Search Results (159,497)

Background: This study investigates the impact of anatomically constrained preprocessing and deep learning architecture selection on benign versus malignant breast lesion classification in contrast-enhanced mammography (CEM), with the goal of improving robustness and clinical reliability across heterogeneous data sources. Methods: In this retrospective multicenter study, CEM images from 300 patients (314 lesions) were combined with 1003 publicly available CEM images, yielding a total of 1120 breast cases. Automatic breast segmentation was performed using the LIBRA framework to generate breast-mask images. Eleven deep learning models, including classical convolutional neural networks, attention-based networks, hybrid convolutional neural networks (CNNs), Transformer architectures, and mammography-specific models, were trained and evaluated using both original DICOM images and breast-mask inputs. Performance was assessed using accuracy, balanced accuracy, sensitivity, specificity, AUROC, and AUPRC on cross-validation and independent test sets. Hyperparameter optimization was conducted for the best-performing architecture. Results: Models trained on breast-mask images consistently outperformed those trained on original DICOM images across all architectures and metrics, with AUROC improvements ranging from +0.06 to +0.21. Among all models, ResNet50 trained on breast-mask images achieved the best performance (AUROC = 0.931; AUPRC = 0.933; balanced accuracy = 0.834), further improved after optimization (balanced accuracy = 0.886; sensitivity = 0.842; specificity = 0.930). Classical CNN architectures demonstrated performance comparable to or exceeding that of more complex hybrid CNN–Transformer models when anatomically focused preprocessing and rigorous optimization were applied. Conclusions: Anatomically constrained preprocessing through breast-mask segmentation substantially enhances deep learning performance and stability in CEM-based breast lesion classification. These findings indicate that input representation quality and training optimization are critical determinants of clinically relevant performance, often outweighing architectural complexity, and may support more reliable AI-assisted decision support in CEM workflows. Full article

Background/Objectives: Immune checkpoint inhibitors (ICIs) are increasingly used in hepatocellular carcinoma (HCC), but their application around liver transplantation (LT) remains controversial because checkpoint blockade may enhance antitumor immunity while disrupting graft tolerance. We systematically reviewed the available evidence on ICI exposure before LT and ICI therapy after LT for recurrent HCC. Methods: A PRISMA-guided systematic review with qualitative synthesis was performed. PubMed/MEDLINE, Embase, and Web of Science were searched from inception to 15 March 2026. Studies including adult patients with HCC treated with PD-1-, PD-L1-, and/or CTLA-4-targeting ICIs before LT or after LT for recurrent HCC were eligible. Results: Fifty-one studies were included. In the pre-LT setting, 25 studies reported 576 transplanted patients. Acute allograft rejection occurred in approximately 22% and graft loss in 3.8%, and shorter washout intervals were consistently associated with higher rejection risk. In the post-LT setting, 26 studies reported 117 recipients treated with ICIs; at least 22 rejection episodes (18.8%) were described, usually within 2–4 weeks of treatment initiation, with limited and inconsistent antitumor benefit. Conclusions: Pre-LT ICI use appears feasible in selected patients when adequate washout is respected. Post-LT ICI therapy remains high risk and should be reserved for highly selected cases within a multidisciplinary framework. Full article

Background: Frailty is a major determinant of adverse outcomes in older adults with cardiovascular disease. Automated digital tools may facilitate routine frailty assessment in hospital settings; however, their validity and prognostic relevance in acutely hospitalized patients remain insufficiently established. Methods: In this prospective cohort study, 113 hospitalized cardiology patients underwent frailty assessment using both manual Short Physical Performance Battery (mSPPB) and an automated electronic SPPB (eSPPB) system. Agreement between methods was evaluated using Pearson correlation, intraclass correlation coefficients (ICCs), and Bland–Altman analysis. Frailty was defined as SPPB < 5. The association between frailty and 30-day mortality was assessed using logistic regression and Kaplan–Meier survival analysis. Results: Seventeen patients (15.0%) were classified as frail. Automated and manual SPPB scores were highly correlated (r = 0.994, p < 0.001) and demonstrated good agreement (ICC = 0.80). Bland–Altman analysis showed a mean difference of −1.63 points (95% limits of agreement −4.41 to 1.16). Frailty was associated with significantly higher 30-day mortality (17.6% vs. 2.1%, p = 0.009), corresponding to a tenfold increase in mortality odds (OR 10.07; 95% CI 1.5–67.5). An exploratory model showed apparent discriminative performance (AUC 0.83; 95% CI 0.71–0.95). Conclusions: Automated eSPPB demonstrated good agreement with manual assessment and was significantly associated with short-term mortality in hospitalized cardiovascular patients. These findings support the validity and potential clinical utility of automated frailty assessment for risk stratification in acute cardiology settings. Full article

Managing blood glucose in type 1 diabetes (T1D) remains a daily clinical challenge, and accurate short-term prediction of glucose levels can meaningfully improve insulin dosing decisions while reducing the risk of dangerous hypoglycaemic episodes. Although numerous machine learning approaches have been proposed for this task, comparing their relative merits is difficult because published studies differ widely in datasets, preprocessing choices, and evaluation criteria. In this work, we address this research gap by benchmarking ten machine learning methods—from a naïve persistence baseline through classical linear regressors, gradient-boosted ensembles, and recurrent neural networks to a novel hybrid that couples LightGBM with stochastic differential equation (SDE)-based glucose–insulin simulation—on two multi-patient datasets comprising 34 T1D subjects, across prediction horizons of 15, 30, 60, and 120 min. Every method is trained and tested under identical preprocessing and temporal splitting conditions to ensure a fair comparison. The proposed Hybrid LightGBM-SDE model consistently outperforms all alternatives, recording RMSE values of 22.42 mg/dL at 15 min, 28.74 mg/dL at 30 min, 33.89 mg/dL at 60 min, and 37.22 mg/dL at 120 min—an improvement of between 13.6% and 27.0% relative to standalone LightGBM. At the clinically important 30 min horizon, 99.7% of predictions lie within the acceptable A and B zones of the Clarke Error Grid. Wilcoxon signed-rank tests confirm that performance differences are statistically significant (p < 10⁻¹⁰), and SHAP-based analysis shows that the SDE-derived simulation features are among the most influential predictors, especially at longer horizons. All source code and evaluation scripts are publicly released to support reproducibility. Due to temporary data access constraints, all experiments reported here use physics-based synthetic datasets generated from the Bergman minimal model, replicating the structural properties of the D1NAMO and HUPA-UCM collections; validation on the original clinical recordings is planned. Among the two synthetic datasets, the D1NAMO-equivalent cohort (nine patients) proves more challenging, with systematically higher per-patient RMSE variance. The clinically acceptable prediction accuracy at the 30 min horizon (99.7% in Clarke zones A + B) suggests potential for integration into insulin dosing decision-support systems. Full article

Background/Objectives: Nipple-sparing mastectomy with immediate reconstruction is a preferred option for selected patients undergoing prophylactic or therapeutic mastectomy. Optimizing postoperative wound care is essential to support healing, preserve the nipple–areola complex, and prevent delays in oncologic treatments. This retrospective observational study aimed to evaluate the clinical outcomes associated with the use of the NovoX^® Cup medical device in post-NSM surgical wound management, assessing clinical–surgical outcomes and quality of life (QoL). Methods: We conducted a retrospective observational study on 54 patients who underwent NSM with immediate reconstruction at AOUI Verona between January 2025 and January 2026; Novox^® Cup was applied intraoperatively and changed every 48 h according to protocol. Surgeon-reported outcomes were assessed by the skin flap viability scale and the complications by Clavien–Dindo classification. Patient-reported outcomes were assessed via the Wound-QoL17 questionnaire at 7, 30, and 90 days. Clinical outcomes were supported by photographic documentation. Results: Mean age was 51.5 years; BMI averaged 23.9 kg/m². Local complications occurred in 30.4% of cases (infections 12%, dehiscence 10%, seromas 4%). Mean healing time was 15 days, with 87.4% of patients having drains removed by day 14. One patient required surgical revision, and one (1.8%) experienced delayed adjuvant therapy. Wound-QoL17 responses showed minimal discomfort and high satisfaction. Clinical evaluation revealed favorable wound appearance and preserved NAC perfusion within 48 h. Conclusions: Novox^® Cup appears effective in supporting wound healing and NAC preservation after NSM, with high patient satisfaction and minimal treatment delays. Its integration into postoperative care may enhance outcomes and maintain oncologic timelines. Full article

Background: Robot-assisted surgery has become increasingly used across multiple specialties; however, its integration into aesthetic plastic surgery remains limited. Individualized patient requirements, such as concealed scar placement, superficial soft tissue dissection, and patient-specific docking angles, are major challenges to its adoption, unlike in other specialties. This review aimed to evaluate the current use of robotic systems in plastic surgery, with a particular focus on aesthetic procedures, operative outcomes, and existing technological limitations. Methods: Multiple databases, including PubMed, Scopus, and Google Scholar, were extensively searched to identify studies published between 2011 and 2026. Data on robotic platforms, operative duration, rehabilitation outcomes, and aesthetic indications were extracted and analyzed. Robotic systems such as da Vinci, Symani, MUSA, and ARTAS demonstrated feasibility across reconstructive subspecialties. However, their clinical application remains limited, as purely aesthetic procedures are rare, highlighting a significant lack of standardized docking methods and dedicated instruments. Results: The data show that robotic platforms offer great advantages, such as precision and minimally invasive access; however, their high costs, bulky instrumentation, and limited docking methods represent barriers to their adoption in aesthetic surgery. Conclusions: Robot-assisted aesthetic plastic surgery remains in the early stage of development. Further research is required to establish reproducible docking standards and expand its clinical indications. Advancements in single-port systems, artificial intelligence integration, and surgeon training will facilitate broader clinical implementation. Full article

Objective: Existing deep learning (DL) studies on brain metastasis have largely focused on algorithm or reader performance in controlled settings, whereas its role in routine teleradiology quality support remains unestablished. We evaluated the report-level impact of DL assistance on brain metastasis interpretation in a real-world teleradiology workflow using dual-sequence MRI. Materials and Methods: In this retrospective study, 600 patients who underwent contrast-enhanced dual-sequence brain MRI during two consecutive 3-month periods before (pre-DL, n = 286) and after (post-DL, n = 314) DL integration into teleradiology workflow were analyzed. Ten board-certified teleradiologists interpreted all the cases with or without DL-generated overlays. Report-level diagnostic metrics were assessed against a consensus reference standard established by faculty neuroradiologists. Subsequently, exploratory case-level stratified sensitivity analyses were performed for metastasis-positive examinations based on lesion multiplicity and the largest lesion size. Teleradiologists’ perceptions were assessed using a post-interpretation survey. Results: Compared with the pre-DL group, the post-DL group showed higher sensitivity (77.7% vs. 90.8%, p < 0.001), specificity (82.3% vs. 90.8%, p = 0.002), accuracy (80.8% vs. 90.8%, p < 0.001), positive predictive value (68.2% vs. 85.7%, p < 0.001), and negative predictive value (88.3% vs. 94.2%, p = 0.011). False-positive and false-negative rates were lower after DL implementation (11.9% vs. 5.7%, p = 0.009; 7.3% vs. 3.5%, p = 0.045). Sensitivity gains were most pronounced for cases with single metastasis (74.6% vs. 91.2%, p = 0.007) and with the largest lesion ≤ 5 mm (74.3% vs. 92.0%, p = 0.004), whereas sensitivity was similar for multiple metastases and for cases with a largest lesion > 5 mm. Survey responses suggested favorable usability and diagnostic support. Conclusions: In this real-world teleradiology workflow, DL implementation was associated with higher report-level diagnostic metrics and fewer false interpretations. DL assistance may help support quality control for brain metastasis interpretation, particularly in more subtle and diagnostically challenging cases, although radiologist judgment remains essential for subtle or borderline lesions. Full article

Background and objectives: Increasing daily steps during weight management programs remains one of the most common recommendations; however, why, when and how many is still unclear. To clarify this, we aim to conduct a systematic review and meta-analysis. Methods: The study was conducted in adherence to the PRISMA guidelines on randomized controlled trials (RCTs), that included weight loss (WL) interventions based on lifestyle modification programs (LSMs), compared to “as usual care” considered as controls, to whom both daily steps and WL% were reported or retrievable at baseline (Time 0), end of WL phase (Time 1, WL1%), and weight maintenance phase (Time 2, WL2%), for both arms. Results: A total of 18 RCTs met the inclusion criteria and were included in the systematic review. Of those, 14 underwent meta-analysis and five main findings were revealed: (i) at baseline (Time 0), no significant difference was observed in mean daily steps between the LSM and controls (7280 vs. 7180, p = 0.336), reflecting a similar lifestyle between arms; (ii) at Time 1, the mean duration was 7.88 months (range = 3–12 months), and the LSM arm showed a significant increase in daily steps with respect to baseline (8454 vs. 7486 steps, p = 0.017) and a significant WL (WL1% = 4.39%, p < 0.001); (iii) at Time 2, the mean duration was 10.27 months (range = 3–24 months), and the LSM arm maintained the level of daily steps achieved by the end of WL phase (8241 vs. 8454 steps, p > 0.05), and also a significant WL% (WL2% = 3.28%, p = 0.001); (iv) the control arm showed no significant changes in daily steps and weight status at all times of assessment; and (v) the meta-regression showed in the LSM arm a positive relationship between daily steps at Time 1 (β = 1.33, p = 0.03) and Time 2 (β = 1.10, p = 0.02), both with WL2%. Conclusions: Our preliminary study results support that during LSM programs, patients should be encouraged to increase their daily steps during the WL phase, targeting approximately 8500 steps/day and maintaining these levels during the maintenance phase, since this strategy appears to be a useful behavioral approach associated with maintaining significant WL in the long term. Full article

Background/Objectives: Nurses and dialysis technicians are primarily responsible for cannulation in in-center and satellite dialysis units. Despite being a core component of hemodialysis care, existing clinical guidelines offer limited standardization, resulting in practice variability across facilities. Therefore, clinical expertise and adherence to consistent standards are essential to ensure safe and effective vascular access management. The study aimed to investigate the variables related to patients and nurses that contribute to unsuccessful vascular access cannulations, as well as the actions taken in response to cannulation failure, in a tertiary dialysis center in the Eastern Region of Saudi Arabia. Methods: This retrospective analytic study reviewed the records of 228 adult hemodialysis patients at King Fahad Military Medical Complex from 2020 to 2024, analyzing demographic, clinical, vascular access, and nursing variables associated with cannulation failure using descriptive statistics, the chi-square test, and t-tests. Ethical approval was obtained, and data were de-identified and manually extracted from nursing and dialysis documentation. Results: Most patients had hypertension and diabetes, with significant comorbidity burdens. Infiltration (61%) and clot formation (30.7%) were the primary complications of cannulation failure. Significant associations emerged with recurrent stroke and peripheral vascular disease, but not with nurse or patient demographics, suggesting vascular factors outweigh staff variables in cannulation risk. Cannulation failures were most common in patients with vascular comorbidities, while staff experience and education had no significant impact. Conclusions: Recommendations include implementing tailored protocols, providing ongoing nurse education, conducting systematic vascular assessments, and holding regular team reviews to enhance access outcomes and patient safety. Full article