Although anatomically resectable hepatocellular carcinoma (HCC) is generally treated with surgery alone, a substantial proportion of patients experience early recurrence, suggesting underlying biological aggressiveness. We aimed to identify a biologically high-risk subgroup among anatomically resectable HCCs and to develop a simple preoperative biology-based risk score. In this multicenter retrospective cohort study, patients who underwent curative-intent hepatectomy for anatomically resectable HCC were temporally divided into training and validation cohorts. Early recurrence was defined as recurrence within 2 years after surgery. Independent predictors were identified using multivariable logistic regression and incorporated into a risk score. Predictive performance was assessed by recurrence-free survival (RFS), AUC, Akaike information criterion, and calibration analysis. Alpha-fetoprotein (AFP), des-gamma-carboxy prothrombin (DCP), and the fibrosis-related host factor FIB-3 were independently associated with early recurrence and were integrated into the ADF (AFP, DCP and FIB-3) score (range, 0-3). In the validation cohort, higher ADF scores were associated with progressively worse RFS (p < 0.001). The ADF score outperformed each individual component alone, showed good calibration, and the addition of tumor size improved discrimination. The ADF score identifies a biologically high-risk subgroup within anatomically resectable HCC that is prone to early postoperative recurrence. This biology-based framework may complement anatomical assessment and help refine preoperative risk stratification.
To evaluate the clinical role, durability, anatomical response, and safety for suprachoroidal triamcinolone acetonide (SCTA) (Xipere) in routine clinical practice. Data were manually extracted from health records at a high-volume retina practice in Cleveland, Ohio. All patients with noninfectious inflammatory macular edema undergoing SCTA (≥1 billing code) between October 25, 2021, and July 17, 2025, were included. Patients with known systemic autoimmune or infectious associations were excluded. Unique eyes were defined by MRN + laterality, with "OU" entries split into OD/OS. Patient demographics, underlying diagnosis categorization, time between first and second injection (for eyes with ≥2 injections) were summarized. Pre- versus post-treatment central retinal thickness (CRT) and intraocular pressure (IOP) were compared via paired t-tests. A total of 177 patients (195 eyes) received 340 SCTA injections for macular edema associated with an underlying diagnosis of intermediate uveitis (10%), pseudophakic cystoid macular edema (31%), and posterior uveitis (59%). The time between first and second injection was approximately 5 months [SD ± 72.5; median 144 (min 35, max 427)]. From baseline to follow-up, mean CRT decreased by 103 µm [95% CI: (-122,-83.2), P < 0.001], while IOP increased minimally [mean difference 0.7 mmHg, 95% CI: (-0.02, 1.43), P = 0.057]; and was medically managed when elevated. In this real-world cohort, findings suggest that SCTA achieves robust anatomical improvement, a favorable IOP profile, and prolonged durability across patients with inflammation-related macular edema.
This study aimed to elucidate the effects of the tibialis anterior (TA) muscle contraction on the medial longitudinal arch (MLA) and hallux valgus (HV) angle, primarily focusing on the potential functional role of the TA in foot alignment. Twenty-five healthy adults (mean age: 20.7 ± 1.0 years) participated. Electrical stimulation was applied to the TA without ankle dorsiflexion. A triaxial accelerometer was attached to the skin over the navicular bone, and electrogoniometers were placed over the interphalangeal joints of the hallux and ankle. The navicular displacement was calculated through double integration of the acceleration data. Contraction of the TA caused navicular displacement in the supination direction (upward, posterior, and medial), suggesting a temporary elevation of the MLA. The hallux showed an average varus shift of 1.36°; the shift was observed in 24 of 25 participants (p < 0.001). A significant positive correlation was found between upward displacement of the navicular bone and the extent of varus change in the HV angle (r = 0.463, p = 0.020). Additionally, a greater initial HV angle was associated with a larger varus shift in the hallux after TA contraction (r = 0.433, p = 0.030). No significant correlations were observed between the HV angle change and ankle dorsiflexion, suggesting that the medial hallux shift may result from structural changes in the MLA rather than isolated joint motion. These findings provide novel evidence that TA contractions can elevate the MLA and induce varus displacement of the hallux. These findings further suggest that the TA contributes to foot biomechanics not only as an isolated muscle but also as part of a coordinated muscle system involved in arch dynamics and toe alignment. This suggests the potential clinical applications of TA activation in noninvasive interventions for mild-to-moderate HV, such as exercise therapy or neuromuscular electrical stimulation.
暂无摘要（点击查看详情）
The coronary sinus and its tributaries constitute the anatomical foundation for successful cardiac resynchronization therapy. Despite the emergence of conduction system pacing as a physiological alternative, conventional transvenous CRT retains its important role-not all clinical scenarios can be addressed by left bundle branch area or His bundle pacing alone. Understanding coronary venous anatomy therefore remains essential for every implanting physician. This review provides a clinically oriented analysis of coronary venous anatomy as it pertains to left ventricular lead implantation. We examine the embryological origins of the coronary sinus, explaining why variants such as persistent left superior vena cava and obstructive Thebesian and Vieussens valves occur. The gross anatomy section details coronary sinus dimensions, ostial localization, and tributary classification using attitudinally correct nomenclature. Microanatomical considerations include wall thickness gradients, subepicardial adipose tissue thickness, and myocardial sleeve distribution. We review imaging modalities-fluoroscopy, computed tomography, and magnetic resonance imaging-emphasizing their complementary roles in preprocedural planning and real-time guidance. Catheter and wire handling techniques are discussed, from cannulation strategies to lead delivery. Electrophysiological parameters including Q-LV and S-QRS intervals are examined in the context of anatomical lead positioning. Finally, we analyze how anatomy influences outcomes, complications, and non-response. Understanding the coronary venous system transforms cardiac resynchronization therapy from a technical procedure into an anatomically informed intervention where success depends on matching therapeutic goals with individual patient anatomy.
Grade 4 glioma is inherently lethal due to inevitable recurrence. Current radiotherapy guidelines recommend uniform target volume margins, disregarding molecular and clinical variables. We hypothesized that spatiotemporal recurrence may be predicted by molecular and phenotypic features present already at diagnosis. We analyzed 390 paired MRIs of grade 4 gliomas from Norwegian multi-centers and TCIA (2015-2025). Deep learning segmentation identified contrast-enhancing (CEcore) and non-enhancing (NE) tumor compartments and afflicted anatomical regions. We quantified primary CEcore/NE volume ratio, anatomical tumor trajectories, and spatiotemporal progression using Hausdorff-95-analyzed in relation to survival, MGMT, IDH, age, extent of resection, sex, and anatomical location, using machine learning and Cox regression. Primary tumor composition was independently prognostic: CEcore/NE volume ratio ≤0.324 predicted improved overall survival (adjusted HR = 0.56, 95% CI 0.37-0.84, p = 0.006), independent of age, MGMT-status, and individual compartments. Within IDH-wildtype, MGMT-unmethylated patients, low volume ratio ≤0.324 demonstrated a 4.3 months survival benefit (median 17.6 vs 13.3 months, p = 0.0209). Longer time to progression correlated with increased HD95 distances (p < 0.03). Tumors originating in occipital lobe had highest propensity to migrate to new sites (57.1%) and the shortest time to progression (adjusted HR = 1.90, p = 0.026). These findings support molecular-demographic-anatomical risk stratification that may inform personalized margin-determination in radiotherapy planning.
The posterolateral (PL) column is often the most difficult one to access and manage in tibial plateau fractures. Although multiple approaches have been described, many provide limited flexibility to address diverse fracture patterns. The Versatile Extended Anterolateral (EAL) approach with four progressive windows has been previously introduced within the Main Deformity Direction (MDD) framework. In this cadaveric study, we provide a detailed anatomical and technical description of this modified approach, characterized by an optimized skin incision and four stepwise deep windows for PL column management. The primary aim was to assess the technical feasibility of this approach. Potential reduction and fixation options associated with each window were also explored descriptively. Four fresh-frozen cadaveric specimens were used to evaluate surgical exposure and relationships with relevant anatomical structures. Based on the anatomical exposure achieved with each window, potential reduction and fixation strategies were explored descriptively. Representative clinical cases were included to illustrate fixation techniques. The skin incision allowed access to all windows, achieving safe progressive PL column exposure. The four windows (pre-fibular, supra-fibular, retro-fibular, trans-fibular) expanded reduction and fixation possibilities stepwise. The trans-fibular window enabled complete articular and metaphyseal exposure. The anterior tibial artery crossing, located at a mean of 3.5 cm from the tip of the fibular head, represented the distal anatomical limit of posterior exposure. The Versatile EAL approach is a feasible option for managing lateral and PL columns in tibial plateau fractures. Its progressive four-window strategy may allow tailored exposure and fixation according to fracture-specific requirements. IV.
Total knee arthroplasty (TKA) is associated with significant postoperative pain. Regional analgesic techniques targeting femoral nerve (FN) and adductor canal (AC) are commonly employed; however, uncertainty persists regarding optimal injection site within the AC. Cadaveric investigations provide anatomical insights into injectate spread patterns, but their interpretation requires caution when extrapolating to clinical practice. Twelve thighs from six embalmed human cadavers underwent ultrasound-guided injections of different-colored dyes, with equal volumes administered at three predefined locations: FN, proximal AC, and distal AC. Subsequent anatomical dissections were performed to assess nerve staining patterns. Injection sites and volumes were standardized to allow for comparative anatomical evaluation. Femoral-level injections stained the femoral nerve and its major branches. Proximal AC injections consistently stained both the saphenous nerve and nerve to vastus medialis, whereas distal canal injections predominantly stained the saphenous nerve with only variable distal obturator involvement. The vaso-adductor membrane stained consistently after proximal canal injection. No sciatic nerve staining was observed. Equal-volume dye injections at the FN, proximal AC, and distal AC demonstrate distinct patterns of neural staining. Proximal AC injections resulted in consistent staining of both SN and NVM, whereas distal injections predominantly involved SN. These findings provide anatomical relevance regarding adductor canal block techniques but do not imply clinical superiority of any technique. Clinical correlation is required before extrapolating these observations to clinical practice.
This study aimed to compare the percentage of local anesthetic spread into the obturator canal between the nerve stimulation-guided interadductor approach (NS-INTAD) and the ultrasound-guided interadductor approach (US-INTAD), and to evaluate anatomical factors influencing the effectiveness of the obturator nerve block (ONB). Eighty patients scheduled for transurethral resection of bladder tumor were randomized to receive either NS-INTAD (n = 40) or US-INTAD (n = 40). In the lithotomy position, a mixture of local anesthetic and contrast medium was injected via the interadductor approach from the medial thigh. Injectate spread into the obturator canal was assessed by fluoroscopy, and the success rate and injection volume were compared between groups. Enhancement in the obturator canal was observed in 75% of patients in the NS-INTAD group and 90% in the US-INTAD group. Although the percentage was higher in the US-INTAD group, the difference was not statistically significant (P = 0.139). The volume of local anesthetic used did not differ significantly between groups (P = 0.07). In patients where the posterior branch of the obturator nerve coursed between obturator externus muscle fascicles, local anesthetic propagation into the obturator canal appeared limited. The percentage of injectate spread into the obturator canal tended to be higher with US-INTAD than with NS-INTAD. Our findings suggest that the lithotomy-position approach facilitates more cephalad local anesthetic spread than the supine-position approach. Furthermore, the development of the superior fascicle of the external obturator muscle appears to be an anatomical factor that inhibits injectate spread into the obturator canal. UMIN Clinical Trials Registry (UMIN-CTR) UMIN000027762; registered 15 June 2017.
The aim of this study was to investigate hydrocephalus shunt insertion dynamics and variability across a group of clinicians using a polyvinyl alcohol (PVA)-Phytagel-based brain-mimicking model for ventricular catheter placement. The rate of catheter insertion through a simplified PVA-Phytagel-based brain mimic was assessed in a group of clinicians at an academic pediatric neurosurgery division. Lateral and longitudinal catheter position and motion throughout insertion depth were assessed and analyzed. Additionally, survey data were collected to better understand the intentionality and thought process underlying each surgeon's insertion approach. A total of 57 insertions were collected across 19 participants, of whom 16 were neurosurgeons and included in the final analysis. Based on survey results, the authors found that there was a positive correlation between the surgeons' attempt to speed up at earlier anatomical landmarks (ependyma) and slow down at deeper anatomical landmarks (deeper ventricular levels) throughout the depth of insertion. However, in practice, there was no correlation between the intended insertion rate and the actual insertion rate of the surgeons. Additionally, the authors found a high degree of variability in both the longitudinal and lateral directions for both position and velocity, independent of trial number, individual surgeon, or insertion depth. In the brain mimic of compliant uniform hydrogel, the surgeon's insertion rate throughout insertion and across samples was found to not be uniform, with catheter insertion rates being highly heterogeneous and nonrepeatable. Analogous fields within the neural implant space have regulated output for both insertion rates and dynamics in order to optimize implant outcome and minimize neuroinflammatory response. The authors believe this study sets a precedent to begin exploration of hydrocephalic tissue response in vivo to differential longitudinal and lateral catheter velocities.
Situs inversus totalis (SIT) is a rare congenital anatomical anomaly characterized by mirror-image reversal of thoracic and abdominal organs. Perioperative management becomes particularly challenging when SIT coexists with inferior vena cava (IVC) malformation (e.g., agenesis of the hepatic segment), decompensated cirrhosis, and respiratory insufficiency-a complex scenario for which relevant anesthesia experience is exceedingly limited. A 66-year-old female patient with SIT complicated by polysplenia syndrome and agenesis of the hepatic segment of the IVC presented for surgical management of intrahepatic bile duct stones, cholangitis, and obstructive jaundice. Preoperatively, she had Child-Pugh C cirrhosis and respiratory insufficiency. The patient underwent laparoscopic cholecystectomy, choledochoscopic-assisted intrahepatic bile duct stone extraction, and T-tube drainage. The anesthesia strategy comprised mirror-image ECG lead placement, transthoracic echocardiography (TTE) -guided preload assessment utilizing left ventricular end-diastolic area with subsequent optimization of fluid therapy, preferential selection of hepatically non-metabolized drugs, and employing lung-protective ventilation with permissive hypercapnia targeting PaCO₂ ≤ 65 mmHg. The procedure lasted 320 min with stable hemodynamics. The patient was transferred to the intensive care unit (ICU) with the endotracheal tube in situ and was successfully extubated 1 h after admission. She was discharged from the ICU on postoperative day 6, showing improvement in liver function and overall clinical status compared with her preoperative baseline. To our knowledge, this is the first description of anesthesia management for a patient with SIT complicated by Child-Pugh C cirrhosis and respiratory insufficiency, highlighting TTE as an alternative for preload monitoring in the presence of IVC anomalies. Anticipation of anatomical variations and implementation of an individualized anesthesia strategy are critical to ensuring perioperative safety.
Extraperitoneal access strategies have been successfully applied to left-sided colorectal resections, but their application to right-sided colectomy with complete mesocolic excision (CME) has not been described in accordance with the IDEAL framework. The EXPERTS (EXtraPEritoneal coloRecTal Surgery) with Complete Mesocolic Excision (CME) approach aims to utilise the superior oncological advantages of CME with the safety of extraperitoneal approach to central vascular structures. As part of IDEAL Stage 0 (preclinical development), a structured cadaveric Programme was undertaken to assess the anatomical feasibility, reproducibility and visualisation of critical structures during EXPERTS right hemicolectomy with CME. A total of 11 cadavers were dissected, including nine preserved using the Thiel method and two using fresh frozen preservation. All dissections were video recorded and analysed. In nine out of 11 cadavers (9 Thiel-preserved and 2 fresh frozen cadavers), consistent identification of key anatomical structures-including the right kidney, duodenum, superior mesenteric artery (SMA), superior mesenteric vein (SMV) and their major branches-was achieved via an extraperitoneal approach. Fresh frozen cadavers demonstrated more variable tissue quality and plane definition. High-quality operative videos were obtained for further analysis. This IDEAL Stage 0 study demonstrates the anatomical feasibility of an extraperitoneal approach to right hemicolectomy with CME and supports progression to assess its safety and effectiveness in a clinical setting.
Perceptually rich visualizations are known to pose cognitive demands. Cognitive load theory suggests that design choices can mitigate negative effects on learning performance. We aim to evaluate whether cueing is able to reduce the cognitive demands of a parallax effect induced by constrained interactivity in learning with perceptually rich visualizations. 191 university students were included in the final sample. Participants were randomly assigned to learn with a perceptually rich anatomical visualization in one of four versions resulting from the factors constrained interactivity (static vs. interactive parallax effect) and visual boundary cues (without vs. with cues). The constrained interactivity allowed participants to slightly rotate an anatomical model. Subjective cognitive load during learning was assessed using a survey. Participants completed image-based and text-based retention tests. Contrary to the notion that cues help learners facing perceptually rich task, we found that cues only benefit learning with the static visualization. As indicated by an interaction effect, cueing lowered retention performance in the constrained interactive condition. Germane cognitive load ratings suggest that participants recognized the cues as helpful in the interactive parallax effect condition while preferring the variant without cues in the static condition. The cognitive demands of minimally interactive perceptually rich visualizations cannot simply be reduced by inserting boundary cues. Instead, visual cues may even add to the perceptual richness of visualizations if they feature subtle forms of interactivity. The perceptual demands of a slightly adjustable visualization that was extended by cues may have been too high.
Nasal obstruction is a common complaint in otolaryngology. Choosing between septoplasty or septorhinoplasty may be challenging due to anatomical variability. Computational fluid dynamics (CFD) has emerged as a complementary, patient-specific tool enabling bilateral simulation and virtual planning, allowing simulation of post-surgical outcomes under physiological conditions. Virtual surgeries were performed on patients with nasal obstruction using Flowgy® software. For each patient, three anatomical models were simulated: baseline, virtual septoplasty, and virtual septorhinoplasty (with turbinoplasty when indicated). CFD analysis was used to assess pressure drop (ΔP), flow symmetry (Φ), and nasal resistance (R), among other airflow parameters. Comparative data were analysed to determine the functional impact of each intervention. In all five cases, septoplasty resulted in improvements in pressure drop ranging from 10.5% to 50.9%, as well as changes in resistance ranging from +2.98% to -64.3%. Septorhinoplasty provided additional functional gains in pressure (up to 44.8%) and resistance (up to 51%) in selected cases. Based on the analysis, surgical benefit thresholds were proposed: ≥30% improvement in ΔP or R from baseline to justify septoplasty, and ≥20-30% additional benefit over septoplasty to indicate septorhinoplasty. CFD analysis is a valuable adjunct in functional nasal surgery, enabling personalised decision-making. While not a substitute for clinical judgement, CFD-guided planning may enhance the selection of surgical strategy and reduce unnecessary interventions.
Chest diseases remain a major cause of global morbidity and mortality, and accurate detection from chest X-ray images is critical for early diagnosis and clinical decision-making. However, large variations in lesion scale, morphology, and spatial distribution pose significant challenges for automated detection systems, particularly in identifying small lesions and achieving precise localization. To address these issues, we propose a multi-scale chest lesion detection method based on adaptive collaborative feature fusion and shape-aware optimization. The method enhances multi-scale feature modeling and introduces shape structural constraints to improve detection accuracy and localization robustness in complex anatomical environments. Specifically, a Context-Embedded Feature Enhancement Network is designed to jointly capture global anatomical context and local lesion characteristics, strengthening lesion representation. An Adaptive Feature Focusing Network further improves multi-scale feature representation through adaptive spatial feature aggregation, enabling more effective detection of small lesions. In addition, a shape-aware optimization strategy integrating normalized Wasserstein distance with shape-weighted constraints improves localization stability and bounding box regression accuracy for irregular lesions. Compared with state-of-the-art methods, the proposed method achieves improvements of 6.3% in mAP, 6.4% in small-lesion mAP, and 8.6% in mean recall on VinDr-CXR, as well as improvements of 4.2% in mAP and 7.8% in mean recall on ChestX-ray8, demonstrating its effectiveness and generalization capability for multi-scale chest lesion detection.
This study presents a radiographic repeat analysis conducted in public hospitals in the Souss-Massa region of Morocco. It aimed to assess repeat rates, identify the most frequent causes of image repetition and the most affected anatomical regions, and examine variations in recurrence rates. A cross-sectional study was conducted during March and April 2025 in four public hospitals in the Souss-Massa region of Morocco. The number of acquired and repeated images was recorded using a standardized observation grid, supplemented by repeated image extraction for retrospective analysis. Descriptive statistics and chi-square tests were used to assess repeat rates, associated causes, anatomical regions involved, and variations by hospital and day of the week. Of the 15,493 radiographic images analyzed, 1063 were repeated, resulting in an overall repeat rate of 6.86%, which falls within the acceptable range recommended by the American Association of Physicists in Medicine (AAPM). The highest specific repeat rate was observed at Hospital 2 (11.94%), followed by Hospital 1 (4.63%), Hospital 4 (4.33%), and Hospital 3 (3.92%). The most frequently repeated examinations were chest, pelvis, abdomen, lumbar spine, and shoulder. Across all examination types, the main causes included positioning errors, metallic artifacts, collimation errors, and under- or over-exposure. Daily variability in repeat rates was also observed. Although the repeat rate was within acceptable limits, indicating reasonably adequate radiographic image production, variations between hospitals and the predominance of preventable professional errors highlight the need for improved technical consistency. Implementing standardized protocols, extending repeat-rate assessment and regular monitoring, establishing a reference threshold, improving understanding of human and organizational factors, and harmonizing image acceptance criteria would help reduce repetition and strengthen consistency in assessments across professionals.
To report the incidence and spectrum of vitreoretinal complications (VRCs) in consecutive patients implanted with 2 designs of type 1 keratoprosthesis (KPro). Retrospective analysis of electronic medical records. Demographic and clinical data were collected from patients who developed VRCs. The main outcome measure was the proportion of VRCs and their management. Eighty-four eyes from 81 patients with KPro were analyzed (Boston = 28; Lucia = 56). VRCs occurred in 32 cases (38.10%). The mean age of patients with VRCs was 55.15 ± 16.67 years (range: 27-90), and the mean follow-up duration was 33.90 ± 16.31 months. The most frequent VRC was retroprosthetic membrane formation requiring pars plana vitrectomy (n = 10, 11.90%), followed by retinal detachment (n = 9, 10.71%), choroidal detachment (n = 8, 9.52%), endophthalmitis (n = 8, 9.52%), vitreous hemorrhage (n = 7, 8.33%), cystoid macular edema (n = 2, 2.38%), asteroid hyalosis (n = 1, 1.19%), epiretinal membrane (n = 1, 1.19%), macular atrophy (n = 1, 1.19%), and sterile vitritis (n = 1, 1.19%). The median best-corrected visual acuity (BCVA) achieved after KPro implantation was 0.6 LogMAR (IQR: 0.18-1.3); in contrast, the median BCVA after any VRC was 2.4 LogMAR (IQR: 0.58-2.7). In all cases (100%), VRCs led to a deterioration in BCVA. Anatomical failure occurred in 7 cases (21.88%) and functional failure in 20 cases (62.50%), after a mean duration of 8.83 months. VRCs occurred in 38.10% of cases after type 1 KPro implantation and were associated with poorer outcomes, including increased anatomical and functional failure.
Breast cancer is the most common malignancy in women worldwide, and bone-targeting agents (BTAs) - particularly zoledronic acid and denosumab - are widely administered for bone metastasis, carrying a recognized risk of medication-related osteonecrosis of the jaw (MRONJ) with reported incidence rates of 1-15% in oncologic populations. Drug-dependent differences in MRONJ characteristics, anatomical distribution, and the causal role of dentoalveolar procedures remain incompletely characterized in breast cancer patients. This single-center retrospective cohort study included 101 breast cancer patients diagnosed with MRONJ per 2022 AAOMS criteria, who received zoledronic acid (n = 66) or denosumab (n = 35) exclusively at this institution between January 2009 and December 2025. All diagnoses were confirmed by a single board-certified oral and maxillofacial surgeon. Between-group comparisons were performed using the Mann-Whitney U, chi-square, and Fisher's exact tests as appropriate. Cumulative BTA dose count at MRONJ onset did not differ significantly between groups (mean 38.9 vs. 25.7; p = 0.076). The denosumab group developed MRONJ at an older mean age than the zoledronic acid group (59.5 vs. 55.7 years; p = 0.013). A significant drug-dependent anatomical difference was observed: mandibular predominance in the zoledronic acid group (59.1%) versus maxillary predominance in the denosumab group (54.3%; p = 0.015). The majority of patients developed MRONJ without prior dentoalveolar procedures (53.0% zoledronic acid; 65.7% denosumab). Periodontal disease prevalence was comparable between groups (60.5%; p = 1.000). Cumulative injection count alone does not reliably predict MRONJ onset in oncologic populations. The majority of cases arose without prior dentoalveolar procedures, challenging the paradigm of tooth extraction as the primary causative trigger. A drug-dependent MRONJ predilection in jaws- mandibular in zoledronic acid users, maxillary in denosumab users - warrants validation in larger prospective cohorts.
Limited-angle cone-beam computed tomography (LA-CBCT) enables rapid imaging and reduced radiation exposure, but its severely incomplete projection data lead to ill-posed reconstructions with prominent artifacts, limiting clinical applicability. Recent advances in 3D Gaussian Splatting (3D-GS) have shown promise for efficient tomographic reconstruction, yet its performance remains highly sensitive to initialization. In this work, we present SPARK (Structurally-Informed Projection-Accelerated Reconstruction), a two-stage framework that introduces a generative, structurally informed initialization for 3D-GS. In the first stage, a geometry-conditioned network directly predicts complete 3D Gaussian parameters from a sparse subset of projections, embedding learned anatomical priors to mitigate artifact propagation. In the second stage, the generated scene is refined through physics-based 3D-GS optimization, yielding high-fidelity reconstructions consistent with measured projections. Extensive experiments on public datasets demonstrate that SPARK substantially improves both image quality and convergence speed, achieving superior PSNR/SSIM in severely limited-angle scenarios compared with analytical, iterative, and deep learning baselines. Moreover, SPARK reconstructions provide enhanced inputs for downstream post-processing networks, further boosting image fidelity. These results suggest that SPARK is a promising prior-informed 3D-GS framework for simulated LA-CBCT reconstruction under limited angular coverage, providing an effective bridge between data-driven anatomical priors and physics-based projection-domain refinement.
Intraoperative functional brain mapping is an essential and intricate technique in modern-day glioma surgery. This article is not a review of the literature but of the technical protocol at our institution that has evolved over the recent decades to the current time and is intended to highlight details that enable us to perform maximal safe resection of gliomas. Prior to surgery, anatomical and functional imaging protocols are obtained to determine the tumor to be resected within its anatomical and functional environment. Preoperative assessments are used to determine which mapping procedures and tasks are most appropriate. Cortical and subcortical motor and language mapping using low and high frequency stimulation paradigms are applied when appropriate during resection. Methods to interpret findings and troubleshoot issues are reviewed herein. All preoperative imaging including magnetic resonance imaging, magnetoencephalography of functional cortex, and diffusion tensor imaging of subcortical tracts are uploaded into the neuronavigation station and used throughout surgery for guidance. The decision to continue with tumor resection is based on constant feedback from the mapping paradigms as functional pathways are approached in real time. Both awake and asleep anesthesia regimens are utilized depending on the type of testing required to assess and preserve functional areas during tumor resection. Postoperatively, deficits are assessed using MRI along with clinical exam to predict whether they will be temporary or permanent. The standard of care for all gliomas is maximal safe resection. In this review, we describe brain mapping methods that have been developed, refined, and utilized over decades at a single institution, which have allowed us to achieve this goal safely.