To evaluate whether the use of intraoperative 3D fluoroscopy during acetabular fracture fixation reduces the rate of unplanned returns to the operating room and improves radiographic quality of reduction without increasing operative time or blood loss. Design: Retrospective comparative study. Single urban Level I trauma center. Patients with operatively treated acetabular fractures (OTA/AO 62) between 2017 and 2025. Patients were divided into two cohorts: those treated before and after the implementation of intraoperative 3D fluoroscopy (Cios Spin, Siemens, Germany).Outcome Measures and Comparisons: Primary outcome was the rate of unplanned return to the operating room. Secondary outcomes included postoperative residual joint gap, presence of intra-articular loose bodies, implant malposition, operative time, and estimated blood loss (EBL). These outcomes were compared between the two cohorts. 177 patients were included between 2017 and 2025, 148 patients treated before (control group) and 29 after the implementation of intraoperative 3D fluoroscopy (3D fluoro group). The mean age was 41.1 ±17.4 years old in the control group with 117 men (79%) vs 42.8 ±18.1 years old (p=0.55) in the 3D fluoro group, with 20 men (68.9%, p=0.34). The rate of unplanned return to the operating room was significantly lower in the 3D fluoroscopy group (0%) compared to the control group (15.5%) (p = 0.01). Postoperative CT scan (obtained in 148/148 in the control group and 22/29 in the 3D fluoroscopy group) demonstrated better fracture reduction in the 3D fluoroscopy group (mean residual gap: 3.4 ± 1.9 mm vs 4.5 ± 2.4 mm; p = 0.03). There were no significant differences in operative time (200.6 ±99.8 vs 221.2 ±114.4; p=0.49) or EBL (641.6 ±637.1 vs 674 ±572; p=0.67) between groups. Intraoperative findings seen on the 3D fluoroscopy such as malreduction, loose bodies or misplaced hardware led to modification of the surgical plan in 29% of cases where the 3D fluoroscopy was used. The use of intraoperative 3D fluoroscopy allowed intraoperative identification and correction of malreductions, implant malposition and removal of loose bodies, reducing the need for revision surgery and improving quality of reduction in acetabular fracture fixation. Its use did not increase operative time or blood loss, suggesting clinical and potentially economic benefits that may support broader adoption. Therapeutic Level III.

Emerging technologies to improve transbronchial sampling of lung lesions include mobile C-arm cone-beam CT (CBCT) and robotic assistance. Surgical Process Modeling (SPM) was used to quantify performance in such procedures performed using a conventional bronchoscope with guidance via 2D fluoroscopy and radial probe endobronchial ultrasound (RP-EBUS) ("Conventional Bronchoscopy") compared to robot-assisted bronchoscopy with CBCT guidance ("CBCT-Guided RAB"). Statistical SPMs were implemented for Conventional Bronchoscopy and CBCT-Guided RAB for simulation and analysis of procedural outcomes, including cycle time, radiation dose, and geometric accuracy. The SPMs were parameterized and validated with respect to clinical observation, published literature, and expert input. 9000 simulation runs were computed for each method, analyzing differences in performance and evaluating the influence of body mass index (BMI), lesion location (upper, middle, or lower lobe), and lesion size. The SPMs exhibited reasonable agreement with retrospective clinical evaluation of cycle time and dose, and variations in geometric accuracy were consistent with clinical literature. CBCT-Guided RAB resulted in a 14% increase in median cycle time (45.3 min) compared to Conventional Bronchoscopy (39.6 min) and increased median dose to the patient by 3.2 × (41.6 Gy cm2 compared to 12.9 Gy cm2). Geometric targeting improved with CBCT-Guided RAB, reducing the rate of geometric miss from 22% under Conventional Bronchoscopy to 2%. 3D visualization of individual runs gave clear depiction of median and outlier performance and a basis for communicating and standardizing complex workflows. SPMs yielded quantitative performance comparison in lung lesion biopsy by conventional and robot-assisted bronchoscopy. The approach quantified increases in cycle time and dose for CBCT-Guided RAB, accompanied by substantial gains in geometric accuracy. Such modeling provided valuable insight on the benefits of emerging technologies at early stages of implementation and a means to optimize and standardize clinical workflow.

Navik 3D (APN Health, Waukesha, WI, USA) is a navigation software program that uses two-dimensional (2D) fluoroscopy images to provide three-dimensional (3D) information. Left bundle branch area (LBBA) pacing (LBBAP) is a novel physiologic pacing technique where the lead is placed in the right ventricular (RV) basal septum to capture the left bundle branch (LBB). Precise lead placement in this region can be challenging using 2D fluoroscopy. We studied the feasibility of using Navik 3D to identify the location, plane, and depth of the lead in the septum to assist with LBBAP procedures. This observational, prospective single-center study included 14 patients undergoing LBBAP. Navik 3D was used to identify the LBBA, RV septum, RV apex, and lead position in three dimensions using two orthogonal 2D views. The 3D images were overlaid on real-time, gated fluoroscopic images for navigation of the lead. Images of the 3D locations and successful or unsuccessful lead locations were projected onto 2D fluoroscopic images, allowing for repositioning if necessary. All attempted patients had successful LBBA lead implants. An LBB potential was recorded in 61.5% of the patients. Selective LBBAP was achieved in 85% of the patients. The mean QRS duration postimplant was 129.8 ± 13.1 ms. The mean left ventricular activation time (stimulus R-wave peak in V6) postimplant was 75 ± 12 ms. No acute complications were recorded. 3D localization of the LBBA using the Navik 3D mapping system was feasible and may assist with more appropriate LBBA lead placement.

To assess image quality and reliability of intraoperative 3D fluoroscopy in assessing acetabular reduction. Design: Retrospective chart review. Academic, level 1 trauma center. Adult patients with acetabular fractures involving weight-bearing dome and/or posterior wall (OTA/AO 62A, 62B, 62C) between 2017-2023 with intraoperative 3D fluoroscopy and postoperative computed tomography (CT) included. Gap and step-off measurements made on intraoperative 3D fluoroscopy and postoperative CT by two graders using standardized technique. Assessments of image quality made using a scale of 1 (uninterpretable) to 5 (excellent). Interclass correlation coefficients (ICC) used to compare 3D fluoroscopy to CT and assessed intra/interobserver reliability for 3D fluoroscopy and CT images. Appropriate tests of significance performed. 54 patients with mean age of 48.4 ± 21.0 included. 72.2% of patients were male. 56/106 (52.8%) 3D fluoroscopic scans rated as 1/5 or 2/5. All CT images rated 3/5 or higher (p<0.001). 70% of CT measurements made with high confidence while 2.8% of 3D fluoroscopy measurements received this rating (p<0.001). Moderate agreement for gap (ICC=0.518, p<0.01) and step-off (ICC=0.420, p<0.01) when comparing 3D fluoroscopy to CT. CT showed good intra-reliability (ICC=0.747, 0.864; p<0.001) while 3D fluoroscopy showed moderate intra-reliability (ICC=0.638, 0.604; p<0.001). CT showed greater inter-reliability for largest gap (ICC 0.621, p<0.001) compared to fluoroscopy (ICC 0.219, p=0.05). When assessing acetabular fracture reduction involving the weight-bearing dome and/or posterior wall, intraoperative 3D fluoroscopy performed worse than postoperative CT on measures of image quality and confidence in reduction assessment. 3D fluoroscopy showed poor ICC across all views compared to CT and performed worse for interrater reliability. 3D fluoroscopy has limitations when making final acetabular reduction assessments; it may be advisable to acquire postoperative CT. Level III, diagnostic.

BackgroundInsufficient reduction of intra-articular fractures can lead to posttraumatic arthritis. An accurate intraoperative assessment of the articular surface is essential to guide surgical decision-making. This study aims to compare the sensitivity of 2D and 3D fluoroscopy for detecting intra-articular gaps and step-offs in distal radius fractures.MethodsIntra-articular distal radius fractures were induced in 6 cadaveric forearms and intra-articular incongruities (gap/step-off) of 0, 1 and 2 mm were created using 3D-printed repositioning guides. Images were acquired by 2D and 3D fluoroscopy for each step and evaluated by 3 blinded observers. Sensitivity and specificity for the detection of intra-articular incongruities were analyzed. Additionally, the inter-rater agreement for the decision to perform revision surgery and dose area product (DAP) were evaluated.ResultsThe overall sensitivity of 3D fluoroscopy for the detection of intra-articular incongruities was 90.28% compared to 55.56% for 2D fluoroscopy. Sensitivities for the detection of 2 mm incongruities were higher than for 1 mm especially in 2D fluoroscopy (1 mm: 38.89%; 2 mm: 74.72%). In 41.67% of cases with 1 mm incongruities where no revision was deemed necessary in 2D fluoroscopy, 3D imaging prompted a change in decision to perform revision. The dose area product was significantly higher for 3D fluoroscopy (41.87 cGycm2) compared to 2D fluoroscopy (3.13 cGycm2) (p < 0.0001).ConclusionThe results suggest that 3D fluoroscopy can facilitate the detection of intra-articular incongruities in distal radius fractures and may prompt intra-operative revisions. Further clinical studies are needed to determine the effect on surgical outcomes.

Awake deep brain stimulation (DBS) surgery has been associated with extended procedural time and patient discomfort. Asleep DBS, performed under general anesthesia with imaging-based guidance and confirmation, has emerged as an effective alternative. We describe a novel asleep DBS implantation technique using a Leksell stereotactic and the ARTIS Pheno robotic C-ARM with intraoperative 3D (3-dimensional) fluoroscopy (C-ARM 3DF) and assess accuracy and feasibility. Retrospective review of 9 patients who underwent asleep DBS lead implantation using a Leksell stereotactic frame and intraoperative C-ARM 3DF between January 2024 and February 2025 was performed. Planned lead coordinates in BrainLab software were compared with actual lead positions on postoperative computed tomography, and accuracy was quantified using Cartesian and radial errors. Among 9 patients (mean age 59.89 ± 19.23 years), the average Cartesian error between planned and actual lead positions was 1.25 ± 0.62 ± mm, and the mean radial error was 0.78 ± 0.31 mm. To assess intraoperative imaging reliability, we compared intraoperative C-ARM 3DF scans with postoperative computed tomography, which demonstrated a mean Cartesian deviation of 1.00 ± 0.80 mm. The average operative time was 105.44 ± 45.92 minutes overall, with bilateral cases averaging 122.57 ± 35.51 minutes and unilateral cases averaging 45.50 ± 7.78 minutes. Asleep DBS lead implantation using a Leksell stereotactic frame and intraoperative C-ARM 3DF achieves localization accuracy comparable with established awake and asleep methods. This technique is efficient and may be a viable alternative in centers with access to 3D fluoroscopic imaging.

Achieving anatomical reduction in tibial plateau fractures is essential to restore joint congruity and minimize long-term complications. Intraoperative assessment remains challenging, especially in complex fracture patterns, prompting the evolution of various imaging and adjunctive techniques. This narrative review summarizes advancements in intraoperative tools used for evaluating reduction in tibial plateau fractures. Techniques discussed include 2D fluoroscopy, intraoperative 3D imaging, arthroscopy, and emerging adjuncts such as 3D printing and cone-beam CT. We conducted a targeted literature search across PubMed, Embase, and Scopus (2000-2025) to identify studies reporting on the clinical application, diagnostic utility, and limitations of intraoperative imaging modalities in tibial plateau fracture management. 2D fluoroscopy remains widely used but is limited in detecting subtle incongruities. Intraoperative 3D imaging enhances accuracy and may reduce reoperations. Arthroscopy offers direct joint visualization, especially useful for assessing depression and posterior injuries. Emerging tools like 3D printing and cone-beam CT present novel intraoperative aids but require further validation. This review provides a consolidated overview of intraoperative strategies described in the literature. By highlighting the current capabilities and limitations of available technologies, we aim to inform ongoing clinical practice and future directions in fracture management.

Full-endoscopic extraforaminal lumbar discectomy: Use of 3-D image-guidance can mitigate risks and overcome steep learning curve.

Background/Objectives: Dorsal screw protrusion or intra-articular screw penetration at the distal radius can cause extensor tendon injuries or articular surface damage. Despite the use of various views, the detection of screw misplacement remains limited in 2D fluoroscopy. This study compares the sensitivity of 2D and 3D fluoroscopy for detecting screw misplacement at the distal radius. Methods: Volar locking plates were placed in six cadaveric forearms, and dorsal or intra-articular screw misplacement was induced. For each screw position, images were acquired by 2D and 3D fluoroscopy and assessed by three blinded observers. Sensitivity and specificity, inter-rater agreement, and observer confidence were evaluated. The dose area product (DAP) was measured separately for 2D and 3D fluoroscopy. Results: Three-dimensional fluoroscopy showed higher sensitivities for detecting dorsal (97.22%) and intra-articular (95.83%) screw misplacements than two-dimensional fluoroscopy. In 2D fluoroscopy, sensitivity for detecting dorsal screw protrusions improved from 63.89 to 75.00-77.78% with the inclusion of tangential views. For intra-articular penetrations, sensitivity in 2D fluoroscopy increased from 79.17 to 83.33% with the addition of oblique views. Observer confidence was higher in 3D fluoroscopy. DAP was significantly higher in 3D (42.4 ± 0.4 cGycm2) compared to 2D fluoroscopy (14.2 ± 3.7 cGycm2) (p < 0.0001). Conclusions: Compared to 2D fluoroscopy, 3D fluoroscopy improves the detection of screw misplacement at the distal radius. However, its routine use is constrained by increased radiation exposure and limited availability. If 3D fluoroscopy is not accessible, the addition of dorsal tangential and oblique views may improve the sensitivity of 2D fluoroscopy.

The use of 3D computer-assisted navigation and its influence on radiation exposure and operation time in the surgical treatment of fragility fractures of the pelvis.

IntroductionAugmented reality (AugR) is becoming a widely recognized and innovative platform in global healthcare. AugR has revolutionized cardiology by enhancing the understanding of cardiac structure and function. This review highlights its applications in diagnosis, surgical planning, cardiac procedures, training, rehabilitation, and the future impact of AugR-related technology.MethodsThis review compiles original research and review articles on AugR in cardiology from PubMed till 2024.ResultsAdvancements in visualization and image processing techniques facilitate the development of AugR tools using holographic displays, enhancing diagnostic accuracy and pre-surgical planning. Current AugR tools offer 3D heart imaging for diagnostic procedures, such as assessing Left Ventricular Ejection Fraction (LVEF). AugR enables real-time visualization for congenital and structural heart diseases, aiding in catheter navigation, transcatheter valve procedures, and arrhythmia treatments. Its effectiveness extends to cardiac resynchronization therapy, ventricular tachycardia ablation, and ultrasound-guided catheterization. AugR surpasses standard 2D fluoroscopy in surgical interventions by optimizing fluoroscopic angles, improving pacemaker placement, reducing X-ray exposure, and increasing procedural accuracy. It also enhances medical training by providing immersive experiences for residents and fellows, improving emergency response training. User-friendly AugR technologies effectively engage patients, promote physical activity, and enhance outcomes in cardiac rehabilitation. Further testing of AugR could serve as a pivotal surgical navigation tool in cardiac transplantology. Mixed reality enhances procedural planning and intraoperative navigation in cardiac electrophysiology by providing real-time 3D visualization and spatial orientation. Holographic visualization techniques combined with 3D and 4D printing hold future potential in cardiac care, particularly for designing patient-specific prosthetics. However, widespread clinical adoption of AugR in many healthcare institutions is limited by technical challenges and high costs related to specialized hardware, software, and maintenance.ConclusionAugR holds great promise in transforming cardiac care, but its clinical integration depends on rigorous trials to validate its effectiveness. While much research remains theoretical, increased human testing is essential for real-world applications. Advancing AugR, alongside technologies like 3D/4D printing and holography, could pave the way for a safer and more precise future in cardiology.

There is growing but so far inconclusive evidence about the value of three-dimensional (3D) fluoroscopy in calcaneal fracture care. The aim of this study was to retrospectively review our own cohort of calcaneus fractures before and after the implementation of 3D fluoroscopy and compare their outcome in terms of quality of reduction and amendments made based on the additional information gained by 3D imaging. Between March 2012 and October 2022, we operated on 28 calcaneal fractures in 25 patients of whom eight were treated with the aid of 3D fluoroscopy (= hybrid group). In all cases, a sinus tarsi approach (STA) and an angular stable plate were used. Intra- and postoperative scans were analyzed based on the Delphi Consensus criteria evaluating quality of reduction and radiographic outcome. Statistical analyses were performed to check for statistical significance and inter-rater reliability. We found no difference between the two groups (one group treated with conventional fluoroscopy and the other with 3D fluoroscopy) in terms of quality of reduction and radiographic outcome. Nevertheless, in more than half of our patients in the hybrid group, amendments were made based on the information gained through 3D fluoroscopy. Median radiation dose and surgical time was prolonged in the hybrid group but not leading to an increased infection rate. We found satisfactory outcomes in osteosynthesis of calcaneus fractures using the STA, regardless of whether conventional or 3D fluoroscopy was used intraoperatively. However, no significant improvement due to usage of intraoperative 3D fluoroscopy imaging in terms of restoration of the anatomy and quality of reduction could be found. Nevertheless, 3D fluoroscopy helps to avoid revision surgeries due to intra-articular placed screws and does not seem to lead to more cases of infection despite slightly prolonged operation times.

Percutaneous fixation of pelvic fractures using sacroiliac (SI) screws is recognized as a less invasive alternative to open surgery. Despite advancements in this technique, complications still occur. This study aims to evaluate the accuracy of IS screw placement assisted by intraoperative 3D navigation. A retrospective study was conducted on patients treated for sacral fractures or SI joint dislocations between 2016 and 2023 at a single institution. All procedures were performed using O-Arm 3D fluoroscopy and StealthStation S8 navigation platform (Medtronic). For each screw, all accuracy measurements-including length, angulation, and cortical perforation-were performed on the final 3D scans. Of the 241 screws implanted, 83.2% showed no perforation. Screw angulation was generally less accurate, with 26.5% of screws showing a minor deviation of 5° from the ideal trajectory. Accuracy was significantly lower in the axial plane compared to the coronal plane. On average, the implanted screws reached 78% of the maximum length the S1 corridor could accommodate. This study represents the largest series to date of SI screw implantation using 3D navigation. Navigation proved to be safe for IS screw placement. Although overall accuracy in our series was slightly lower than that reported in other studies using the O-Arm, our sample included six times more patients, multiple surgeons, and a heterogeneous cohort. 3D navigation with O-Arm proved more helpful in avoiding cortical perforations than in improving screw angulation.

To summarize current intra-operative techniques for assessing tibial plateau fracture reduction, with a particular focus on their feasibility and applicability in low-resource settings. This narrative review synthesizes literature describing intraoperative assessment strategies for tibial plateau fractures. Although no single study specifically addressed reduction assessment in low-resource environments, relevant articles discussing imaging-based assessments (e.g. fluoroscopy, 3D imaging), direct visualization approaches (e.g. arthrotomy), and tactile reduction techniques were included based on clinical relevance and applicability. Conventional 2D fluoroscopy remains the most widely used intraoperative imaging modality, despite its limited sensitivity for subtle articular incongruities. Direct visualization techniques, including arthrotomy, provide reliable tactile and visual assessment without requiring advanced imaging. Emerging innovations such as smartphone-assisted tools and 3D printing hold promise but currently lack validation in low-resource settings. There remains a notable gap in literature addressing tailored intraoperative assessment protocols for resource-constrained environments. In the absence of advanced intraoperative imaging, reliance on fundamental principles-such as thorough preoperative planning, direct visual and tactile assessment, and structured surgical technique-remains essential. There is an urgent need for context-specific research and pragmatic innovations to improve intraoperative fracture assessment and surgical outcomes in low-resource settings.

Accurate patient dosimetry estimates from fluoroscopically-guided interventions (FGIs) are hindered by limited knowledge of the specific anatomy that was irradiated. Current methods use data reported by the equipment to estimate the patient anatomy exposed during each irradiation event. We propose a deep learning algorithm to automatically match 2D fluoroscopic images with corresponding anatomical regions in computational phantoms, enabling more precise patient dose estimates. Our method involves two main steps: (1) simulating 2D fluoroscopic images, and (2) developing a deep learning algorithm to predict anatomical coordinates from these images. For part (1), we utilized DeepDRR for fast and realistic simulation of 2D x-ray images from 3D computed tomography datasets. We generated a diverse set of simulated fluoroscopic images from various regions with different field sizes. In part (2), we employed a Residual Neural Network (ResNet) architecture combined with metadata processing to effectively integrate patient-specific information (age and gender) to learn the transformation between 2D images and specific anatomical coordinates in each representative phantom. For the Modified ResNet model, we defined an allowable error range of ± 10mm. The proposed method achieved over 90% of predictions within ± 10mm, with strong alignment between predicted and true coordinates as confirmed by Bland-Altman analysis. Most errors were within ± 2%, with outliers beyond ± 5% primarily in Z-coordinates for infant phantoms due to their limited representation in the training data. These findings highlight the model's accuracy and its potential for precise spatial localization, while emphasizing the need for improved performance in specific anatomical regions. In this work, a comprehensive simulated 2D fluoroscopy image dataset was developed, addressing the scarcity of real clinical datasets and enabling effective training of deep-learning models. The modified ResNet successfully achieved precise prediction of anatomical coordinates from the simulated fluoroscopic images, enabling the goal of more accurate patient-specific dosimetry.

BackgroundThree-dimensional rotational angiography (3DRA) allows real-time 3D reconstruction of the left atrium (LA) and integration in the fluoroscopy image. As such it provides anatomical guidance which can be valuable during single shot ablation procedures. There are no data estimating efficiency and effectiveness in the context of pulsed field ablation (PFA) of atrial fibrillation (AF).PurposeWe sought to investigate the impact of 3DRA on the procedural parameters in consecutive patients who underwent AF ablation with PFA technology.MethodsAll consecutive patients who underwent AF ablation with PFA at two experienced centers from November 2023 until October 2024 were eligible for this study. Operators performed pulmonary vein isolation (PVI) using device-specific instructions resulting in eight applications (four basket and 4 flower poses) for each vein. Additional lesions were done at the operator discretion. Procedural metrics were analysed according to the use of 3DRA which was used according to institutional standards.ResultsFour hundred and four patients were included in this study. 62.9% of the population had paroxysmal AF, and 37.1% had persistent AF. 3DRA was used in 239 cases (59.2%) (Figure 1). Patients with peri-procedural 3DRA received fewer additional applications compared to patients without (applications per patients 38 [32-45] vs 40 [34-53], p =0.012). Median fluoroscopy time was 15 [10-22] min, median LA dwell time was 30 [25-37] min, and median skin-to-skin time was 41 [35-51] min. The use of 3DRA improved the procedural metrics (3DRA vs no 3DRA: 39 [34-48] min vs 45 [38-53] min for skin-to-skin time, p <0.001; 11 [9-16] min vs 20 [15-24] min for total fluoroscopy time p <0.001, and 9 [7-13] vs 20 [15-24] for fluoroscopy time during PFA) (Figure 2).ConclusionIn our study, the integration of 3DRA during PVI with PFA may provide advantages in terms of procedural metrics, such as the decrease in both the total procedural time and the total fluoroscopy dose. These improvements may be a result of the anatomical information provided by the 3D model.

This study evaluated the accuracy of an automated 2D-3D registration software for matching preoperative 3D models of the pelvis and acetabular component to intraoperative 2D fluoroscopy images in total hip arthroplasty (THA). We developed a 2D-3D registration software that registers a 3D model of the pelvis from preoperative CT and a 3D model of the acetabular implant to intraoperative fluoroscopic imaging, thereby calculating the implant position relative to the 3D pelvic reference frame. A total of 145 datasets were used including 65 digitally reconstructed radiographs, 20 dry bone phantoms datasets and 60 clinical datasets with preoperative CT and intraoperative fluoroscopy imaging. Achieved acetabular positions from the clinical images were determined from post-operative CT using a 3D/3D registration method. Accuracy was assessed by comparing the calculated acetabular position from the 2D-3D software to the ground truth data. Mean absolute difference between ground truth and the 2D-3D software was 1.9° [signed error range: -4.4, 4.8] for inclination, 1.5° [-7.3, 4.1] for anteversion, 1.6mm [-5, 3.8] for cup height and 1.8mm [-7.3, 4.1] for depth across all datasets. In total, 100% of inclination results and 98% of anteversion results were within 5° while 90% of height and 81% of depth results were within 3mm. We validated the accuracy of an automated 2D-3D registration software for use in THA. While our method requires preoperative data from CT, the results are comparable to robotics and image-based navigation, and present a promising, simple technology that can be easily integrated into an operating room for THA.

The accurate segmentation of guidewires in interventional cardiac fluoroscopy videos is crucial for computer-aided navigation tasks. Although deep learning methods have demonstrated high accuracy and robustness in wire segmentation, they require substantial annotated datasets for generalizability, underscoring the need for extensive labeled data to enhance model performance. To address this challenge, we propose the Segmentation-guided Frame-consistency Video Diffusion Model (SF-VD) to generate large collections of labeled fluoroscopy videos, augmenting the training data for wire segmentation networks. SF-VD leverages videos with limited annotations by independently modeling scene distribution and motion distribution. It first samples the scene distribution by generating 2D fluoroscopy images with wires positioned according to a specified input mask, and then samples the motion distribution by progressively generating subsequent frames, ensuring frame-to-frame coherence through a frame-consistency strategy. A segmentation-guided mechanism further refines the process by adjusting wire contrast, ensuring a diverse range of visibility in the synthesized image. Evaluation on a fluoroscopy dataset confirms the superior quality of the generated videos and shows significant improvements in guidewire segmentation.

IntroductionPercutaneous sacroiliac screw fixation of pelvic fragility fractures is increasingly being used to maintain mobility and reduce pain in the elderly patient population. Traditionally, this is performed using 2D fluoroscopy. Several newer, navigated techniques have emerged that may further facilitate this procedure. It, however, remains unclear whether there is a benefit regarding accuracy, radiation exposure and complications of these new navigation techniques when compared to the traditional 2D fluoroscopy.MethodsA systematic review and meta-analysis were performed. PubMed, CENTRAL and Embase were searched for both randomized controlled trials and observational studies comparing new navigation techniques to 2D fluoroscopy for percutaneous sacroiliac screw fixation. Effect estimates were pooled (random effects) and presented as odds ratio, mean difference and standardized mean difference with a 95% confidence interval.Results19 studies were included. The 2D fluoroscopy group had 642 patients and the new navigation group 663 patients. Accuracy was significantly higher in the new navigation group (OR 2.44, 95% CI 1.53–3.90), especially O-Arm, 3D CT and Robotic navigation. On average, accuracy was 82% in the 2D group and 92% in the new navigation group, which was significant. Also, fluoroscopy time (MD 71.89 s, 95% CI 51.37–92.41) and frequency (MD 17.22 images in total, 95% CI 7.73–26.70) were significantly reduced in the new navigation group. Complications are acceptably low, however, poorly reported in both groups.ConclusionThis meta-analysis demonstrated a higher accuracy, lower fluoroscopic frequency and time for new navigation techniques compared to 2D fluoroscopy. More advanced navigation techniques, such as 3D CT and robotic navigation, appeared to be even better.

BackgroundDorsal screw protrusion can lead to complications such as extensor pollicis longus (EPL) tear or rupture after volar locking plate (VLP) fixation. Previous studies displayed that intraoperative 3D fluoroscopy and skyline view had similar diagnostic accuracy. This study investigated the efficacy of intraoperative 3D fluoroscopy compared to skyline view for detecting dorsal cortex screw protrusion in VLP procedures for unstable intra-articular distal radius fractures (DRF). We used postoperative computed tomography (CT) to assess the efficacy and addressed the limitations of previous methods in evaluating screw penetration accurately.MethodsWe utilized the ICUC database, a prospective cohort of patients with surgically treated DRF, to collect cases with available images, including skyline views, intraoperative 3D fluoroscopy, and postoperative CT scans. The postoperative CT confirmed whether the screw protruded through the dorsal cortex. The interrater reliability was assessed using Cohen’s Kappa, and a diagnostic test was utilized to compare the two intraoperative imaging modalities.ResultsTwenty-one unstable DRFs were included in the study. The agreement between skyline view and postoperative CT was moderate agreement, with a kappa value of 0.481 (95% CI: 0.297–0.652, N = 84), identifying 10 uncertain, 56 shorter screws, and 18 screw penetrations. Intraoperative 3D fluoroscopy demonstrated almost perfect agreement with postoperative CT, with a kappa of 0.839 (95% CI: 0.703–0.975, N = 84), identifying 62 shorter screws and 22 screw penetrations. The sensitivity and specificity of intraoperative 3D fluoroscopy in detecting dorsal screw protrusion were 81.8% and 98.4%, respectively, while the skyline view’s sensitivity and specificity were 72.2% and 90.9%.Conclusion3D fluoroscopy offers an almost perfect evaluation, whereas the skyline views provide only moderate agreement. 3D fluoroscopy could reduce cumulative radiation exposure of surgeon and patient compared to skyline view. Clinically, 3D fluoroscopy would be beneficial for surgeons to evaluate dorsal screw protrusion precisely and safely.