Search Results (3,635)

If we do not urgently educate current and future medical professionals to critically evaluate and distinguish credible AI-assisted diagnostic tools from those whose performance is artificially inflated by data leakage or improper validation, we risk undermining clinician trust in all AI diagnostics and jeopardizing future advances in patient care. For instance, machine learning models have shown high accuracy in diagnosing Parkinson’s Disease when trained on clinical features that are themselves diagnostic, such as tremor and rigidity. This study systematically investigates the impact of data leakage and feature selection on the true clinical utility of machine learning models for early Parkinson’s Disease detection. We constructed two experimental pipelines: one excluding all overt motor symptoms to simulate a subclinical scenario and a control including these features. Nine machine learning algorithms were evaluated using a robust three-way data split and comprehensive metric analysis. Results reveal that, without overt features, all models exhibited superficially acceptable F1 scores but failed catastrophically in specificity, misclassifying most healthy controls as Parkinson’s Disease. The inclusion of overt features dramatically improved performance, confirming that high accuracy was due to data leakage rather than genuine predictive power. These findings underscore the necessity of rigorous experimental design, transparent reporting, and critical evaluation of machine learning models in clinically realistic settings. Our work highlights the risks of overestimating model utility due to data leakage and provides guidance for developing robust, clinically meaningful machine learning tools for early disease detection. Full article

The aim of this work is to evaluate the effectiveness of antimicrobial photodynamic therapy (aPDT) against oral MDR (multi-drug-resistant) Candida spp. infections related to orthodontic treatment with palatal expanders through in vitro study. Methods: PDT protocol: Curcumin + H₂O₂ was used as a photosensitizer activated by a 460 nm diode LED lamp, with an 8 mm blunt tip for 2 min in each spot of interest. In vitro simulation: A palatal expander sterile device was inserted into a custom-designed orthodontic bioreactor, realized with 10 mL of Sabouraud dextrose broth plus 10% human saliva and infected with an MDR C. albicans clinical isolate CA95 strain to reproduce an oral palatal expander infection. After 48 h of incubation at 37 °C, the device was treated with the PDT protocol. Two samples before and 5 min after the PDT process were taken and used to contaminate a Petri dish with a Sabouraud field to evaluate Candida spp. CFUs (colony-forming units). Results: A nearly 99% reduction in C. albicans colonies in the palatal expander biofilm was found after PDT. Conclusion: The data showed the effectiveness of using aPDT to treat palatal infection; however, specific patient oral micro-environment reproduction (Ph values, salivary flow, mucosal adhesion of photosensitizer) must be further analyzed. Full article

The evidence for photobiomodulation in reducing postoperative pain after endodontic instrumentation is classified as low or very low certainty, indicating a need for further research. Longitudinal pain assessments over 24 h are crucial, and studies should explore these pain periods. Background/Objectives: This double-blind, randomized controlled clinical trial evaluated the effect of PBM on pain following single-visit endodontic treatment of maxillary molars at 4, 8, 12, and 24 h. Primary outcomes included pain at 24 h; secondary outcomes included pain at 4, 8, and 12 h, pain during palpation/percussion, OHIP-14 analysis, and frequencies of pain. Methods: Approved by the Research Ethics Committee (5.598.290) and registered in Clinical Trials (NCT06253767), the study recruited adults (21–70 years) requiring endodontic treatment in maxillary molars. Fifty-eight molars were randomly assigned to two groups: the PBM Group (n = 29), receiving conventional endodontic treatment with PBM (100 mW, 333 mW/cm², 9 J distributed at 3 points near root apices), and the control group (n = 29), receiving conventional treatment with PBM simulation. Pain was assessed using the Visual Analog Scale. Results: Statistical analyses used chi-square and Mann–Whitney tests, with explained variance (η²). Ten participants were excluded, leaving 48 patients for analysis. No significant differences were observed in postoperative pain at 24, 4, 8, or 12 h, or in palpation/percussion or OHIP-14 scores. Pain frequencies ranged from 12.5% to 25%. Conclusions: PBM does not influence post-treatment pain in maxillary molars under these conditions. These results emphasize the importance of relying on well-designed clinical trials to guide treatment decisions, and future research should focus on personalized dosimetry adapted to the anatomical characteristics of the treated dental region to enhance the accuracy and efficacy of therapeutic protocols. Full article

Background: The knee joint performs a vital function in human movement, supporting significant loads and ensuring stability during daily activities. Methods: The objective of this study was to develop and validate a subject-specific framework to model knee flexion–extension by integrating 3D gait data with individualized musculoskeletal (MS) and finite element (FE) models. In this proof of concept, gait data were collected from a 52-year-old woman using Xsens inertial sensors. The MS model was based on the same subject to define realistic loading, while the 3D knee FE model, built from another individual’s MRI, included all major anatomical structures, as subject-specific morphing was not possible due to unavailable scans. Results: The FE simulation showed principal stresses from –28.67 to +44.95 MPa, with compressive stresses between 2 and 8 MPa predominating in the tibial plateaus, consistent with normal gait. In the ACL, peak stress of 1.45 MPa occurred near the femoral insertion, decreasing non-uniformly with a compressive dip around –3.0 MPa. Displacement reached 0.99 mm in the distal tibia and decreased proximally. ACL displacement ranged from 0.45 to 0.80 mm, following a non-linear pattern likely due to ligament geometry and local constraints. Conclusions: These results support the model’s ability to replicate realistic, patient-specific joint mechanics. Full article

Large, high-quality patient datasets are essential for applications like economic modeling and patient simulation. However, real-world data is often inaccessible or incomplete. Synthetic patient data offers an alternative, and current methods often fail to preserve clinical plausibility, real-world correlations, and logical consistency. This study presents a patient cohort generator designed to produce realistic, statistically valid synthetic datasets. The generator uses predefined probability distributions and Cholesky decomposition to reflect real-world correlations. A dependency matrix handles variable relationships in the right order. Hard limits block unrealistic values, and binary variables are set using percentiles to match expected rates. Validation used two datasets, NHANES (2021–2023) and the Framingham Heart Study, evaluating cohort diversity (general, cardiac, low-dimensional), data sparsity (five correlation scenarios), and model performance (MSE, RMSE, R², SSE, correlation plots). Results demonstrated strong alignment with real-world data in central tendency, dispersion, and correlation structures. Scenario A (empirical correlations) performed best (R² = 86.8–99.6%, lowest SSE and MAE). Scenario B (physician-estimated correlations) also performed well, especially in a low-dimensions population (R² = 80.7%). Scenario E (no correlation) performed worst. Overall, the proposed model provides a scalable, customizable solution for generating synthetic patient cohorts, supporting reliable simulations and research when real-world data is limited. While deep learning approaches have been proposed for this task, they require access to large-scale real datasets and offer limited control over statistical dependencies or clinical logic. Our approach addresses this gap. Full article

Purpose: To present the findings of our preliminary experience using daily image-guided radiotherapy (IGRT) supported by implanted fiducial markers (FMs) in the radiotherapy of the vaginal cuff, in a cohort of post-surgery endometrial cancer patients. Methods: Patients with vaginal cuff cancer requiring adjuvant radiation with external beams were enrolled. Five patients underwent radiation therapy targeting the pelvic disease and positive lymph nodes, with doses of 50.4 Gy in twenty-eight fractions and a subsequent stereotactic boost on the vaginal vault at a dose of 5 Gy in a single fraction. One patient was administered 30 Gy in five fractions to the vaginal vault. These patients underwent external beam RT following the implantation of three 0.40 × 10 mm gold fiducial markers (FMs). Our IGRT strategy involved real-time 2D kV image-based monitoring of the fiducial markers during the treatment delivery as a surrogate of the vaginal cuff. To explore the potential role of FMs throughout the treatment process, we analyzed cine movies of the 2D kV-triggered images during delivery, as well as the image registration between pre- and post-treatment CBCT scans and the planning CT (pCT). Each CBCT used to trigger fraction delivery was segmented to define the rectum, bladder, and vaginal cuff. We calculated a standard metric to assess the similarity among the images (Dice index). Results: All the patients completed radiotherapy and experienced good tolerance without any reported acute or long-term toxicity. We did not observe any loss of FMs during or before treatment. A total of twenty CBCTs were analyzed across ten fractions. The observed trend showed a relatively emptier bladder compared to the simulation phase, with the bladder filling during the delivery. This resulted in a final median Dice similarity coefficient (DSC) of 0.90, indicating strong performance. The rectum reproducibility revealed greater variability, negatively affecting the quality of the delivery. Only in two patients, FMs showed intrafractional shift > 5 mm, probably associated with considerable rectal volume changes. Target coverage was preserved due to a safe CTV-to-PTV margin (10 mm). Conclusions: In our preliminary study, CBCT in combination with the use of fiducial markers to guide the delivery proved to be a feasible method for IGRT both before and during the treatment of post-operative gynecological cancer. In particular, this approach seems to be promising in selected patients to facilitate the use of SBRT instead of BRT (brachytherapy), thanks to margin reduction and adaptive strategies to optimize dose delivery while minimizing toxicity. A larger sample of patients is needed to confirm our results. Full article

Background/Objectives: Accurate and accessible perioperative health information empowers patients and enhances recovery outcomes. Artificial intelligence tools, such as ChatGPT, have garnered attention for their potential in health communication. This study evaluates the accuracy and readability of responses generated by ChatGPT to questions commonly asked about breast cancer. Methods: Fifteen simulated patient queries about breast cancer surgery preparation and recovery were prepared. Responses generated by ChatGPT (4o version) were evaluated for accuracy by a pool of breast surgeons using a 4-point Likert scale. Readability was assessed with the Flesch–Kincaid Grade Level (FKGL). Descriptive statistics were used to summarize the findings. Results: Of the 15 responses evaluated, 11 were rated as “accurate and comprehensive”, while 4 out of 15 were deemed “correct but incomplete”. No responses were classified as “partially incorrect” or “completely incorrect”. The median FKGL score was 11.2, indicating a high school reading level. While most responses were technically accurate, the complexity of language exceeded the recommended readability levels for patient-directed materials. Conclusions: The model shows potential as a complementary resource for patient education in breast cancer surgery, but should not replace direct interaction with healthcare providers. Future research should focus on enhancing language models’ ability to generate accessible and patient-friendly content. Full article

Background: Hepatocellular carcinoma (HCC) is one of the most common and fatal malignancies worldwide, characterized by remarkable molecular heterogeneity and poor clinical outcomes. Despite advancements in diagnosis and treatment, the prognosis for HCC remains dismal, largely due to late-stage diagnosis and limited therapeutic efficacy. Therefore, there is a critical need to identify novel therapeutic targets and explore alternative strategies, such as drug repurposing, to improve patient outcomes. Methods: In this study, we employed network pharmacology, molecular docking, and molecular dynamics (MD) simulations to explore the potential therapeutic targets of Nirmatrelvir in HCC. Results: Nirmatrelvir targets were predicted through SwissTarget (101 targets), SuperPred (1111 targets), and Way2Drug (38 targets). Concurrently, HCC-associated genes (5726) were retrieved from DisGeNet. Cross-referencing the two datasets identified 29 overlapping proteins. A protein–protein interaction (PPI) network constructed from the overlapping proteins was analyzed using CytoHubba, identifying 10 hub genes, with HDAC1, HDAC3, and STAT3 achieving the highest degree scores. Molecular docking revealed a strong binding affinity of Nirmatrelvir to HDAC1 (docking score = −7.319 kcal/mol), HDAC3 (−6.026 kcal/mol), and STAT3 (−6.304 kcal/mol). Moreover, Nirmatrelvir displayed stable dynamic behavior in repeated 200 ns simulation analyses. Binding free energy calculations using MM/GBSA showed values of −23.692 kcal/mol for the HDAC1–Nirmatrelvir complex, −33.360 kcal/mol for HDAC3, and −21.167 kcal/mol for STAT3. MM/PBSA analysis yielded −17.987 kcal/mol for HDAC1, −27.767 kcal/mol for HDAC3, and −16.986 kcal/mol for STAT3. Conclusions: The findings demonstrate Nirmatrelvir’s strong binding affinity towards HDAC3, underscoring its potential for future drug development. Collectively, the data provide computational evidence for repurposing Nirmatrelvir as a multi-target inhibitor in HCC therapy, warranting in vitro and in vivo studies to confirm its clinical efficacy and safety and elucidate its mechanisms of action in HCC. Full article

Aims/Background: Polypharmacy, or the concurrent intake of five or more medications, is a significant issue in clinical practice, particularly in multimorbid elderly individuals. Despite its importance for patient safety, medical education often lacks systematic training in recognising and managing polypharmacy within the framework of patient-centred care. We investigated the impact of a structured learning intervention introducing polypharmacy as a chronic condition, assessing whether it enhances medical students’ diagnostic competence, confidence, and interprofessional collaboration. Methods: A prospective cohort study was conducted with 50 final-year medical students who received a three-phase educational intervention. Phase 1 was interactive workshops on the principles of polypharmacy, its dangers, and diagnostic tools. Phase 2 involved simulated patient consultations and medication review exercises with pharmacists. Phase 3 involved reflection through debriefing sessions, reflective diaries, and standardised patient feedback. Student knowledge, confidence, and attitudes towards polypharmacy management were assessed using pre- and post-intervention questionnaires. Quantitative data were analysed through paired t-tests, and qualitative data were analysed thematically from reflective diaries. Results: Students demonstrated considerable improvement after the intervention in identifying symptoms of polypharmacy, suggesting deprescribing strategies, and working in multidisciplinary teams. Confidence in prioritising polypharmacy as a primary diagnostic problem increased from 32% to 86% (p < 0.01), and knowledge of diagnostic tools increased from 3.1 ± 0.6 to 4.7 ± 0.3 (p < 0.01). Standardised patients felt communication and patient-centredness had improved, with satisfaction scores increasing from 3.5 ± 0.8 to 4.8 ± 0.4 (p < 0.01). Reflective diaries indicated a shift towards more holistic thinking regarding medication burden. The small sample size limits the generalisability of the results. Conclusions: Teaching polypharmacy as a chronic condition in medical school enhances diagnostic competence, interprofessional teamwork, and patient safety. Education is a structured way of integrating the management of polypharmacy into routine clinical practice. This model provides valuable insights for designing medical curricula. Future research must assess the impact of such training on patient outcomes and clinical decision-making in the long term. Full article

Background/Objectives: There is significant interest in using texture features to extract hidden image-based information. In medical imaging applications using radiomics, AI, or personalized medicine, the quest is to extract patient or disease specific information while being insensitive to other system or processing variables. While we use digital breast tomosynthesis (DBT) to show these effects, our results would be generally applicable to a wider range of other imaging modalities and applications. Methods: We examine factors in texture estimation methods, such as quantization, pixel distance offset, and region of interest (ROI) size, that influence the magnitudes of these readily computable and widely used image texture features (specifically Haralick’s gray level co-occurrence matrix (GLCM) textural features). Results: Our results indicate that quantization is the most influential of these parameters, as it controls the size of the GLCM and range of values. We propose a new multi-resolution normalization (by either fixing ROI size or pixel offset) that can significantly reduce quantization magnitude disparities. We show reduction in mean differences in feature values by orders of magnitude; for example, reducing it to 7.34% between quantizations of 8–128, while preserving trends. Conclusions: When combining images from multiple vendors in a common analysis, large variations in texture magnitudes can arise due to differences in post-processing methods like filters. We show that significant changes in GLCM magnitude variations may arise simply due to the filter type or strength. These trends can also vary based on estimation variables (like offset distance or ROI) that can further complicate analysis and robustness. We show pathways to reduce sensitivity to such variations due to estimation methods while increasing the desired sensitivity to patient-specific information such as breast density. Finally, we show that our results obtained from simulated DBT images are consistent with what we see when applied to clinical DBT images. Full article

This study presents a DIKWP-driven artificial consciousness framework for IoT-enabled smart healthcare, integrating a Data–Information–Knowledge–Wisdom–Purpose (DIKWP) cognitive architecture with a software-defined IoT infrastructure. The proposed system deploys DIKWP agents at edge and cloud nodes to transform raw sensor data into high-level knowledge and purpose-driven actions. This is achieved through a structured DIKWP pipeline—from data acquisition and information processing to knowledge extraction, wisdom inference, and purpose-driven decision-making—that enables semantic reasoning, adaptive goal-driven responses, and privacy-preserving decision-making in healthcare environments. The architecture integrates wearable sensors, edge computing nodes, and cloud services to enable dynamic task orchestration and secure data fusion. For evaluation, a smart healthcare scenario for early anomaly detection (e.g., arrhythmia and fever) was implemented using wearable devices with coordinated edge–cloud analytics. Simulated experiments on synthetic vital sign datasets achieved approximately 98% anomaly detection accuracy and up to 90% reduction in communication overhead compared to cloud-centric solutions. Results also demonstrate enhanced explainability via traceable decisions across DIKWP layers and robust performance under intermittent connectivity. These findings indicate that the DIKWP-driven approach can significantly advance IoT-based healthcare by providing secure, explainable, and adaptive services aligned with clinical objectives and patient-centric care. Full article

Medical education often emphasizes theoretical knowledge, limiting students’ opportunities to practice history taking, a structured interview that elicits relevant patient information before clinical decision making. Large language models (LLMs) offer novel solutions by generating simulated patient interviews. This study evaluated the educational potential of LLM-generated history-taking dialogues, focusing on clinical validity and diagnostic diversity. Chest pain was chosen as a representative case given its frequent presentation and importance for differential diagnosis. A fine-tuned Gemma-3-27B, specialized for medical interviews, was compared with GPT-4o-mini, a freely accessible LLM, in generating multi-branching history-taking dialogues, with Claude-3.5 Sonnet inferring diagnoses from these dialogues. The dialogues were assessed using a Chest Pain Checklist (CPC) and entropy-based metrics. Gemma-3-27B outperformed GPT-4o-mini, generating significantly more high-quality dialogues (90.7% vs. 76.5%). Gemma-3-27B produced diverse and focused diagnoses, whereas GPT-4o-mini generated broader but less specific patterns. For demographic information, such as age and sex, Gemma-3-27B showed significant shifts in dialogue patterns and diagnoses aligned with real-world epidemiological trends. These findings suggest that LLMs, particularly those fine-tuned for medical tasks, are promising educational tools for generating diverse, clinically valid interview scenarios that enhance clinical reasoning in history taking. Full article

Orthognathic surgery (OS) is a complex procedure commonly used to treat dentofacial deformities (DFDs). These conditions, related to jaw position or size and often involving malocclusion, affect approximately 15% of the population. Due to the complexity of OS, accurate planning is essential. Digital assessment using computer-aided design (CAD) and computer-aided manufacturing (CAM) tools enhances surgical predictability. However, limitations in soft tissue simulation often require surgeon input to optimize aesthetic results and minimize surgical impact. This study aimed to evaluate the accuracy of virtual surgery planning (VSP) by analyzing the relationship between planning deviations and surgical satisfaction. A single-center, retrospective study was conducted on 16 patients who underwent OS at San Pedro University Hospital of La Rioja. VSP was based on CT scans using Dolphin Imaging software (v12.0, Patterson Dental, St. Paul, MN, USA) and surgeries were guided by VSP-designed occlusal splints. Outcomes were assessed using the Orthognathic Quality of Life (OQOL) questionnaire and deviations were measured through pre- and postoperative imaging. The results showed high satisfaction scores and good overall outcomes, despite moderate deviations from the virtual plan in many cases, particularly among Class II patients. A total of 63% of patients required VSP modifications due to poor soft tissue fitting, with 72% of these being Class II DFDs. Most deviations involved less maxillary advancement than planned, while maintaining optimal occlusion. This suggests that VSP may overestimate advancement needs, especially in Class II cases. No significant differences in satisfaction were observed between patients with low (<2 mm) and high (>2 mm) deviations. These findings support the use of VSP as a valuable planning tool for OS. However, surgeon experience remains essential, especially in managing soft tissue behavior. Improvements in soft tissue prediction are needed to enhance accuracy, particularly for Class II DFDs. Full article

Surgery remains a healthcare intervention with significant risks for patients. Novel technologies can now enhance the peri-operative workflow, with artificial intelligence (AI) and extended reality (XR) to assist with pre-operative planning. This review focuses on innovation in AI, XR and imaging for hepato-biliary surgery planning. The clinical challenges in hepato-biliary surgery arise from heterogeneity of clinical presentations, the need for multiple imaging modalities and highly variable local anatomy. AI-based models have been developed for risk prediction and multi-disciplinary tumor (MDT) board meetings. The future could involve an on-demand and highly accurate AI-powered decision tool for hepato-biliary surgery, assisting the surgeon to make the most informed decision on the treatment plan, conferring the best possible outcome for individual patients. Advances in AI can also be used to automate image interpretation and 3D modelling, enabling fast and accurate 3D reconstructions of patient anatomy. Surgical navigation systems utilizing XR are already in development, showing an early signal towards improved patient outcomes when used for hepato-biliary surgery. Live visualization of hepato-biliary anatomy in the operating theatre is likely to improve operative safety and performance. The technological advances in AI and XR provide new applications in pre-operative planning with potential for patient benefit. Their use in surgical simulation could accelerate learning curves for surgeons in training. Future research must focus on standardization of AI and XR study reporting, robust databases that are ethically and data protection-compliant, and development of inter-disciplinary tools for various healthcare applications and systems. Full article

Rehabilitation devices such as actuated lower-limb exoskeletons can provide essential mobility assistance for pediatric patients with gait impairments. Enhancing their control systems under conditions of user variability and dynamic disturbances remains a significant challenge, particularly in active-assist modes. This study presents a human-in-the-loop control architecture for a pediatric lower-limb exoskeleton, combining outer-loop admittance control with robust inner-loop trajectory tracking via a non-singular terminal sliding-mode (NSTSM) controller. Designed for active-assist gait rehabilitation in children aged 8–12 years, the exoskeleton dynamically responds to user interaction forces while ensuring finite-time convergence under system uncertainties. To enhance adaptability, we augment the inner-loop control with a twin delayed deep deterministic policy gradient (TD3) reinforcement learning framework. The actor–critic RL agent tunes NSTSM gains in real-time, enabling personalized model-free adaptation to subject-specific gait dynamics and external disturbances. The numerical simulations show improved trajectory tracking, with RMSE reductions of 27.82% (hip) and 5.43% (knee), and IAE improvements of 40.85% and 10.20%, respectively, over the baseline NSTSM controller. The proposed approach also reduced the peak interaction torques across all the joints, suggesting more compliant and comfortable assistance for users. While minor degradation is observed at the ankle joint, the TD3-NSTSM controller demonstrates improved responsiveness and stability, particularly in high-load joints. This research contributes to advancing pediatric gait rehabilitation using RL-enhanced control, offering improved mobility support and adaptive rehabilitation outcomes. Full article