Search Results (381)

Radiotherapy simulation is essential for accurately targeting tumors while preserving healthy tissue, ensuring treatment precision and safety. This study aimed to develop an integrated radiotherapy simulation system capable of automated segmentation, dose estimation, and collision detection within a virtual planning environment to enhance efficiency and reduce costs in radiotherapy treatment planning. The Point Transformer model was applied to organ point cloud data derived from CT medical imaging for automated segmentation. Farthest point sampling (FPS) was employed to downsample the data before training. To enhance the accuracy and anatomical fidelity of the AI-generated segmentation results, reconstruction and refinement algorithms, including k-d tree, outlier removal, marching cubes, and surface smoothing, were implemented. Beam penetration simulation with the ray casting algorithm was employed for correction-based dose estimation. A collision detection module was incorporated to identify potential machine–machine or machine–patient interactions. The entire workflow was executed within a Unity 3D-based virtual simulation environment. As a result, the Point Transformer model demonstrated high segmentation accuracy, achieving Dice scores of 93.86 $\pm$ 1.50% for single-organ and 91.86 $\pm$ 3.25% for multi-organ cases, surpassing the performance of PointNet++. Applying ray casting for the refined surface meshes generated through post-processing enabled accurate dose estimation with discrepancies of 3.5% (brain), 5.9% (liver), and 13.8% (lung) compared to a Pinnacle TPS. The proposed method provides a low-cost and adaptable solution that enables easy modification and further development, making it particularly suitable for widespread applications in radiotherapy research, education, and clinical workflow optimization. Full article

Background: Drug shortages represent a growing challenge to healthcare systems worldwide, affecting treatment continuity and patient outcomes. This study assessed the burden and perceived impact of drug shortages from both healthcare professionals’ and patients’ perspectives in a Saudi tertiary hospital. Methods: A cross-sectional survey was conducted in April 2025 at King Abdulaziz Medical City, Riyadh, Saudi Arabia. Convenience sampling was used to recruit healthcare professionals with at least two years of experience and adult outpatients. Structured questionnaires assessed shortage frequency, affected drug classes, perceived impacts, and management practices. The findings were descriptively analyzed and compared with the Saudi Food and Drug Authority (SFDA) national shortage data for the corresponding 12-month period. Results: A total of 230 healthcare professionals and 243 patients participated. Among healthcare professionals, 89.1% reported experiencing at least one drug shortage, with 38.3% encountering shortages more than ten times annually. Anti-infectives (36.5%) and analgesics (35.7%) were the most frequently reported classes. The most common response was prescribing alternative medications (77.4%), with 55.3% perceived as adequately effective and 30.8% as less effective. Delayed care was the most frequently reported consequence (44.0%). Among patients, 30.9% reported experiencing shortages, 46.7% reported some degree of health impact, and 28.1% incurred additional costs. Awareness and utilization of the SFDA reporting system were low in both groups. Comparison with SFDA data revealed discrepancies between hospital-reported and nationally reported shortages. Conclusions: Drug shortages were frequently reported and associated with perceived clinical and economic consequences. Gaps between hospital experiences and national reporting highlight limitations in current surveillance systems. Strengthening reporting mechanisms, communication, and supply chain coordination may improve the management of drug shortages. Full article

Increasing use of intrauterine devices (IUDs) makes effective pain management essential for high-quality care. However, gaps between patient and clinician pain perceptions, along with limited training in managing IUD-related acute pain, contribute to barriers in IUD use and patient–provider mistrust. This narrative review aims at (1) summarizing current IUD-related pain management, (2) evaluating the emphasis on pain education and management, including IUD insertion pain, in medical education, and (3) examining how gaps in pain education and management may affect clinicians’ preparedness to manage patients’ IUD-related pain. Relevant literature was identified through keyword searches across major databases and national organizations. Studies on IUD pain management approaches and pain education in undergraduate medical education (UME), graduate medical education (GME), and residency programs were included. The results section outlines the current IUD insertion pain-management options available and their limitations, alongside evidence of insufficient emphasis on pain education and its management in medical school and residency training. In conclusion, the study indicates that, although all pain-management approaches reduce perceived IUD-related pain, significant gaps in national guidelines, clinician–patient pain-perception non-alignment, and pain-education across UME and GME may be contributary to inconsistent clinical practices. A multi-pronged approach that includes strengthening pain-education throughout medical training could be one way to reduce these disparities by improving clinicians’ competence to manage pain effectively. Full article

Background: Echocardiography is a non-invasive test that is readily used to detect pulmonary hypertension associated with bronchopulmonary dysplasia (BPD-PH) and right ventricular failure (RVF). However, the most feasible, reproducible and accurate parameters to measure and use for guidance in addressing patient care have not been established and may differ between subspecialties. Methods: We surveyed members of the BPD Collaborative to determine how different care providers clinically evaluate infants for BPD-PH and RVF. Perceived challenges and obstacles that limit the utility of echocardiography are also reported. Results: Of the 108 survey respondents from ~45 centers, 55.6% were neonatologists, 18.5% were pediatric pulmonologists or pediatric intensive care physicians, 15.7% were pediatric cardiologists or pulmonary hypertension specialists, and 10.2% were other providers. Responses revealed discrepancies between specialists concerning the use of standard echocardiographic protocols and parameters that can be measured serially with relative ease, metrics that should be used to best define and distinguish the severity of BPD-PH or RVF, and parameter values that should be used to determine whether changes in PH-targeted medical therapies, hemodynamic or respiratory support are needed. Free text responses identified patient-, protocol-, cardiology-, technician-, and BPD-PH definition-related obstacles that may limit the reliable utility of echocardiography. Conclusions: Although most providers agree that echocardiography is feasible and of value, variability exists between subspecialists and centers, suggesting the need for improved standardization of imaging protocols and BPD-PH definition, consistent test interpretation, and effective communication of results to improve the reproducibility and accuracy of echocardiography in infants with BPD. Full article

Cross-domain few-shot learning (CD-FSL) remains challenging in medical imaging, where labeled data are scarce and source–target domain gaps are often large due to modality differences. In particular, existing few-shot learning methods rely on source–target domain similarity, which limits their effectiveness in cross-modality settings such as MRI-to-CT transfer. To address this problem, this paper proposes an adapter-based Vision Transformer framework for cross-domain few-shot brain tumor classification. Lightweight adapter modules are inserted into a pretrained Vision Transformer to enable parameter-efficient domain adaptation without fine-tuning the entire backbone. In addition, a Prototypical Network is employed to construct class prototypes from limited labeled samples, while a prototype-level Maximum Mean Discrepancy (MMD) loss is introduced to align feature distributions across domains. Unlike prior approaches, the proposed framework introduces a unified prototype-level alignment strategy within an episodic learning paradigm, enabling direct class-wise cross-modal alignment. This design improves generalization under large modality gaps and limited labeled data by jointly optimizing representation learning and domain adaptation. The proposed framework is evaluated on MRI-to-CT brain tumor classification as well as several heterogeneous cross-domain benchmarks, including Chest X-ray, ISIC, CropDisease, and EuroSAT. Experimental results demonstrate that the proposed method achieves competitive performance compared to existing few-shot learning baselines, showing strong robustness under significant domain shifts. Full article

Multivariate time series anomaly detection holds critical application value in key domains such as industrial system monitoring, financial risk management, and medical surveillance. However, existing approaches face two major challenges: reconstruction-based or prediction-based models tend to adapt to anomalous patterns during training, thereby weakening the distinction between normal and abnormal samples; furthermore, the non-stationary nature of time series leads to distribution shifts between training and testing data, impairing model generalization. To address these issues, this paper proposes the TFCID model. The model innovatively leverages diffusion principles to effectively impute missing time series data while capturing significant frequency-domain features. In the temporal processing stream, an unconditional diffusion model combined with imputation masking is employed to achieve high-precision imputation of randomly missing values, effectively preventing anomalies from interfering with model training. In the frequency-domain processing stream, an amplitude-aware frequency-domain masked autoencoder is introduced to specifically capture periodic or trend-based pattern anomalies. The model mitigates distribution shift by constraining the discrepancy between temporal and frequency-domain representations via adversarial contrastive learning, and uses this discrepancy as a robust anomaly scoring metric. Experimental results on five public benchmark datasets show that TFCID significantly outperforms state-of-the-art methods in detection accuracy (F1-Score), validating its effectiveness in anomaly detection tasks. Full article

Background: In 2022, the German healthcare system generated 400,000 tons of waste. Reducing this number could lower greenhouse gas emissions. The waste management plan at the University Medical Center Regensburg, and those of other comparable German facilities, require that glass, cardboard/paper, residual waste, and other non-hazardous materials are collected separately. Objectives: To assess the personal interest, proficiency, opinion, and awareness of waste management among German dermatology staff to develop customized, resource-saving process optimization and training programs. Methods: An online cross-sectional survey was conducted among German dermatology healthcare professionals between 27 February and 4 October 2024. Out of the 100 responses, 84 were complete and subsequently analyzed. Respondents included staff at dermatology wards, outpatient units, and private practices. Data were analyzed descriptively; comparisons were made between clinics and outpatient units, and correlations were identified among the items. Results: Most respondents perceived the amount of waste generated during wound dressing changes as high; more than 60% expressed an interest in receiving further training on sustainability and waste reduction. Although many respondents reported having a good understanding of waste separation, they identified time pressure and stress as the two main obstacles to consistent implementation. Higher self-reported knowledge did not correspond with greater confidence in recycling as an effective waste reduction measure. Conclusions: The findings suggest a discrepancy between awareness and practice regarding sustainable waste management in dermatology. Combining structural and organizational measures with targeted training and workflow optimization could promote more sustainable clinical practices. Full article

Background/Objectives: The increasing prevalence of foot and ankle procedures in the United States has coincided with rising costs of care, exposing socioeconomic disparities within this area of orthopedics. Insurance payor status is one such factor that can affect the quality and accessibility of medical care and has been implicated in orthopaedic patient outcomes. While previous studies in other fields of orthopaedics have demonstrated an association between insurance status and access to treatment, length of hospital stay, post-operative outcomes and complication rates, no comprehensive review has yet explored this relationship in foot and ankle surgery. Thus, the goal of this study is to examine the association between insurance payor status and outcomes in foot and ankle procedures. Methods: A systematic review of five databases was conducted, focusing on the interplay between insurance coverage and foot/ankle procedures. Included studies reported on insurance payor status and patient outcomes following foot and ankle surgery. Extracted outcomes included time to be seen by a provider, complication and revision rates, post-operative Emergency Department (ED)/Urgent Care utilization, readmission rates, hospital length of stay, pain, functional scores, discharge destinations, return to work/activity, and follow-up. Meta-analyses were performed using the Revman 5.3 software. Studies that did not qualify for meta-analyses were described qualitatively. Results: Of 1401 studies identified, 24 texts met inclusion and exclusion criteria. Across the 24 studies, there were a total of 20,950 patients. Noninsured patients had a 59% higher risk of ED/urgent care utilization within 30 days of surgery compared to insured patients [Risk Ratio (RR) = 1.59, 95% Confidence Interval (CI) = 1.18 to 2.12, p < 0.05]. Privately insured patients were seen 3.65 days earlier than patients with government insurance [95% CI = 2.02 to 5.27, p < 0.0001]. Worker’s Compensation patients had statistically significant findings for poorer outcomes, higher pain scores, and lower functional scores. Similarly, Medicaid patients also fared worse on functional scores and had delayed access to appointments and treatments. Conclusions: Patients without private insurance have worse pain and functional outcomes, delayed access to care, and increased utilization of emergency resources following foot and ankle procedures. It is crucial for providers to be cognizant of these discrepancies when caring for patients. Further research is needed to better understand the nuances of these insurance-related disparities within foot and ankle subspecialties. Full article

Background/Objectives: Mental health conditions, including stress, anxiety, depression, and sleep problems, represent a significant health concern globally. Mounting evidence suggests a link between mental health and the gut microbiome via the gut–brain axis. However, discrepancies in human microbiome data exist due to the heterogeneity in study design and analytical approaches. Thus, this study aimed to explore the gut microbial characteristics associated with self-reported mental health symptoms using multiple analytical methods. Methods: A total of 44 healthy adults, defined as individuals without any major chronic medical conditions, were assessed for mental health symptoms using self-reported questionnaire data. To evaluate gut microbial characteristics, stool samples were collected at six time points over 28 days and underwent 16S rRNA gene sequencing. Differential abundance was assessed via ANCOM-BC, and a random forest classifier was implemented to rank features important for the classification of mental health symptoms. Participants who did not report anxiety, stress, depression, or sleep problems served as the reference group for microbiome comparisons. Results: The proportion of participants with self-reported mental health symptoms was 11.4% (stress), 27.3% (depression), 31.8% (anxiety), and 15.9% (sleep problems). Participants reporting mental health symptoms showed differences in gut microbiome composition compared to asymptomatic participants, including variation in alpha- and beta-diversity. Differential analysis identified specific taxa with higher or lower relative abundance in participants reporting specific mental health symptoms. Random forest feature ranking identified partially overlapping taxa across methods, suggesting candidate associations warranting further investigation. Conclusions: These exploratory findings suggest that self-reported mental health symptoms in otherwise healthy adults are associated with differences in the gut microbiome. The taxa identified in this study represent candidates for validation in larger, independent cohorts. Full article

Background: This scoping review aimed to identify gaps in the literature regarding medication therapy problems (MTPs) among hospitalized adults with decreased kidney function. Specifically, it aimed to answer the following questions: (1) What types of MTPs have been reported? (2) What is the reported prevalence of MTPs? (3) Do MTPs differ by type of kidney disease? (4) What gaps exist regarding MTPs and pharmacists’ involvement? Methods: Studies involving adult patients with decreased kidney function that investigated MTPs were included. Studies focused exclusively on post-transplant care, chemotherapy, or a single MTP type were excluded. Literature searches were conducted in PubMed, EMBASE, Cochrane Library, Web of Science, and International Pharmaceutical Abstracts. Two independent reviewers screened and extracted data, with a third reviewer resolving discrepancies. All identified MTPs were re-categorized using the Pharmacy Quality Alliance (PQA) framework and the Pharmaceutical Care Network Europe (PCNE) classification. Results: A total of 23 studies met the inclusion criteria, including two conference proceedings, encompassing 7151 patients. The most common MTP framework was the PCNE classification (13 studies, 57%). Reclassification using the PQA yielded 10,596 MTPs, most frequently “Safety—dosage too high” (n = 2464) and “Effectiveness—dosage too low” (n = 2262). Reclassification using the PCNE yielded 11,574 MTPs, most frequently “Drug selection” (n = 6974) and “Dose selection” (n = 2636). All studies involved patients with chronic kidney disease (CKD), and two also included acute kidney injury (AKI). Conclusions: Dosage-related MTPs were most prevalent among hospitalized patients with decreased kidney function. Variability in MTP definitions, limited representation of patients with AKI and AKD, and minimal reporting on pharmacists’ roles reveal important gaps. Addressing these gaps through standardized MTP classification and further research in understudied kidney disease populations may enhance patient safety and support clinical pharmacists’ contributions to optimizing medication safety across the kidney disease continuum. Full article

Background/Objectives: Large language models (LLMs) are increasingly used as decision support tools in clinical nutrition, including meal planning for individuals with type 2 diabetes mellitus (T2DM). However, the clinical safety, quantitative accuracy, and guideline adherence of AI-generated dietary plans remain uncertain. This study aimed to evaluate systematic bias and agreement between LLM-generated diets and a guideline-concordant reference diet, and to assess whether current LLMs can function as reliable clinical nutrition decision support tools in T2DM. Methods: Six widely used LLMs generated standardized three-day, 1800 kcal dietary plans for T2DM using an identical prompt. Each day was treated as an independent observation (n = 18). Energy and macronutrient contents were analyzed using professional nutrition software and compared with a dietitian-designed reference diet based on ADA, EASD, IDF, and national guidelines. Agreement was evaluated using Bland–Altman analysis, proportional bias assessment, and intraclass correlation coefficients. Guideline adherence and clinical appropriateness were independently scored by registered dietitians. Results: Most LLM-generated diets systematically deviated from the reference diet, with lower total energy, reduced carbohydrate and fiber content, and variable protein distribution. Bland–Altman analyses demonstrated significant bias and wide limits of agreement for key nutrients, indicating clinically meaningful discrepancies. Guideline adherence scores varied substantially across models, with only one model showing relatively consistent performance. Inter-rater reliability between dietitians was high (ICC = 0.806). Conclusions: Current LLMs exhibit systematic quantitative bias and inconsistent guideline adherence when used for T2DM meal planning. AI-generated dietary plans are not interchangeable with dietitian-guided medical nutrition therapy and may pose clinical risks if used without professional oversight. Careful validation, domain-specific fine-tuning, and integration within supervised clinical workflows are required before implementation in diabetes care. Full article

Pharmacotherapy is a key component of obesity management, yet treatment failure remains a prevalent challenge in clinical practice. Such failure may present as insufficient pharmacological response, early discontinuation, or post-treatment weight regain, underscoring the discrepancy between clinical trial efficacy and real-world outcomes. The effectiveness of anti-obesity medications (AOMs) is influenced by psychiatric comorbidities, including depression, anxiety, and disordered eating patterns, as well as environmental and socioeconomic factors such as limited healthcare access, weight-related stigma, and high medication costs. Individual characteristics, including physical activity, body composition, visceral adiposity, and microbiome profile, further modulate treatment outcomes. Pharmacokinetic and pharmacotherapeutic limitations such as drug-phenotype mismatch, route of administration, suboptimal formulations, and exposure to counterfeit products also compromise efficacy. No less important are genetic and immunological factors, comprising pharmacogenomic variants of both incretin and melanocortin receptors along with antidrug antibodies (ADAs), which may constitute therapy resistance. Concomitant medications and comorbid endocrine disorders can additionally attenuate weight-loss effects. The objective of this review is to characterize the multifactorial nature of resistance and refractoriness to anti-obesity therapy, and the importance of identifying pretreatment predictive factors for recognizing individuals at risk of inadequate or lack of response, thereby enabling personalized management strategies and improving long-term clinical outcomes, particularly in “difficult-to-treat” patients. Full article

Background: Tetrasomy X (48, XXXX) is an extremely rare sex chromosome aneuploidy characterized by highly variable phenotypic manifestations. It includes various medical issues, a wide range of developmental delays, and neurocognitive deficits. Methods: The present case report provides a comprehensive neurodevelopmental profile of a 4.5-year-old girl with Tetrasomy X, with the aim of contributing to phenotype delineation, exploring genotype–phenotype associations, and emphasizing the importance of early, targeted intervention. A multidisciplinary assessment was conducted, encompassing cognitive, speech–language, motor, sensory, adaptive, and socioemotional functioning, using a battery of standardized and culturally adapted instruments. Results: Results revealed borderline intellectual functioning and mild global developmental delay, with marked intra-individual variability across domains. Motor development was significantly delayed and speech and language assessment demonstrated a pronounced receptive–expressive discrepancy. Sensory processing evaluation revealed a pattern of global sensory under-responsiveness, representing a novel and underreported feature in Tetrasomy X. Adaptive functioning was uneven, with relative strengths in daily living skills and weaknesses in motor abilities. Conclusions: This detailed early developmental characterization highlights the heterogeneity of Tetrasomy X and challenges some of the previous assumptions. The findings underscore the necessity of individualized, multidisciplinary assessment and early intervention to optimize developmental outcomes and quality of life in affected individuals. Full article

Background: Large language models (LLMs) and vision-language models (VLMs) have recently been applied to ophthalmology for patient education, diagnosis, and surgical decision support. Their ability to generate, interpret, and synthesize medical information positions them as promising assistive tools in glaucoma care. This scoping review aims to consolidate current evidence on the applications of LLMs and VLMSs in glaucoma, summarizing their tasks, inputs, performance metrics, and limitations to guide future clinical and research developments. Methods: A systematic search was conducted in PubMed, Scopus, Web of Science, arXiv, and IEEE Xplore from 2014 to July 2025. Eligible studies included original research and research letters employing LLMs or VLMs/MM-LLMs in any glaucoma-related application, including diagnostic reasoning, image interpretation, patient education, or surgical decision support. Screening and full-text review were independently performed by two reviewers following PRISMA-ScR methodology, with discrepancies resolved by consensus. Results: In total, 316 records were identified across five databases, with 27 studies meeting the inclusion criteria. The selected studies focused on three main domains: patient education (n = 11), diagnosis and risk prediction (n = 10), and surgical management (n = 6). Conclusions: Current LLMs serve best as assistive rather than autonomous tools in glaucoma care. They demonstrate strong potential in patient communication and text-based clinical decision support but remain constrained by variable accuracy, limited multimodal integration, and a lack of ophthalmology-specific fine-tuning. Future research should focus on developing domain-trained and retrieval-augmented LLMs, enhancing multimodal (text-image) fusion, ensuring readability adaptation for patients, and establishing ethical and regulatory frameworks for clinical implementation. Full article

Background/Objectives: The study primarily examined whether an IoT-based medication assistance system enhances ART adherence relative to standard care, and secondarily evaluated device feasibility and error patterns over time. Methods: This prospective study was conducted between June 2022 and October 2023 at a tertiary hospital in South Korea. Adults (≥19 years) living with HIV and prescribed ART were included; those with comorbid hepatitis B or C were excluded. People living with HIV who agreed to use the IoT-based InPHRPILL system (Sofnet Inc., Seoul, Republic of Korea) were assigned to the intervention group, whereas those who declined were assigned to the control group. Viral suppression, CD4⁺ cell counts, and adherence rates were measured. Additional analyses evaluated 12-month longitudinal adherence using pill-count data in both groups, and device-measured adherence and device-associated error rates in the intervention group. Results: Thirty-five participants (12 in the intervention group and 23 in the control group) were included. The intervention group demonstrated marginally shorter durations since HIV diagnosis and ART initiation at study enrollment, as well as slightly higher baseline HIV-RNA levels; however, these differences did not reach statistical significance. The median pill-counting and IoT device adherence rates were 100% and 87.4%, respectively (median deviation error rate = 4.4%). Poisson regression revealed significantly reduced error rates over time (β = −0.06493, p < 0.01), suggesting improved device use proficiency. Conclusions: IoT-based medication assistance systems may provide objective, real-time monitoring of ART adherence and facilitate identification of discrepancies between clinical evaluations and actual adherence patterns. Larger studies targeting individuals with suboptimal adherence are warranted to determine whether such systems can enhance adherence outcomes. Full article