Volumetric-modulated arc therapy (VMAT) represents a standard of care for head and neck cancer (HNC). However, treatment plan quality depends heavily on the planner's expertise, thereby challenging the standardization of treatment quality. While the clinical application of artificial intelligence (AI) in radiotherapy planning is rapidly increasing, the robustness of these models-for both typical and irregular cases-remains insufficiently understood.

To evaluate the clinical utility and generalizability of an artificial intelligence (AI)-assisted volumetric-modulated arc therapy (VMAT) planning workflow for head and neck cancer (HNC). This study focused on the robustness of AI-assisted planning using dose distribution prediction for various cases including complex anatomical variations, such as bulky tumors and irregular tumor ruptures, and explored the feasibility of quality standardization through a "human-in-the-loop" intervention approach via re-optimization.

Twelve VMAT plans from 11 HNC patients, including complex anatomical scenarios with bulky masses and post-rupture anatomical changes, were analyzed. Clinical plans developed by an experienced planner were compared with AI-assisted plans generated using RatoGuide (AiRato Inc.). The AI-assisted planning workflow utilized deep learning-based dose distribution prediction to generate dose-derived ring contours for optimization. A human-in-the-loop intervention was implemented, wherein planners performed manual fine-tuning if initial AI-assisted plans violated the institutional dose constraint of Dmax <115%. Dosimetric metrics for the planning target volume (PTV) and organs at risk (OARs), along with Paddick conformity index and monitor units (MU), were evaluated.

AI-assisted plans achieved PTV coverage comparable to clinical plans while improving OAR sparing, particularly for the spinal cord and brainstem. The mean spinal cord PRV Dmax was reduced from 27.50 Gy in clinical plans to 23.97 Gy in AI-assisted plans. Even in cases with bulky or ruptured tumors, AI-assisted planning maintained high conformity and the overall mean Paddick conformity index was comparable to clinical plans. Although initial AI plans exhibited hotspots (Dmax > 115%) in three cases near the body surface, manual fine-tuning of optimization successfully suppressed these hotspots, meeting clinical criteria. AI-assisted planning maintained comparable mean MU values, and reduced the total planning time from approximately 2 h to 30 min.

The AI-assisted planning workflow demonstrated the potential to produce high-quality VMAT plans, even for complex HNC cases. By integrating physicist human-in-the-loop intervention with AI efficiency, this approach could facilitate the standardization of radiotherapy treatment planning quality while supporting patient safety.

Radiotherapy for vulvar carcinoma often employs the "frog-leg" position to minimize skin toxicity, but this position is biomechanically unstable and prone to setup errors using conventional skin markings.

To evaluate the clinical efficacy of Surface-Guided Radiotherapy (SGRT) for setup accuracy in vulvar cancer patients treated in the frog-leg position and to quantify the biomechanical correlation between Body Mass Index (BMI) and setup errors.

A retrospective cohort study of 20 vulvar cancer patients treated in the frog-leg position was conducted. All patients were treated using a uRT-linac 506c accelerator with integrated helical Fan-Beam CT (FBCT). Daily image guidance was performed using a two-step registration protocol (bony anatomy followed by soft-tissue fine-tuning). Setup errors from 500 FBCT datasets were compared between the conventional laser group (n = 10) and SGRT group (n = 10). The correlation between BMI and setup errors was analyzed using Spearman's rank correlation.

The SGRT group demonstrated significantly superior setup accuracy across multiple key dimensions (p < 0.001). Notably, the Pitch error was reduced from 1.21° ± 0.48° in the Conventional Group to 0.07° ± 0.32° in the SGRT Group (Cohen's d = 2.79). Longitudinal (LNG) error decreased from 0.49 ± 0.26 cm to 0.04 ± 0.24 cm (d = 1.82), and vertical (VRT) errors were similarly reduced.

In the Conventional Group, BMI showed a strong significant correlation with Pitch and Yaw errors (ρ = -0.81, P = 0.005), indicating that obesity may exacerbate pelvic rotational instability. Conversely, in the SGRT Group, no statistically significant correlation was observed between BMI and setup errors in any dimension (p > 0.05), demonstrating the system's robustness against body habitus variations.

The frog-leg position is characterized by significant biomechanical instability due to distinct anatomical factors. Traditional skin marking methods are susceptible to the "skin traction effect" and interference from BMI. SGRT, through real-time topographical matching, effectively corrects pelvic tilt and longitudinal sliding. Crucially, it eliminates the setup uncertainty associated with BMI, providing a solid evidence-based foundation for precision radiotherapy and individualized PTV margin reduction.

Intraventricular meningiomas (IVMs) are rare adult tumors in deep-seated ventricular locations with challenging surgical anatomy and heterogeneous management. Contemporary pooled benchmarks for extent of resection, complications, and recurrence in adult intraventricular meningiomas remain limited.

A systematic review and meta-analysis was performed following PRISMA 2020 and MOOSE guidelines. Studies published from 2019 to 2025 including adults with histologically confirmed IVMs were analyzed. Data on demographics, tumor location, histopathology, surgical management, and outcomes were extracted. Random-effects models pooled means and proportions; heterogeneity was assessed with I².

Thirty-four studies comprising 981 patients were included. Most tumors involved the lateral ventricles (83%), with a mean age of 48 years (95% CI, 45-51) and 63% female predominance. The pooled gross total resection (GTR) rate was 88% (95% CI, 82-93; I² = 52%), and the recurrence rate-in studies with ≥ 12 months follow-up-was 9% (95% CI, 4-19; I² = 0). Postoperative complications were reported descriptively across studies and consisted mainly of transient visual or motor deficits, hemorrhage, edema, and hydrocephalus requiring CSF diversion. Stereotactic radiosurgery achieved durable control for small residual or recurrent lesions. Among 297 tumors with available grading data, CNS WHO grade 1 tumors accounted for 74.1% and CNS WHO grade 2 tumors for 24.2%.

Adult IVMs are predominantly lateral-ventricle tumors affecting middle-aged women and are associated with generally favorable outcomes in specialized centers. Gross total resection remains the preferred goal when functionally feasible, whereas planned subtotal resection followed by early stereotactic radiosurgery may offer durable control in anatomically high-risk cases. These pooled benchmarks help contextualize current practice and highlight the need for improved molecular and neurofunctional reporting in future prospective studies.

Robotic-assisted surgical platforms are now routinely utilized in numerous surgical specialties. In breast surgery, robotic-assisted procedures are being actively explored and applied. However, there are currently few published studies specifically addressing robotic breast-conserving surgery (R-BCS). The surgical and oncologic outcomes, patient-reported aesthetic results and learning curve, of R-BCS for breast cancer were evaluated in this study. This study retrospectively analyzed data from all patients who received R-BCS at a single medical center during the period from March 2024 to December 2025. The preliminary results were reported. A total of 88 individuals participated in this study and underwent robotic-assisted R-BCS. The mean age of patients was 53.6 years. Notably, all surgical interventions were successfully performed robotically, with no cases requiring intraoperative conversion to alternative approaches. The mean operative duration was 124.40 ± 36.79 min. Learning curve analysis demonstrated a marked decrease in operative time after the first 12 consecutive cases. The overall complication rate was 10.2%. The positive surgical margin rate was 1.1%. In the Breast Q score results, patients with tumors smaller than 2 cm or tumors located in the upper quadrant had higher postoperative satisfaction with breasts and sexual well-being. Compared with preoperative scores, patients had higher postoperative psychosocial well-being, but their physical well-being: chest scores were lower than preoperative levels. The follow-up duration was 14.43 ± 4.65 months. At the time of the most recent follow-up, none of the patients developed local recurrence, distant metastasis, or died from any cause. This study confirmed the short-term safety and feasibility of R-BCS. Preliminary findings suggest that this surgical approach demonstrated the potential to achieve a low incidence of complications, a low positive margin rate and favorable aesthetic outcomes.

Taxane-induced peripheral neuropathy (TIPN) remains a frequent, dose-limiting toxicity in breast cancer, and reliable pre-treatment predictors are lacking. Nailfold capillaroscopy (NFC) offers a non-invasive window into systemic microvascular morphology, which may reflect vascular susceptibility relevant to neuropathy risk. The objective of the study was to evaluate whether baseline NFC morphology is associated with taxane-related neuropathy severity and symptom burden, and to explore ROC-derived thresholds of NFC parameters for identifying clinically significant neuropathy.

In this pilot cohort (N = 48) of breast cancer patients receiving taxane-based regimens (docetaxel [DOC], paclitaxel [PAC], or carboplatin-paclitaxel [CPAC]), baseline NFC parameters-capillary density (CD), capillary width (CW), capillary loop diameter (CLD), and capillary length (CL)-were recorded. Neuropathy outcomes were assessed longitudinally using NCI-CTCAE score and EORTC QLQ-CIPN20. Regimen-stratified associations were examined using Spearman's correlation. ROC analyses were conducted in the overall cohort using NCI ≥ 2 vs 0-1 as the binary endpoint; optimal cut-offs were defined by Youden's index.

Regimen-specific patterns were observed. In CPAC (N = 13), CD correlated with NCI (ρ = 0.567, p = 0.043) and CIPN20 (ρ = 0.608, p = 0.028). In PAC (N = 16), no significant correlations were found between NFC parameters and neuropathy outcomes (all p > 0.05), although CIPN20 and NCI remained strongly correlated (ρ = 0.778, p < 0.001) and CLD-NCI showed a trend (ρ = 0.486, p = 0.056). In DOC (N = 19), CL correlated with NCI (ρ = 0.509, p = 0.026) and CIPN20 (ρ = 0.484, p = 0.036), while CW was inversely correlated with CIPN20 (ρ =  - 0.510, p = 0.026). In overall ROC analyses, optimal thresholds for predicting NCI ≥ 2 were CD ≥ 8.5 capillaries/mm (sensitivity 77.3%, specificity 42.3%), CW ≥ 31.0 (90.9%, 26.9%), CLD ≥ 18.45 (90.9%, 42.3; highest Youden index 0.332), and CL ≥ 160.1 (72.7%, 46.2).

Baseline NFC morphology demonstrated regimen-dependent associations with taxane-related neuropathy outcomes and yielded exploratory ROC thresholds for identifying clinically significant neuropathy (NCI ≥ 2). CLD showed the strongest overall discrimination, while subgroup analyses highlighted distinct candidate markers (CD in CPAC; CL in DOC). These findings are hypothesis-generating and warrant validation in larger, independent cohorts.

Stereotactic Centralized Ablative Radiation Therapy (SCART) is a spatially fractionated radiotherapy strategy designed for large or bulky tumors by delivering an intensified ablative dose to a centrally located subvolume while allowing controlled dose reduction toward the tumor periphery. However, clinical implementation of SCART critically depends on reproducible definition of the stereotactic treatment volume (STV), which currently lacks a standardized geometric framework.

To develop and preliminarily evaluate a geometry-based scaling model for STV definition based on internal dose fall-off characteristics under stereotactic volumetric modulated arc therapy (VMAT) delivery.

A phantom-based stereotactic VMAT planning study was performed using idealized cylindrical STVs with radii of 1-3 cm and axial lengths of 3-10 cm. VMAT plans were generated using prescription doses of 15-24 Gy per fraction over three fractions. Internal dose fall-off was quantified using the equivalent radius (r₁₅) of the 15 Gy isodose surface, defined as the radius of a volume-equivalent cylindrical isodose distribution. The predicted STV radius (d) was subsequently modeled from outer reference target geometry using dose-specific linear regression analysis.

The extracted 15 Gy isodose surface demonstrated reproducible geometry-associated attenuation behavior across the investigated stereotactic VMAT configurations. The equivalent radius (r₁₅) increased monotonically with STV size, ranging from approximately 2.0-2.1 cm to 4.1-4.3 cm at 15 Gy × 3 and from approximately 3.0-3.3 cm to 6.3-6.6 cm at 24 Gy × 3 as STV radius increased from 1 to 3 cm. Variation associated with axial target length generally remained within approximately 0.2-0.4 cm for a given prescription level. Increasing arc number produced only modest effects on dose attenuation behavior, with a median inter-configuration difference in r₁₅ of approximately 0.12 cm. Dose-specific regression models demonstrated excellent agreement with phantom-derived measurements, with R² values ranging from 0.991 to 0.999.

This study establishes a geometry-based STV scaling framework for SCART based on reproducible internal dose fall-off characteristics under stereotactic VMAT delivery. The proposed dose-specific regression model provides a practical and physically interpretable approach for inward STV scaling from measurable gross tumor volume geometry and may support more reproducible implementation of SCART planning.

Vulvar cancer is a rare malignancy, accounting for 5%-8% of gynecological cancers, most commonly vulvar squamous cell carcinoma (VSCC). Achieving adequate tumor-free margins is crucial for local control. Although current guidelines recommend margins of at least 8 mm, the optimal cut-off remains uncertain, as published evidence is inconsistent. Other clinicopathological factors may also influence recurrence, but data are limited given the rarity of VSCC. This meta-analysis aimed to evaluate the effect of margin width and additional prognostic factors on recurrence risk.

We conducted a PRISMA-compliant systematic review and meta-analysis, registered in PROSPERO (CRD420251135659). PubMed, Scopus, and Cochrane were searched through September 2025 for surgical VSCC studies reporting tumor-free margin width and local recurrence. Pooled risk ratio (RR) and hazard ratio (HR) were estimated using random-effects models, with subgroup analyses and risk of bias assessed using ROBINS-I.

Twenty-eight cohort studies were included. Margins < 8 mm were associated with increased local recurrence (LR) (RR 1.42; 24.8% vs. 15.6% for ≥ 8 mm). No significant difference was found between 3-8 mm and < 3 mm, though recurrence was higher with < 3 mm. Margin width was not associated with regional or distant metastasis. Among patients with < 8 mm margins, adjuvant treatment were radiotherapy (22.5%) and re-excision (12.5%). Predictors of LR included tumor size (HR 1.53), lateral site (HR 1.60), depth of invasion (HR 2.51), lymph node positivity (HR 2.52), and higher FIGO stage (HR 1.77).

Tumor-free margins ≥ 8 mm are associated with a lower risk of local recurrence in VSCC. However, recurrence remains multifactorial, influenced by tumor size, depth of invasion, nodal involvement, and stage. Therefore, optimal management should combine adequate surgical margins with individualized adjuvant therapy tailored to the patient's risk profile.

PROSPERO (CRD420251135659).

Infectious complications in patients with cancer are contributors to hospitalisations, treatment disruption, and mortality. This study aimed to develop machine learning (ML) models for risk stratification of infection-related hospitalisations (IRHs) among patients with cancer.

Adult patients diagnosed with and treated for lymphoma, multiple myeloma (MM), chronic lymphocytic leukaemia (CLL), colorectal, or lung cancer from 2013 to 2023 in the North Denmark Region were included. Data from national cancer registries and patient health records were used. Serious infections were defined as sepsis, positive blood culture, intensive care admission, or death during hospitalisation. Models were developed using multiple ML methodologies across three different data setups: ML-using data up to hospital admission; ML₄₈-using data up to 48 h after admission; and ML_48,reduced-a pruned version of ML₄₈ with fewer features. Data were split into training, validation and held-out test sets.

Among 9874 patients 2661 IRHs occurred, including 691 (21.5%) serious infections. On the held-out test set, ML achieved the best performance (ROC-AUC: 0.79) versus ML₄₈ (0.76), with ML₄₈ having higher specificity at matched sensitivity (50.9% vs 44.9% for sensitivity ~85%). ML_48,reduced reached a ROC-AUC of 0.75 (specificity 46.5% at sensitivity ~85%). Subgroup analysis showed the best performance for lymphoma (ROC-AUC: 0.86) and lowest for lung cancer (ROC-AUC: 0.73). The model performed better in patients with ECOG score 0-1 compared to ≥ 2 (ROC-AUC 0.88 vs. 0.77). SHAP analysis identified previous blood cultures, carbamide, and neutrophil counts as key predictive features across all models.

The ML models showed moderate performance in risk stratification of IRHs in patients with cancer. Such models could provide decision support in hospitalisation decisions and discharge.

To review evidence and practical considerations for integrating fluorescence imaging into robotic head and neck surgery, emphasizing technical parameters, optical limitations, and clinical utility. We additionally incorporate insights derived from our institutional experience to contextualize real-world implementation challenges. Transoral robotic surgery (TORS) has expanded minimally invasive treatment for head and neck squamous cell carcinoma, improving functional outcomes and reducing morbidity. However, absent tactile feedback and limited depth perception remain challenges, particularly during margin assessment. Fluorescence-guided surgery (FGS) has emerged as a potential solution for real-time tumor visualization. This narrative review synthesizes literature on nonspecific dyes (indocyanine green) and tumor-targeted agents (panitumumab-IRDye800CW), evaluating compatibility with the da Vinci Firefly system. Institutional intraoperative observations complement existing data and highlight workflow considerations. Fluorescence imaging improves intraoperative tumor delineation and deep margin assessment. Critical technical variables include imaging mode selection, scope angle, illumination characteristics, field uniformity, and autoscaling behavior. Tumor-targeted tracers demonstrate superior contrast in Sensitive Firefly mode but require standardized protocols for optimal performance. Fluorescence-guided robotic surgery can improve oncologic precision in TORS by compensating for inherent sensory limitations. Consistent clinical benefit requires further development of targeted agents, workflow standardization, and robust validation studies.

Splenosis is defined as the heterotopic implantation of functional splenic tissue in extra-splenic compartments, usually secondary to traumatic or iatrogenic splenic rupture (1). Although it is most commonly found in the abdominal and pelvic cavities, implants have been reported in less common sites such as the thorax, liver, and subcutaneous tissue. Thoracic splenosis represents a diagnostic challenge, as it is usually an incidental finding, rarely symptomatic, and, from an imaging standpoint, can mimic malignant entities such as primary neoplasms or metastatic disease. We present two cases of splenosis diagnosed using nuclear medicine techniques, including the case of a 65-year-old woman with a history of gunshot wound, who required laparotomy, splenectomy, and left nephrectomy, in whom benign pleural and abdominal lesions were identified. Initial metabolic characterization using PET/CT and subsequent confirmation with 99mTc-HDRBC (Heat-Damaged Red Blood Cells) scintigraphy allowed us to establish the definitive diagnosis, avoiding unnecessary invasive procedures. Additionally, we describe a 58-year-old male with a history of blunt chest trauma and prostate cancer, in whom 18F-PSMA PET/CT suggested malignancy, but 99mTc-HDRBC scintigraphy confirmed multi-compartmental splenosis. This report highlights the importance of considering splenosis in the differential diagnosis of thoracic and abdominal nodular lesions in patients with a history of splenic trauma, as well as the value of functional imaging techniques in its accurate identification.