used as input to the proposed framework. In order to create a control dataset where no functional changes were expected, we also shuffled the time points for the 11 patients in every possible way that discarded as much temporal information as possible resulting in 128 functional map sequences.
Results: In the patient dataset, 3/11 patients were identified with significant functional decline and 4/11 with functional increase that was associated with tumor regression. Finally, in the control dataset the frequency of occurrence of significant changes was 1.6\% (4/256) compared to 32\% (7/22) for the original patient dataset.
Conclusion: We have developed a framework for analyzing functional ventilation changes from longitudinal data. The results of the lung cancer patient dataset indicate that significant functional increase and decline can occur during the course of radiotherapy treatment. More generally, the developed framework can be used to assess ventilation changes with the potential of guiding adaptive treatment strategies.&#xD.

CBCT-Guided Pencil Beam Scanning Proton Therapy for Moving Lung Tumours: PROMO Study- Standardised Workflow and Early Outcomes From a Consecutive Real-World Cohort.

Study on the impact of tumor volume change and radiation dose on 4D-CT-based lung ventilation function at mid-treatment adaptive radiotherapy in stage III non-small cell lung cancer.

Study on seepage characteristics of concrete cracks based on CT 3D reconstruction and fractal theory

Background: Wrist arthroscopy is widely used to assess the intactness of ligaments, including the scapholunate ligament (SLL), as it allows direct visualisation and palpation. However, arthroscopy is invasive and costly, while traditional imaging modalities, such as X-rays and MRI, often miss ligament injuries due to their static nature. A promising alternative is four-dimensional computed tomography (4DCT), which allows dynamic, non-invasive visualisation of wrist motion in real time. This study evaluates the diagnostic potential of 4DCT for detecting SLL tears by comparing its findings with arthroscopy and open surgery. Methods: Twenty-one patients with suspected SLL tears underwent 4DCT imaging, with independent reviews by two radiologists. The motion cycles of extension-flexion and radial-ulnar deviation were analysed. Arthroscopy was performed within a few weeks after the scan. Patients with Geissler grade III or IV tears underwent surgical reconstruction. The qualitative assessments of the SLL by radiologists and surgeons were compared, while the kinematic parameters of the injured and contralateral wrists were analysed using a linear mixed model. Results: Radiologists' 4DCT evaluations corresponded with arthroscopy or surgical findings in 14 of 21 patients (67%). Patients with total SLL tears showed greater scapholunate distance and scapholunate angle during all wrist movements compared with uninjured wrists. Notably, 4DCT showed greater diagnostic agreement with surgical findings than arthroscopy alone. Conclusions: These findings demonstrate the potential role of 4DCT in diagnosing SLL tears. Further research should focus on optimising the acquisition and analysis of the 4DCT scan and refining 4DCT parameters to distinguish partial from total SLL tears. Level of Evidence: Level III (Therapeutic).

Robotic systems are transforming image-guided interventions by enhancing accuracy and minimizing radiation exposure. A significant challenge in robotic assistance lies in surgical path planning, which often relies on the registration of intraoperative 2D images with preoperative 3D CT scans. This requirement can be burdensome and costly, particularly in procedures like vertebroplasty, where preoperative CT scans are not routinely performed. To address this issue, we introduce a differentiable rendering-based framework for 3D transpedicular path planning utilizing bi-planar 2D X-rays. Our method integrates differentiable rendering with a vertebral atlas generated through a Statistical Shape Model (SSM) and employs a learned similarity loss to refine the SSM shape and pose dynamically, independent of fixed imaging geometries. We evaluated our framework in two stages: first, through vertebral reconstruction from orthogonal X-rays for benchmarking, and second, via clinician-in-the-loop path planning using arbitrary-view X-rays. Our results indicate that our method outperformed a normalized cross-correlation baseline in reconstruction metrics (DICE: 0.75 vs. 0.65) and achieved comparable performance to the state-of-the-art model ReVerteR (DICE: 0.77), while maintaining generalization to arbitrary views. Success rates for bipedicular planning reached 82% with synthetic data and 75% with cadaver data, exceeding the 66% and 31% rates of a 2D-to-3D baseline, respectively. In conclusion, our framework demonstrates the feasibility of versatile, CT-free 3D path planning for robot-assisted vertebroplasty, accommodating diverse intraoperative imaging conditions without requiring preoperative CT scans.

Introduction: Gnatoschizis (alveolar cleft), is a developmental defect of the maxillary alveolar bone, occurring in up to 75% of patients with cleft lip and palate. This condition may impair mastication and cause persistent oronasal fistulas if it is left untreated. Gnatoplasty with bone grafting is a reliable procedure to restore alveolar continuity and oral function. This report describes the management of bilateral gnatoschizis with bilateral gnatoplasty and bone graft reconstruction. Case Report: A 25-year-old female presented with nasal fluid leakage during meals. Two small openings were foung in the upper labial mucosa, forming oronasal communications. OPG and 3D CT scan confirmed anterior maxillary defects directly communicating with the nasal cavity. The patient treated by bilateral gnatoplasty under general anesthesia. Closure of nasal and oral mucosa were performed to seal the fistulas. Bone grafting followed without membrane coverage using AmGraft® (freeze-dried dentin matrix, allograft) and Bonefill Ortho® (bovine hydroxyapatite, xenograft). Postoperative CBCT demonstrated well-positioned grafts completely filling the defects and patient reported no nasal leakage and regained normal mastication. Discussion: This case highlights the effectiveness of gnatoplasty with bone grafting performed without membranes. This approach simplifies the procedure while still achieving both anatomical repair and functional rehabilitation. Conclusion: Bilateral gnatoplasty with bone graft reconstruction is effective for closing oronasal communication and restoring oral function, without using membranes, making this approach practical for patients with residual cleft defects.

Early identification and appropriate surgery are essential in the management of patients with posteromedial rotatory instability of the elbow to prevent the development of early post-traumatic osteoarthritis. The aim of this study was analyze the radiological findings in order to suggest appropriate treatment strategies. A total of 58 patients with posteromedial rotatory instability of the elbow, with available 3D CT and MRI scans, were included. We investigated the size of the anteromedial coronoid facet fracture and the patterns of injury to the lateral collateral ligament complex and ulnar collateral ligament, relating them to the O'Driscoll subtypes 2-1, 2-2, and 2-3 of injury. Subtypes 2-1, 2-2, and 2-3 comprised eight, 41, and nine patients, respectively. The mean size of the coronoid fracture significantly increased from subtype 2-1 (3.2 mm (SD 1.5)) to 2-3 (10.0 mm (SD 2.2mm)) (p < 0.001). The rate of rupture of the lateral ligament and the requirement for repair decreased significantly from 100% in subtype 2-1 to 33% in subtype 2-3 (p = 0.038, p = 0.005). The rate of rupture of the posterior bundle of the ulnar ligament increased significantly from 25% in subtype 2-1 to 100% in subtype 2-3 (p = 0.006). There are distinct patterns of radiological injury in the subtypes of posteromedial instability. Subtype 2-1 is primarily associated with substantial lateral ligamentous injury, whereas subtype 2-3 features larger coronoid fractures and an increased rate of associated injury to the posterior bundle of the ulnar ligament; subtype 2-2 is intermediate. These findings inform a proposed algorithm for the surgical treatment: lateral ligamentous repair is key for 2-1, and fixation of the coronoid fracture fixation is key for 2-3, with intraoperative assessment guiding further steps. This observational study establishes a theoretical framework for the management of these patients, which will require validation in the future.

Four-dimensional computed tomography (4D CT) imaging is essential for radiation therapy planning in thoracic tumors. However, current protocols tend to acquire more projection data than is strictly necessary for reconstructing the 4D CT, potentially leading to unnecessary radiation exposure and a misalignment with the ALARA (As Low As Reasonably Achievable) principle. We propose a deep learning (DL)-driven approach that uses the patient's breathing signal to guide data acquisition, aiming to acquire only necessary projection data. This retrospective study analyzed 1415 breathing signals from 294 patients, with a 75/25 training/validation split at the patient level. On the basis of the signals, a DL model was trained to predict optimal beam-on events for projection data acquisition. Model testing was performed on 104 independent clinical 4D CT scans. The performance of the model was assessed by measuring temporal alignment between predicted and optimal beam-on events. To assess the impact on the reconstructed images, each 4D CT data set was reconstructed twice: using all clinically acquired projections (reference) and using only the model-selected projections (dose-reduced). Reference and dose-reduced images were compared using Dice coefficients for organ segmentations, deformable image registration-based displacement fields, artifact frequency, and tumor segmentation agreement, the latter evaluated in terms of Hausdorff distance and tumor motion ranges. The proposed approach reduced beam-on time and imaging dose by a median of 29% (IQR, 24%-35%), corresponding to 11.6 mGy dose reduction for a standard 4D CT CTDIvol of 40 mGy. Temporal alignment between predicted and optimal beam-on events showed marginal differences. Similarly, reconstructed dose-reduced images showed only minimal differences to the reference images, demonstrated by high lung and liver segmentation Dice values, small-magnitude (deformable image registration) displacement fields, and unchanged artifact frequency. Minor deviations of tumor segmentation and motion ranges compared with the reference suggest only minimal impact of the proposed approach on treatment planning. The proposed DL-driven data acquisition approach has the ability to reduce radiation exposure during 4D CT imaging while preserving diagnostic quality, offering a clinically viable, ALARA-adhering solution for 4D CT imaging.

Although previous research has examined the first lumbar (L1) vertebra for sex estimation, studies focusing on its role in age estimation have been limited. This study expands on earlier work by investigating the morphological variations in shape and size of the L1 vertebra across different age groups within the Malaysian population. A sample of 440 abdominal CT images was collected from the Radiology Department at Universiti Kebangsaan Malaysia Medical Centre. These images included adults aged 18 to 80, who visited the department in 2019. Twenty-seven 3D landmarks were marked on each L1 vertebra using digitized 3D CT scan images. Statistical analyses were performed using a geometric morphometric approach to evaluate age-related variations in the shape and size of the L1 vertebra. Principal Component Analysis identified 74 shape variables describing the shape of the L1 vertebra, with the first five principal components explaining 38.27% of the variance. The Canonical Variate Analysis scatter plot showed slight separation among the confidence ellipses for the three age groups, with significant p-values (p < 0.001). Procrustes ANOVA revealed significant differences in both the size and shape of the L1 vertebra across all age groups. Additionally, multivariate regression of shape on continuous age revealed a significant, biologically meaningful pattern (R² = 0.022, p = 0.001). This study shows that the size and shape of the L1 vertebra differ across various age groups. In elderly individuals, the L1 vertebra is characterized by longer spinous processes and shorter, flatter vertebral bodies. Conversely, younger individuals tend to have L1 vertebrae with shorter transverse spinous processes and taller vertebral bodies.

Chronic subdural hematoma (CSDH) commonly affects older adults and remains associated with recurrence despite surgical evacuation. Coagulation of middle meningeal artery (MMA) during burr-hole drainage may be useful where MMA embolization is unavailable, however reliable external landmarks for MMA localization are poorly defined. This study aimed to identify the MMA confluence point on 3D cranial CT scans of elderly Filipino patients and define its relationship to external cranial landmarks to support pre-operative planning. A retrospective cross-sectional morphometric study was performed using plain cranial CT scans of patients aged ≥65 years from 2019 to 2023. Scans with intact calvarial anatomy and adequate visualization of MMA groove were included. Threedimensional reconstructions were generated, and bilateral distances from lateral canthus, external auditory canal, and canthomeatal line to MMA confluence point were measured. Analyses were performed by sex and laterality (p<0.05). A total of 221 patients were included (mean age 76 years; 63% female). The external auditory canal and lateral canthus showed low variability. Male patients had greater external auditory canal and canthomeatal line distances (p<0.05), while confluence measurements did not differ by sex. The confluence point was located farther from the canthomeatal line on the left (p=0.0125). These findings provide a practical framework for MMA localization during pre-operative planning. In settings without MMA embolization, landmark-based localization offers a low-cost method to guide burr-hole placement and potential MMA coagulation, supporting resource-adapted strategy to reduce CSDH recurrence.

This study aimed to evaluate the accuracy of cephalometric analyses performed by deep learning-based AI programs (NemoCeph 2D, OrthoDx, AudaxCeph, and WebCeph) by comparing their results with the gold standard measurements obtained from 3D CT scans in orthognathic surgery patients. Orthognathic surgery candidates underwent pretreatment 3D CT scans. These scans were manually analyzed using 3D cephalometric software to establish gold-standard landmark positions. Two-dimensional cephalometric images were then derived from the 3D scans, and deep learning-based AI programs automatically identified the landmarks on these images. The AI-generated measurements were compared with the 3D gold standard, and the differences were analyzed statistically. While the ANB angle showed no significant differences between the methods (p=0.061), other measurements-including SNA, SNB, Wits appraisal, Y Axis Angle, and various facial height ratios-showed significant discrepancies (p<0.05). AI-based cephalometric analyses showed notable errors compared with the 3D CT gold standard. These findings suggest that deep learning algorithms require further refinement before they can be reliably used for orthognathic surgery planning.

Primary hyperparathyroidism (PHPT) has varied presentations whose successful surgical management demands integration of the expanding literature within a comprehensive decision-making framework. Our objective is to compare the workup and management of variant (i.e., normocalcemic hyperparathyroidism (NCPHPT) and normohormonal hyperparathyroidism (NHPHPT)) to classic hyperparathyroidism (CPHPT), taking into account localization studies, surgical considerations, utility of intraoperative parathyroid hormone (IOPTH), and surveillance to identify gaps in knowledge that limit management outcomes of variant PHPT presentations. A comprehensive search of PubMed and Google Scholar was conducted from June 11, 2024, to March 17, 2025, using MeSH terms and free-text queries focused on imaging, surgery, and outcomes in PHPT and reported using PRISMA ScR guidelines. Peer-reviewed studies addressing surgical management of PHPT in adults, published in English, were included. Two reviewers independently screened all titles, abstracts, and full texts using previously developed inclusion criteria recorded in a shared spreadsheet. Twenty-seven studies were included. Data were synthesized into a flow diagram to identify areas in need of further guidance. Imaging accuracy was highest with 4DCT and PET scans in both classic and variant PHPT, particularly in the setting of multiglandular disease. Not obtaining ≥ 50% IOPTH drop and/or levels below 40 pg/mL are risk factors for persistence or recurrence. PHPT has varied presentations that demand nuanced approaches. Our review highlights knowledge gaps that, if addressed, will improve the care of patients with variant PHPT. Further prospective validation of this algorithmic summary is warranted to improve long-term impact and patient outcomes.

Age estimation plays a vital role in forensic science as it helps in reconstructing the biological profile of a missing person. Cranial suture obliteration is one of the methods used for age estimation. However, the varying accuracy of age assessment through cranial suture obliteration has been found in various populations. The present study aims to generate population-specific regression models by assessing cranial suture obliteration in 3D CT scan images for the estimation of age in the Bengali population of Eastern India using the Acsádi and Nemeskéri method. This study included 221 female and 192 male Bengali individuals of Kolkata, who underwent a 3D CT scan using the DSCT SOMATOM CT scan instrument. Significant correlation ( p &lt; .05) found between age and cranial suture obliteration score. 44%–46.2% of estimated age showed higher accuracy (±2.5 years of actual age) in males with age groups of 1–9 years and 50–59 years, when estimated age was calculated through sagittal suture obliteration score. In females, 43.8% of estimated age showed higher accuracy (±2.5 years of actual age) among the age group (40–59 years), when estimated age was calculated through the coronal suture. The Acsádi and Nemeskéri method is an effective method to estimate age in the Bengali population from cranial suture obliteration, with some error due to adaptive response against intrinsic and extrinsic factors.

Towards 3D-dense ultrasound image simulation from 2D CT scans for ultrasound-guided percutaneous nephrolithotomy: a progressive training approach from basic to advanced simulator complexity.

BackgroundThis study aimed to compare the clinical and radiological outcomes of the embrace fixation (EF) technique with those of suture button fixation (SF) for syndesmotic injuries in ankle fractures, with a focus on postoperative CT-assessed reduction quality, functional scores, and complication rates.MethodsA retrospective cohort study was conducted on 148 patients (86 EF and 62 SF) treated between January 2021 and January 2023. Preoperative and postoperative assessments included radiographies and 3D-CT scans, American Orthopedic Foot & Ankle Society (AOFAS) hindfoot scores, Olerud-Molander Ankle (OMA) scores, and visual analog scale (VAS) pain scores. The primary outcome was the AOFAS hindfoot score at the final follow-up.ResultsBoth groups demonstrated excellent clinical outcomes, with no significant differences in AOFAS, OMA, or VAS scores. Postoperative CT analysis revealed low malreduction rates. The complication rates were comparable and primarily involved minor soft tissue irritation.ConclusionBoth the embrace and suture button techniques demonstrated comparable accuracy and effectiveness for syndesmotic stabilization in ankle fractures and yielded excellent radiological and clinical outcomes. Notably, the embrace technique yielded equivalent outcomes compared with suture button fixation for managing syndesmotic injuries.Level of evidenceLevel III, retrospective case-control study, therapeutic.

The use of alendronate to enhance transcranial transmission of focused ultrasound for successful ablations in brain.