Evaluation of condyle histogram data based on magnetic resonance imaging, condylar morphology, and disc position in patients with jaw deformities.

Purpose To develop a multiparametric MRI-based radiomics model and deep learning-radiomics (DLR) fusion model for preoperative prediction of lymph node metastasis (LNM) in early-stage cervical cancer. Materials and Methods In this multicenter retrospective study (January 2020-December 2022), preoperative MRI data from patients with early-stage cervical cancer were split into training, internal testing, and external testing cohorts. Radiomic and deep learning (DL) features of both the tumor and lymph node were extracted separately from the MRI scans. Multivariable logistic regression was used to construct predictive models for LNM based on tumor and lymph node radiomic features (Rad_T+LN) and based on radiomic and DL features from both the tumor and lymph node (DLR_T+LN). The models' effectiveness and clinical applicability were evaluated using receiver operating characteristic curves, calibration curves, and decision curve analysis. A two-tailed P value of <.05 was considered statistically significant. Results The overall dataset included 862 patients (median age, 53 years [IQR, 45-60 years]). Rad_T+LN resulted in areas under the receiver operating characteristic curve (AUCs) of 0.81 (95% CI: 0.76, 0.86), 0.79 (95% CI: 0.72, 0.87), and 0.77 (95% CI: 0.71, 0.82) in the training, internal testing, and external testing cohorts, respectively. DLR_T+LN achieved AUCs of 0.83 (95% CI: 0.76, 0.91) and 0.79 (95% CI: 0.74, 0.84) in the internal and external testing cohorts, respectively, and did not improve over Rad_T+LN (P > .05). Both models demonstrated good calibration and positive net benefit on decision curve analysis. Conclusion Rad_T+LN and DLR_T+LN exhibited robust diagnostic performance for LNM prediction. Keywords: MR-Diffusion Weighted Imaging, MR Imaging, Genital/Reproductive, Cervix, Metastases, Decision Analysis, Segmentation, Radiomics, Diagnosis, Uterine Cervical Neoplasms, Lymphatic Metastasis, Magnetic Resonance Imaging, Deep Learning Supplemental material is available for this article. © RSNA, 2026.

Associations Between Preoperative Diagnostic Tests and Surgical Recommendation for Carpal Tunnel Syndrome.

With multiple treatment options available for hemorrhoidal disease (HD), identifying factors that influence surgical and perioperative management decisions is essential, particularly in advanced cases. This study aimed to determine the patient and disease-related determinants affecting the choice of surgical technique and perioperative management in patients with Grade III and IV HD, thereby addressing inter-institutional variations in treatment approaches. A secondary analysis was performed on data from a nationwide, multicenter prospective cohort study. The study included 315 patients diagnosed with Grade III (72%) and Grade IV (28%) HD, with a mean age of 43.7 ± 11.4 years and a male predominance (76.7%). Preoperative data, including patient demographics, comorbidities (ASA scores), symptom severity, and clinical findings, were collected at participating governmental and private hospitals. Surgical techniques were classified as excisional or non-excisional, while anesthesia type, use of perianal or pudendal analgesia, and hospitalization duration were determined by the surgical teams. Hospital type was also recorded. Multivariable analyses were conducted to identify factors influencing the choice of surgical techniques, anesthesia, analgesia application, and the decision for outpatient procedures. Multivariate analysis revealed that the presence of thrombosis significantly influenced the choice of surgical technique (OR: 7.2, CI: 2.8-12.7, p = 0.001), while hospital category also played an important role (OR: 5.1, CI: 2.7-9.7, p = 0.001). For anesthesia type, factors such as disease grade (OR: 3.3, CI: 1.6-6.7, p = 0.001), hospital category (OR: 9, CI: 4.1-19.9, p < 0.001), and surgical technique (OR: 6.8, CI: 3-15.3, p < 0.001) were significant determinants. The decision to use perianal or pudendal analgesia was influenced by hospital category (OR: 27.1, CI: 11.7-62.6, p < 0.001) and the presence of incontinence (OR: 0.2, CI: 0.04-0.93, p = 0.04). Outpatient management was associated with disease grade (OR: 2.3, CI: 1.1-4.8, p = 0.023), hospital category (OR: 2, CI: 1.2-3.2, p = 0.011), higher comorbidity (ASA ≥ 3, OR: 3.3, CI: 1.8-3.2, p = 0.007), and the selected surgical technique (OR: 3.1, CI: 1.6-5.8, p = 0.001). Significant inter-institutional variations exist in the management of advanced HD. Among various factors, the presence of thrombosis emerges as the predominant determinant in surgical decision-making, providing valuable insights for standardizing treatment protocols and reducing practice variability.

Facial nerve (FN) palsy is a significant complication of microsurgical resection of vestibular schwannoma (VS) and can profoundly impact patient quality of life. Patients exhibit two distinct trajectories of FN dysfunction: immediate facial nerve palsy (IFNP), occurring early in the postoperative period, and delayed facial nerve palsy (DFNP), characterized by new-onset weakness after an initially normal postoperative examination. In this study, we aimed to identify preoperative, intraoperative, and postoperative predictors of facial nerve trajectory to inform patient counseling and improve long-term functional outcomes. Electronic health records at our institution were reviewed to identify patients who underwent microsurgical resection of VS between July 2016 and April 2024. Preoperative data included demographic characteristics, tumor features, and baseline facial nerve function, classified according to the House-Brackmann (HB) grading scale. Univariable and multivariable logistic regression analyses were performed to identify independent predictors of IFNP and DFNP. Cox regression analysis was used to evaluate factors associated with intrinsic recovery among patients with IFNP. Among 433 patients who underwent vestibular schwannoma resection, 36.3% developed IFNP and 10.9% DFNP. DFNP was typically transient, with 90% of patients (27/30) achieving good recovery within two months. In contrast, only 35% of IFNP patients (N = 55) experienced intrinsic recovery, and approximately 60% of those without recovery (N = 66) required facial reanimation surgery. On multivariable analysis, larger tumor size (> 2cm) independently predicted IFNP, while younger age (< 33 years) independently predicted DFNP. In this study, we found that larger tumors increased the risk of immediate FN palsy, while younger age was associated with delayed onset palsy. High HB scores at discharge predicted slower or incomplete recovery in immediate cases, whereas nearly all delayed cases recovered without intervention. Systematically characterizing these postoperative trajectories can enhance patient counseling, guide risk stratification, and inform timely interventions to optimize long-term facial nerve function.

This study aimed to evaluate the performance of machine learning (ML) algorithms integrated with explainable artificial intelligence (XAI) techniques in predicting outcomes following flexible ureteroscopy (fURSL) in children. By identifying significant preoperative predictors, the goal was to improve individualized surgical risk assessment and planning. A retrospective analysis was conducted on 391 pediatric patients treated with fURSL for urolithiasis across 8 high-volume centers (2017-2021). Preoperative, intraoperative, and postoperative data were collected. 15 ML models were trained to predict five postoperative outcomes: fever, hematuria, sepsis, residual fragments (RF), and reintervention. A multitask artificial neural network (ANN) was also developed. Performance was evaluated using validation accuracy, confusion matrices, and classification reports. SHapley Additive exPlanations values and decision trees were used for model interpretability. Ensemble models outperformed others, with Gradient Boosting achieving 92.4% validation accuracy in predicting postoperative fever, Extra Trees achieving 91.1% for hematuria, and XGBoost reaching 96.0% for sepsis. Predictors included preoperative infections, stone burden, operative duration, and anatomical anomalies. For RF, Gradient Boost and Random Forest yielded strong results with up to 93.7% accuracy. Reintervention was best predicted by Random Forest, with RF as the strongest predictor. XAI techniques provided transparent, clinically interpretable models that aligned with medical reasoning. ML models demonstrated high accuracy in predicting adverse postoperative outcomes in pediatric ureteroscopy, with ensemble methods showing the best performance. Integration with XAI enhanced interpretability, supporting clinical decision-making. These findings underscore the potential of ML and XAI to inform personalized treatment strategies, though further prospective validation is needed to develop robust, generalizable predictive tools.

The decision to perform concomitant oophorectomy at the time of benign hysterectomy may be influenced by factors such as individual characteristics, socioeconomic factors, and surgeon expertise. Although hysterectomy rates have declined, evidence-based guidance on when to perform concomitant oophorectomy remains limited. This study aimed to evaluate local practice patterns and identify factors associated with the decision to perform oophorectomy in women undergoing hysterectomy for benign indications. Women aged 40-55 years who underwent hysterectomy for benign indications with or without concomitant oophorectomy at Başakşehir Çam and Sakura City Hospital between June 2020 and July 2021 were included in this retrospective cohort. Demographic, preoperative, intraoperative, postoperative, and pathologic data were analyzed to evaluate factors associated with concomitant bilateral oophorectomy during hysterectomy. Among 372 hysterectomy cases, 158 (42.5%) underwent concomitant bilateral salpingo-oophorectomy (BSO). Women in the BSO group were significantly older (49.38 ± 3.31 vs 45.37 ± 3.42 y; P < 0.001) and had a higher body mass index (BMI; 30.61 ± 5.09 vs 28.59 ± 4.91; P < 0.001). Ovarian cysts or masses were more frequently observed in the BSO group (84.2% vs 49.2%; P < 0.001). However, the proportion of pathologic findings beyond simple cysts was higher in the BSO group than in the non-BSO group (22% vs 6%; P < 0.001). Multivariable logistic regression showed that age, BMI, the presence of ovarian cysts, and endometrioma were independent predictors of BSO (overall model accuracy 85.9%; Nagelkerke R² = 0.65). No significant differences were observed in operative time, history of prior abdominal surgery, or comorbidities. Concomitant BSO remains commonly performed in perimenopausal women undergoing hysterectomy, and is primarily influenced by women's age, BMI, and intraoperative findings. Given the potential risks associated with surgical menopause, a more selective approach incorporating intraoperative assessment, guideline-based age thresholds, and improved counseling may help optimize clinical decision-making and surgical outcomes.

Introduction: The aim of this registry-based study was to analyse predictors of adverse events and the risk of implant revision surgery up to 10 years after proximal interphalangeal joint surface replacement. Methods: Adverse events in patients who underwent proximal interphalangeal joint surface replacement with the CapFlex-PIP prosthesis between 2010 and 2024 were analysed. Uni- and multivariate logistic regression models were developed to identify predictors of limited range of motion of the proximal interphalangeal joint (below 30°), implant-related problems and revision surgeries. The covariates included in the models were preoperative sociodemographic data, clinical measures, patient-reported outcomes and surgical details. Results: Four-hundred and thirty-one patients with 476 implants were analysed. The risk of reduction of range of movements increased with younger age and prior hand surgeries, but decreased for fingers with lax radial collateral ligament, as well as for the ring finger ( R 2 = 0.10). The little finger had a higher risk of implant-related problems. Greater surgical experience was associated with an 83% reduced risk of implant-related problems and a 76% reduced risk of revision surgery. Conclusion: Greater surgical experience was associated with fewer implant-related issues or the need for revision surgery in patients after proximal interphalangeal joint surface replacement. However, the models explained only a small proportion of the variance in adverse events, suggesting that there could be additional contributing factors. Level of evidence: IV

Abstract Introduction Neuroproliferative vestibulodynia has been proposed as a subtype of vulvodynia characterized by an increase in nerve growth factor (NGF) expression and nerve fiber density in the vulvar vestibule, leading to superficial dyspareunia. Similarly, in endometriosis, retrospective studies have also shown an increase in NGF expression and nerve fiber density around endometriosis lesions that correlates with deep dyspareunia. Objective The primary objective of this study was to validate the association between nerve fiber (bundle) density, NGF expression, and deep dyspareunia in endometriosis by utilizing a prospective registry. Methods Subjects undergoing initial endometriosis surgery at a tertiary referral center between December 2013 and December 2017 were included based on the presence of endometriosis of the posterior pelvis (ie, cul-de-sac, uterosacral ligaments, rectosigmoid colon). Participants were excluded based on post-menopausal status, malignancy, or bladder and/or pelvic floor tenderness to control for non-neuroproliferative deep dyspareunia cases (eg, related to central sensitization). Blinded immunohistochemical (IHC) scoring for protein-gene-project 9.5 (PGP9.5) nerve bundle density around the endometriosis lesions and for NGF expression by the endometriosis epithelium/stroma were linked to baseline (pre-operative) data on pain symptoms from a prospective registry at the center. First, Spearman correlation was used to investigate sample-level associations between PGP9.5 nerve bundle density and NGF expression, as well as by endometriosis subtype. Second, the mean IHC scores between the high deep dyspareunia [defined as: pre-operative deep dyspareunia severity ≥5/10 with reproduced vaginal apex tenderness on pelvic exam by a gynecologist], and low deep dyspareunia [defined as: deep dyspareunia severity &amp;lt;5/10 with no vaginal apex tenderness] groups were compared using Welch’s t-test. For comparison, the same analysis was done for other pain types without consideration of vaginal apex tenderness: superficial dyspareunia, dysmenorrhea, dyschezia, and chronic pelvic pain. Results This study involved 180 subjects with 276 endometriosis tissue samples, as subjects may have more than one sample of endometriosis. NGF expression in the endometriosis epithelium was associated with PGP9.5 nerve bundle density in superficial endometriosis lesions (rho: 0.24, P = 0.007). For pain, those with high deep dyspareunia (N = 64) had significantly higher mean PGP9.5 nerve bundle density (0.27, SD: 0.51) compared to those with low deep dyspareunia (N = 25) (0.12, SD: 0.09, P = 0.02). In contrast, neither NGF expression in the endometriosis epithelium (P = 0.88) nor the stroma (P = 0.89) significantly differed between the deep dyspareunia groups. No other endometriosis-associated pelvic pain symptoms (superficial dyspareunia, dysmenorrhea, dyschezia, chronic pelvic pain) were associated with nerve bundle density. Conclusions The association between deep dyspareunia and PGP9.5 nerve bundle density around endometriosis of the posterior pelvis was validated in this study utilizing a prospective registry. Therefore, we propose “neuroproliferative dyspareunia” as a distinct subtype of endometriosis-associated deep dyspareunia. Disclosure Any of the authors act as a consultant, employee or shareholder of an industry for: AbbVie, Pfizer, Ferring, Baxter International, Helix Biopharma, GSK, AstraZeneca, Merck, Mosaica.

Background: Glioblastoma is an extremely aggressive brain tumor that diffusely infiltrates white matter and alters large-scale brain connectivity. Most prognostic models focus on localized tumor features and clinical variables, overlooking broader effects on the brain's structural connectome. This study addressed this limitation by integrating graph-theoretical analysis of preoperative diffusion tensor imaging (DTI)-derived structural connectomes with machine learning (ML) to improve prediction of overall survival (OS) in newly diagnosed glioblastoma. Methods: Preoperative DTI data from 871 glioblastoma patients from the UPenn-GBM and UCSF-PDGM cohorts were processed to construct whole-brain structural connectomes weighted by tract count and quantitative anisotropy (QA). Global and nodal graph-theoretical network metrics were extracted and combined with demographic and clinical information. Ten ML models were trained and validated on 784 patients (90% of the cohort). The three best-performing algorithms were tested on a held-out cohort of 87 patients (10%). Results: Random forest, adaptive boosting, and KStar showed the strongest validation performance. In held-out internal testing, random forest models using degree and QA-weighted strength achieved accuracies of 0.862 and 0.874, with AUROCs of 0.929 and 0.909, for predicting OS beyond one year. Strength and clustering coefficient were key predictors, with over two-thirds of significant nodes localized in the temporal lobe, particularly the parahippocampal, and superior, middle, and inferior temporal gyri. Conclusions: Graph-theoretical quantification of structural brain network disruption combined with ML allows accurate prediction of OS in glioblastoma. These results support a network-based conceptualization of the disease and indicate that connectome-derived metrics may complement established prognostic frameworks.

Knee osteoarthritis (OA) is a common cause of pain and disability in the elderly, for which total knee arthroplasty (TKA) is an effective treatment. The medial parapatellar approach is widely used but may impair patellar blood supply and delay quadriceps recovery. The subvastus (SV) approach has been introduced to preserve the extensor mechanism and potentially allow faster post-operative rehabilitation. This study compares the SV and medial parapatellar approaches in primary TKA, focusing on early recovery of quadriceps function. This study aims to evaluate and compare the SV and medial parapatellar approaches for total knee replacement (TKR) in patients with OA at a tertiary care center, focusing on the time required for post-operative recovery of quadriceps function. We conducted a retrospective review of medical records from 200 patients with OA who underwent TKR at a tertiary care hospital between January 2024 and December 2025. Patients were categorized into two groups: Group A underwent TKR using the SV approach, while Group B received TKR through the medial parapatellar approach. Pre-operative data, including quadriceps strength, body mass index, and baseline demographic characteristics, were collected from initial workup notes. Demographic data of all the patients were described in Table. Starting from the 1st post-operative day, the time to the first unassisted straight leg raise was recorded as the primary measure of quadriceps function. The collected data were then analyzed to assess post-operative quadriceps recovery. The SV approach was associated with faster recovery of quadriceps function compared to the medial parapatellar approach. Patients in their sixties experienced the most significant improvement. In addition, pre-operative quadriceps strength was found to play a key role in post-operative muscle recovery. The SV approach in TKA promotes faster early recovery, including quicker quadriceps function, less pain, earlier range of motion, and shorter hospital stays compared to the medial parapatellar approach. Long-term functional outcomes are similar between the two techniques. Although technically more demanding, the SV approach is a safe and effective option that enhances early rehabilitation.

The role of artificial intelligence in pre-operative prediction of completeness of cytoreduction for peritoneal surface malignancies: a scoping review.

Differentiating large-duct pancreatic ductal adenocarcinoma from malignant intraductal papillary mucinous neoplasm: MRI characteristics and diagnostic implications.

Is multimodal monitoring effective in detecting neurological changes in cases of severe idiopathic scoliosis after halo gravity traction?: a series of 70 patients.

Assessment of the efficacy of 3D virtual surgical planning compared with traditional planning in maxillofacial reconstruction for facial traumatic deformity.

Hematologic indices in pediatric sleep-disordered breathing: a retrospective case-control study.

Red blood cells are often overprepared for surgery, leading to waste and increased costs, despite the need for cross-matching tests. This study aimed to develop a model for predicting the number of red blood cell units required during cardiovascular surgery using patient characteristics. This retrospective study included patients who underwent cardiovascular surgery at our hospital from April 2022 to October 2022. Multiple regression analysis was performed using preoperative patient attributes and blood data, with red blood cell units used during surgery as the objective variable. Models were developed using patient demographics and blood data, with additional models incorporating specific surgical procedures to assess predictive accuracy. Model 1 included age, sex, weight, hematocrit, prothrombin time-international normalized ratio, serum creatinine, volume of preoperatively donated diluted autologous blood, and history of cardiovascular surgery. Model 2 included the same variables as Model 1, plus aortic aneurysm resection as a surgical procedure. The predictive equations of the study showed superior accuracy for both Model 1 and Model 2 compared to the conventional red blood cell units requested by physicians or those predicted using the surgical blood order equation based on correlation coefficients. Model 2 outperformed both Model 1 and conventional methods in predictive utility. This study developed a clinically useful formula for predicting the number of red blood cell units required during surgery based on preoperative patient-specific data, without restricting the analysis to specific procedures. This formula may improve blood product inventory management and reduce medical costs.

Intraoperative language tasks often inadequately assess complex grammatical processes. This study aims to improve Persian cortical mapping by integrating object naming and sentence completion tasks. Multi-task language (MTL), consisted of an object naming task and a semantically relevant verb, was designed. The content validity ratio (CVR) and index (CVI) were calculated to determine task items retention. Twenty-five right-handed patients with brain mass lesions in language-associated regions were enrolled. Each patient received a medical recommendation for awake surgery, and preoperative fMRI data were collected through verb and syntax generation tasks. Activations in language-related regions for object naming, word generation, and MTL were compared. Lateralization of brain activity was evaluated for each participant by calculating the Laterality Index (LI). The CVR for all items ranged from 0.9 to 1 (0.93 ± 0.09), while the CVI ranged from 0.8 to 1 (0.94 ± 0.07). Significant activations were identified during object naming, particularly in the left precentral gyrus and inferior frontal gyrus (z = 7.53). For word generation, activations were bilateral in the fusiform gyrus and Broca (z = 5.54). The MTL task showed its highest activation in the fusiform and inferior temporal gyrus (z = 6.86). A significant difference was found between the LI of object naming and word generation (p = 0.003), while MTL comparisons were not statistically significant. The developed Persian intraoperative language paradigm, named MTL, demonstrated validity and feasibility, with significant activation in key language areas, particularly the left inferior frontal gyrus.

Somatic and psychological predictors of chronic postsurgical pain in cancer patients: a machine learning approach in a longitudinal two-centre study.

Objective: to present corneal morphological changes in patients with initial keratoconus during 12-month follow-up period after simultaneous transepithelial photorefractive keratectomy with accelerated ultraviolet collagen crosslinking. Patients and methods. The study included 30 patients/eyes aged 18–45 years with progressive stage 1 keratoconus, who underwent laser scanning in vivo confocal microscopy before and after treatment to measure the total, average, minimum and maximum lengths of corneal nerve fibres, as well as the sub-basal nerve plexus density; corneal optical coherence tomography with measurement of the demarcation line depth; Scheimpflug densitometry of the anterior-central layers of corneal stroma at 0–2 and 2–6-mm zones. The examinations were performed 1, 6 and 12 months after surgery. Results. Morphological changes observed by laser scanning in vivo confocal microscopy located in the anterior and middle stroma within 300 μm depth and were temporary: at 1 month — epitheliopathy, subepithelial fibroplasia, honeycomb-type lacunar edema and a sharp decrease in sub-basal nerve plexus density; at 6 months — partial regeneration of the keratocytes and sub-basal nerve plexus, reduction of the lacunar edema; at 12 months — nealrly complete restoration of corneal cellular structure and sub-basal innervation (sub-basal nerve plexus density — 80.5 % of preoperative data, total length of nerve fibres — 82 %). According to Scheimpflug densitometry, the highest values of corneal optical density was detected at 1 month after surgery and subsequently returned to baseline at 12 months, indicating reducing of postoperative corneal haze. According to anterior segment optical coherence tomography, the demarcation line depth was 198.5 (179.25; 214.5) μm visualized at 1 month postoperatively. Conclusion. Simultaneous transepithelial photorefractive keratectomy with accelerated corneal crosslinking can be considered as a safe and effective approach to managing initial keratoconus, that ensures morphological restoration of the cornea and minimal risk of severe fibroblastic complications during one-year follow-up period. Further research is needed to assess the long-term stability of the results.