SEARCH · Search PubMed

Results for “precision transfusion intelligence”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

253 records · Page 4Linked to original sources

PubMed2026

Automated CEAP Classification of Venous Duplex Reports Using Multimodal Artificial Intelligence.

OBJECTIVE: To develop and internally validate a prototype multimodal artificial intelligence system for automated CEAP (Clinical, Etiological, Anatomical and Pathophysiological) classification of venous duplex ultrasound (VDUS) reports, integrating natural language processing of free-text components with computer vision analysis of hand-drawn anatomical diagrams. METHODS: Single centre retrospective observational study using routinely collected clinical data. One thousand consecutive venous duplex ultrasound reports from Cambridge University Hospitals NHS Foundation Trust, UK (July 2024 - May 2025) were labelled according to the CEAP classification, excluding the Etiological component, which could not be reliably determined from duplex reports alone. Transfer learning was applied using ClinicalBERT for text and MobileNetV3 for diagrammatic data. Clinical classes were predicted from request line text. Text- and image-based pathophysiological models were developed for four anatomical territories (Great Saphenous Vein, Small Saphenous Vein, Deep system, Perforators), combined using late fusion with probability averaging. RESULTS: The clinical CEAP model achieved accuracy of 0.91, macro-F1 of 0.82, and macro-AUC of 0.98. Pathophysiological prediction varied, with text models broadly outperforming image models. Fusion yielded heterogeneous benefits, improving SSV performance but reducing Deep system accuracy. The performance of the final pathophysiological CEAP fusion models varied across anatomical territories: accuracy ranged from 0.70-0.92 and macro-AUC from 0.80-0.92. CONCLUSION: This study demonstrates the feasibility of automated CEAP classification from VDUS reports. Despite class imbalance affecting minority class predictions, the strong discriminatory performance validates this multimodal ML model for extracting clinically meaningful information from real-world data. This approach offers potential, pending external validation, to streamline vascular services through automated triage and guideline-compliant decision making.

Artificial intelligence↗

PubMed2026

Artificial Intelligence Cannot Replace Peer Reviewers but May Help Editors Triage: A Comparative Analysis of a Large Language Model and Human Reviewer Recommendations at the American Journal of Sports Medicine.

BACKGROUND: The peer review system faces increasing strain from rising manuscript volumes, reviewer fatigue, and well-documented interreviewer disagreement. Large language models (LLMs) have shown potential to support the peer review process, but their ability to replicate editorial decisions at high-impact medical journals and their utility as manuscript screening tools remain unknown. PURPOSE: To compare the agreement between an LLM and the final editorial decision on manuscripts submitted to the American Journal of Sports Medicine and to evaluate the potential of LLMs as a manuscript screening tool. STUDY DESIGN: Cross-sectional agreement study. METHODS: Fifty-four manuscripts randomly selected from submissions to the American Journal of Sports Medicine (September 2024-October 2024) were reviewed by a locally deployed LLM (Ministral 3 14B; Mistral AI) using a standardized prompt. The artificial intelligence (AI) produced a categorical recommendation (reject, cascade, revision, or accept) and a numerical score (0-100) for each manuscript. Agreement with the final editorial decision was assessed by Cohen kappa (4-category model) for pooled human reviewers (n = 139 reviews) and the AI (n = 54). Screening performance was evaluated by positive predictive value (PPV), sensitivity, and specificity. RESULTS: Pooled human reviewers demonstrated fair agreement with the final decision (κ = 0.181 [P < .001]; 42.4% agreement), while the AI demonstrated slight, nonsignificant agreement (κ = 0.126 [P = .099]; 37.0% agreement). The AI recommended revision for 61.1% of manuscripts, of which 72.7% were ultimately rejected or cascaded, demonstrating systematic "revision bias." When the AI recommended rejection, 54.5% of those manuscripts were ultimately rejected and 27.3% were cascaded; when the AI recommended cascade, 50% were rejected and 50% were cascaded. However, when the AI recommended rejection or cascade (n = 21), 90.5% received a final decision of rejection or cascade (PPV, 90.5%; specificity, 81.8%). Manuscripts with an AI score <70 were rejected or cascaded 88.0% of the time (PPV, 88.0%). CONCLUSION: AI cannot replicate the nuanced judgment of human peer reviewers at a high-impact sports medicine journal. When AI recommended rejection or cascade, 90.5% of manuscripts received that final decision (descriptive PPV, 90.5%; 95% CI, 71.1%-97.3%), suggesting potential utility as an exploratory first-pass screening tool warranting further validation in larger cohorts. However, AI could not reliably distinguish manuscripts destined for outright rejection from those that would be cascaded to a sister journal-an important limitation for editorial triage applications.

Sports Medicine↗

PubMed2026

The impact of artificial intelligence on critical thinking and clinical reasoning in health professions education: A systematic review and meta-analysis.

BACKGROUND: Critical thinking and clinical reasoning underpin healthcare professionals' ability to navigate uncertainties and deliver safe and effective care. With artificial intelligence (AI) advancement and growing adoption, AI-based educational tools are increasingly used to support these cognitive competencies' development. OBJECTIVE: To synthesize randomised and controlled clinical trials on AI-based educational tools in health professions education and examine their effects on critical thinking and clinical reasoning among health professions students. METHODS: Six electronic databases were searched from January 1, 2014 to July 28, 2025 was reviewed: PubMed, Cochrane Central Register of Controlled Trials, CINAHL, Scopus, Embase and Web of Science. Two independent reviewers performed data extraction and quality assessment using standardized JBI checklists. The GRADE approach was used to assess the certainty of evidence. Studies were pooled via random-effects meta-analyses or narrative syntheses. RESULTS: Fourteen randomised controlled trials and seven controlled clinical trials were included (n = 21). Meta-analyses revealed small to medium effect sizes for the surrogate clinical reasoning outcomes of performance-based assessment scores (SMD 0.68; 95% CI [0.38, 0.98], p-value = 0.00; I2 = 38%) and knowledge test scores (SMD 0.39; 95% CI [0.09, 0.69], p-value = 0.01; I2 = 79%). Critical thinking and clinical reasoning skills and dispositions were narratively synthesized, with majority of included studies favouring AI-based interventions but the evidence had low to very low certainty. CONCLUSION: AI-based educational interventions may improve critical thinking and clinical reasoning among health profession students, but the evidence is very uncertain. This review offers preliminary insights but does not allow identification of optimal interventions or discipline-specific recommendations due to small sample sizes and substantial intervention heterogeneity. Further research is required to draw definitive conclusions. PROTOCOL REGISTRATION: CRD42025634074.

Humans↗

PubMed2026

Proteomics in environmental pollution research: Advances, challenges, and future directions.

Environmental proteomics has emerged as a powerful approach for elucidating the molecular mechanisms underlying pollutant-induced biological effects. Although this field has developed rapidly, the systematic review of recent proteomics applications in environmental pollution research remains limited. This review explored the emerging roles of toxicoproteomics in biomarker discovery and mechanistic elucidation, as well as ecotoxicoproteomics in ecological risk assessment and bioremediation strategies. Here, we review the field, highlighting recent trends such as the integration of proteomics with genomics, transcriptomics, and metabolomics to provide a comprehensive view of biological responses to environmental stressors. We further discuss the growing application of artificial intelligence in improving proteomics data interpretation and accelerating biomarker discovery. In addition, recent technological advances in environmental proteomics are highlighted, including next-generation tissue microarray proteomics, nanoscale proteomics, single-cell proteomics, and spatial proteomics. Despite its potential, proteomics faces challenges, such as high operational costs, computational complexity in analysis, and technical limitations in low-abundance protein detection. We propose that the convergence of proteomics with artificial intelligence and multi-omics approaches offers promising solutions to these challenges, enhancing the practical application of proteomics in environmental monitoring and risk assessment.

Proteomics↗

PubMed2026

Artificial intelligence (AI) uses in stereotactic radiosurgery (SRS): diagnosis with brain metastasis (BM) - A systematic review.

BACKGROUND: Brain metastases (BM) are the most common intracranial tumors in adults, and stereotactic radiosurgery (SRS) has become a mainstay of management. However, several diagnostic challenges persist in the SRS pathway, particularly the differentiation of radiation necrosis (RN) from true tumor progression, which conventional MRI and even advanced imaging techniques often cannot reliably resolve. Recent advances in artificial intelligence (AI) offer the potential to address these diagnostic limitations. This systematic review synthesizes current literature on AI applications for MRI-based diagnostic decision support in BM patients undergoing SRS, with a focus on radiomics and deep learning tools for distinguishing RN from progression, classifying molecular and histologic subtypes, and predicting treatment response. METHODS: A systematic review was performed in accordance with PRISMA guidelines. PubMed, Web of Science, and Scopus were searched using a targeted query combining terms related to AI, brain metastasis, diagnosis or imaging, and SRS. After screening 483 records and applying strict inclusion and exclusion criteria, 18 studies published between 2015 and 2025 were included. Data were extracted on study design, cohort characteristics, imaging modality, AI methodology, validation strategy, and reported diagnostic performance. RESULTS: Among the 18 included studies, AI models demonstrated strong performance across diagnostic tasks in the BM-SRS pathway. The differentiation of RN from true tumor progression was the most extensively studied application, addressed by 14 of 18 studies, with reported AUCs ranging from 0.71 to 0.94. Support vector machines, random-forest ensembles, convolutional neural networks, and transformer-based multimodal architectures were widely used. The literature evolved from single-sequence radiomic classifiers in 2018 to multimodal deep learning frameworks fusing imaging with clinical and genomic data in 2025. Contrast-enhanced T1-weighted MRI was the dominant imaging input, and texture-based radiomic features (GLCM, GLSZM, GLDM, and wavelet-derived features) were the most consistently predictive. The highest-performing models reached AUCs of 0.85-0.91 through multimodal integration of imaging with clinical and genomic features, and consistently outperformed expert neuroradiologist read on matched cases. Remaining studies addressed longitudinal segmentation-based detection of local failure and adverse radiation effects, BRAF mutation status in melanoma BM, early Gamma Knife treatment response, and primary tumor histology classification, with more variable performance. CONCLUSION: AI models, particularly those integrating MRI-derived radiomic features with clinical and genomic data, show high accuracy in supporting diagnostic decisions for BM patients treated with SRS. The post-SRS differentiation of radiation necrosis from true tumor progression has reached the greatest level of maturity and is closest to clinical translation, with potential to reduce unnecessary biopsies, personalize surveillance intervals, and rationalize treatment-pathway decisions. Other diagnostic applications, including molecular subtyping and primary tumor histology classification, remain exploratory and require further multicenter validation. Integration of AI tools into multidisciplinary tumor-board workflows, combined with prospective validation and standardized reporting, will be essential to realize the full clinical benefits of AI in SRS for brain metastases.

Humans↗

PubMed2026

Psychological consequences of AI-assisted training and the buffering role of mindfulness.

The integration of artificial intelligence (AI) into athletic training is accelerating, yet its psychological implications for athletes remain insufficiently understood. Drawing on the transactional model of stress and the stress-buffering framework of mindfulness, this study examined whether mindfulness training can mitigate adverse psychological responses associated with AI-assisted training. Using a randomized controlled factorial design, 160 collegiate athletes were assigned to AI-assisted training or standard training, with or without concurrent mindfulness intervention, and assessed at baseline, week 4, and week 8. Athletes exposed to AI-assisted training without psychological support exhibited increases in perceived stress and AI dependence over time. In contrast, these stress increases were substantially attenuated when mindfulness training was implemented alongside AI-assisted training. A significant AI × Mindfulness × Time interaction emerged for perceived stress at post-intervention, and difference-in-differences analyses corroborated a robust buffering effect. Mediation analyses further indicated that mindfulness training reduced stress partially through enhancing mindful awareness; a three-wave cross-lagged analysis showed that mindful awareness and stress were reciprocally related over time, with the hypothesized awareness-to-stress pathway remaining robust. Together, these findings suggest that AI-assisted training introduces a distinct form of evaluative pressure, and that mindfulness training may serve as an effective psychological buffer during the adoption of continuous algorithmic performance evaluation systems.

Humans↗

PubMed2026

Impact of Commercial Artificial Intelligence on Radiologist Reading Time for Pulmonary Nodule Evaluation at Chest CT.

Background Chest CT is a primary method for identifying pulmonary nodules, yet interpreting scans remains time-intensive and demanding. Currently, artificial intelligence (AI) is expected to reduce reading times, but the effect of AI on reporting times in this setting is unknown. Purpose To evaluate the impact of a commercial AI software on radiologists' reading time for pulmonary nodule assessment on chest CT scans within a real-world clinical setting. Materials and Methods This retrospective study included patients who underwent chest CT examinations at a tertiary medical center between September 2021 and May 2024. The study period was divided into pre- and post-AI phases. The primary outcome was radiology reporting time. The association between AI implementation and reporting time was evaluated using a multivariable parametric Weibull shared frailty survival model adjusted for reader function, examination type, patient location, and requesting specialty, with clustering at the radiologist level. Interaction analyses assessed heterogeneity across prespecified subgroups. An exploratory extrapolation estimated projected workforce and financial impact. Results This study included 19 433 patients (mean age, 62 years ± 14.2 [SD]; 21 814 men; 39 323 chest CT examinations, 19 190 pre-AI, and 20 133 post-AI). AI implementation was associated with faster report completion (adjusted hazard ratio, 1.17; 95% CI: 1.14, 1.21; P < .001). The adjusted median reporting time decreased from 21.3 minutes pre-AI to 18.2 minutes post-AI (14.6% reduction; P < .001). Heterogeneity was observed across reader function (P < .001), examination type (P = .048), and requesting specialty (P = .03). The largest relative reductions were observed for CT thorax electrocardiogram-gated examinations (-41.1%; P < .001) and thoracic radiologists (-25.0%; P < .001), whereas emergency department examinations showed increased median reporting time (7.1%; P < .001). At institutional scan volumes (approximately 20 000-22 000 chest CT examinations annually), exploratory modeling suggested an approximate reduction of 0.5 full-time equivalent radiologist workload. Conclusion Implementation of commercial AI-assisted pulmonary nodule assessment on chest CT scans reduced radiologist reporting time in a real-world clinical setting. © The Author(s) 2026. Published by the Radiological Society of North America under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Iwasawa in this issue.

Humans↗

PubMed2026

Electrospun Nanofiber Dressings for Diabetic Wounds: From Single-Layer to Intelligent Composite Systems.

Diabetic chronic wounds have become a major challenge for clinical treatment due to their complex pathological microenvironment, including persistent inflammatory response, angiogenesis disorder, excessive oxidative stress, and susceptible infection. Traditional dressings as a passive barrier have difficulty meeting the above multiple treatment needs. Electrospinning technology, with its ability to mimic the fibrous network structure of the natural extracellular matrix (ECM), offers a high specific surface area, controllable porosity, and excellent drug-loading capacity, making it an ideal platform for developing a new generation of multifunctional wound dressings. This article provides a systematic review of the research progress on electrospun nanofiber dressings in the treatment of diabetic wounds, focusing on the design evolution from basic single-layer structures to advanced complex structures and elucidating the mechanisms of action and quantifiable effects of each structural type in addressing specific pathological challenges. We also compared the current status of clinical translation for electrospun dressings with that of other advanced wound care platforms and proposed a standardized preclinical evaluation framework. A large number of research data show that these advanced designs can effectively improve the quality of healing. Finally, this paper points out the challenges faced by this field, such as scalable fabrication, in vivo reliability of smart systems, and long-term biosafety, and provides theoretical basis and technical reference for the design of efficient and intelligent electrostatic spinning diabetic wound dressings.

Nanofibers↗

PubMed2026

Assessing AI literacy and attitudes among medical students: implications for integration into healthcare practice.

PURPOSE: This study aims to assess AI literacy and attitudes among medical students and explore their implications for integrating AI into healthcare practice. DESIGN/METHODOLOGY/APPROACH: A quantitative research design was employed to comprehensively evaluate AI literacy and attitudes among 374 Lusaka Apex Medical University medical students. Data were collected from April 3, 2024, to April 30, 2024, using a closed-ended questionnaire. The questionnaire covered various aspects of AI literacy, perceived benefits of AI in healthcare, strategies for staying informed about AI, relevant AI applications for future practice, concerns related to AI algorithm training and AI-based chatbots in healthcare. FINDINGS: The study revealed varying levels of AI literacy among medical students with a basic understanding of AI principles. Perceptions regarding AI's role in healthcare varied, with recognition of key benefits such as improved diagnosis accuracy and enhanced treatment planning. Students relied predominantly on online resources to stay informed about AI. Concerns included bias reinforcement, data privacy and over-reliance on technology. ORIGINALITY/VALUE: This study contributes original insights into medical students' AI literacy and attitudes, highlighting the need for targeted educational interventions and ethical considerations in AI integration within medical education and practice.

Students, Medical↗

PubMed2026

Artificial intelligence enabled social robotic interventions (PARO) in Australian dementia care: A systematic review and meta-analysis.

BACKGROUND: Although there is a growing body of research indicating that Personal Robot/Social Robot could be used in various aspects of care for individuals with dementia, little is known about how well these types of interventions work in an actual hospital setting in Australia. AIMS & OBJECTIVES: The objective of the present systematic review and meta-analysis is to assess the effectiveness of PARO-based socially assistive robotic intervention in terms of its effectiveness outcomes towards the reduction of dementia-related behavioural and psychological symptoms in Australian based healthcare settings. METHODS: A systematic search was conducted across five electronic databases, including MEDLINE (PubMed), EMBASE, CINAHL, PsycINFO, and the Cochrane Library, to identify randomised controlled trials (RCTs) investigating PARO-based socially assistive robotic interventions for dementia in Australian healthcare settings. This review was registered with PROSPERO (CRD420251251916) and followed the PRISMA 2020 guidelines. In addition, the Cochrane Risk of Bias tool (RoB 2) was used to evaluate the risk of bias across all studies. Pooled standardised mean differences (SMD) with 95 % confidence intervals (CI) were calculated for agitation, anxiety, and depression. Heterogeneity across studies was evaluated using the I2 statistic. RESULTS: Six RCTs involving 1444 participants were identified for inclusion in this review. AI-enabled socially assistive robotic interventions, specifically the PARO therapeutic robot, significantly reduced agitation and anxiety when compared to standard treatment or control conditions. The pooled analysis showed that agitation [SMD = -0.44 (95 % CI: -0.70, -0.18) p = 0.0008] and anxiety [SMD = -0.59 (95 % CI: -0.91, -0.27) p = 0.0003] were reduced significantly, while the decrease in depression [SMD = -0.44 (95 % CI: -0.95, -0.07) p = 0.09] scores was non-significant among dementia patients receiving PARO-based socially assistive robotic interventions as compared to the control. The overall risk of bias across all six studies was considered low to moderate. CONCLUSION: PARO-based socially assistive robotic interventions may provide preliminary evidence of effectiveness in reducing agitation and anxiety in individuals with dementia in Australian healthcare, but the evidence regarding the reduction of depression remains unclear. Therefore, additional high-quality trials with consistent methodology and extended follow-up will be necessary to determine both the short-term and long-term clinical efficacy and practicality of implementing these interventions into practice.

Humans↗

PubMed2026

Dynamic lysine acetylation and succinylation of platelet proteins regulates platelet storage lesion: mechanistic insights from multi-omics.

OBJECTIVES: Platelet storage lesion (PSL) severely impairs platelet function during storage, presenting a major hurdle in transfusion medicine; however, the dynamic interplay between global proteomic changes and post-translational modifications (PTMs) underlying these functional deteriorations remains insufficiently characterized. Here, we report the first comprehensive multi-omics analysis integrating global proteomics, acetylomics, and succinylomics to dissect the molecular dynamics during platelet storage. METHODS: We performed quantification of global proteomics, acetylome and succinylome based on TMT-labeled LC-MS/MS analysis, combined with antibody-affinity enrichment and purification. Dynamic molecular changes and functional transformation of platelet were also characterized under proper conditions stored for 1, 3, 5, 7 days, respectively. RESULTS: We systematically characterized 3,609 proteins, 1,308 acetylation sites, and 1,947 succinylation sites across multiple storage time points (D1, D3, D5, D7). We distinct temporal patterns of post-translational modifications, with succinylation showing more extensive coverage than acetylation in platelets. Pathway enrichment analysis revealed extensive metabolic reprogramming involving complement activation, energy metabolism, and cellular detoxification processes. The identification of specific motif patterns provided mechanistic insights into the functional specificity of these modifications. Random forest machine learning identified 20 core regulatory proteins representing critical nodes in PSL development. Furthermore, we employed real - time quantitative polymerase chain reaction (RT - QPCR) to measure the expression levels of key genes related to platelet function and PTM - associated pathways. CONCLUSION: By mapping the interplay between proteomic abundance shifts and PTM dynamics, this study provides a multidimensional understanding of PSL, establishing a foundational framework for optimizing storage protocols and enhancing transfusion safety.

Blood Platelets↗

PubMed2026

Applications of quantum AI in brain disorder diagnosis: A systematic review.

BACKGROUND AND OBJECTIVE: Brain disorder diagnosis and prediction remain challenging because neuroimaging, electrophysiological, behavioral, and multimodal data are high-dimensional, noisy, heterogeneous, and limited by small clinical cohorts. This systematic review synthesised applications of quantum artificial intelligence (QAI) for brain disorder diagnosis, prediction, detection, and monitoring. METHODS: Following PRISMA guidelines, studies published from 2016 to 13 January 2026 were retrieved from Scopus, Web of Science, and IEEE Xplore. After screening, 36 studies met the eligibility criteria and were qualitatively analysed according to disorder category, data modality, QAI method, implementation setting, validation strategy, and performance. RESULTS: At the broader disease-group level, neurodegenerative disorders were the most frequently investigated, followed by mental health and psychiatric disorders. At the individual level, Parkinson's disease and schizophrenia were the leading applications, followed by depression, anxiety, Alzheimer's disease, and stress-related tasks. MRI-based modalities were the most frequently used data source, followed by multimodal data and EEG. Methodologically, primary QAI approaches were dominated by quantum neural and QDL architectures, followed by quantum-inspired optimization or feature-selection methods and quantum-kernel/conventional QML classifiers. Qiskit/IBM Quantum and PennyLane were the most frequently reported quantum software frameworks. However, most studies relied on simulators, classical quantum-inspired implementations, or unclear implementation settings, with limited real-hardware evaluation. CONCLUSIONS: QAI shows emerging potential for brain disorder analysis, particularly through hybrid quantum-classical learning, quantum neural architectures, quantum-kernel methods, and quantum-inspired optimization. Nevertheless, current evidence remains preliminary and requires larger datasets, subject-level and external validation, fair classical benchmarking, noise-resilient circuits, real quantum hardware evaluation, explainability, and clinical validation.

Humans↗

PubMed2026

A Sentiment-Based Comparison of AI- and Physician-Generated Empathic Statements in Palliative Care.

CONTEXT: Empathic communication promotes trust in patient-provider relationships. As healthcare integrates artificial intelligence (AI) into patient communication, we have yet to understand how these models' communication compares to that of physicians. OBJECTIVES: Our primary objectives were to examine patient preferences for AI-generated vs. palliative care physician-generated empathic statements addressing fear and anxiety around cancer treatment, and to analyze associations between linguistic features and patient preferences. METHODS: We conducted a secondary analysis of the PALL-AI trial, a randomized controlled survey comparing cancer patients' preferences of AI- to physician-generated empathic statements. Physicians and AI were provided the same prompt with a maximum sentence length. Patient preferences for each statement were measured in blinded surveys. We analyzed sentiment of the statements using the Valence Aware Dictionary and Sentiment Reasoner (VADER) and the National Research Council Canada (NRC) Emotion Lexicon. We evaluated associations between sentiment scores and patient preferences using Spearman's correlation coefficients. RESULTS: A total of 105 patients completed blinded surveys, preferring the AI-generated statement 72.4% of the time. VADER sentiment analysis showed all three AI statements displayed positive sentiment, while all three physician statements displayed negative sentiment. Controlling for statement length, AI statements used twice as many positive words as human statements. However, they contained a similar number of negative words. Of the eight NRC emotions, "trust" and "joy" demonstrated the strongest correlations with patient preference. CONCLUSION: Patients preferred AI-generated statements around cancer care over those from palliative care physicians when standardized for prompt and statement length. Analysis shows AI-generated statements contain more positive language which may be the factor driving patient preference toward AI.

Humans↗

PubMed2026

An AI-assisted Clinical Decision Support System for Green Classification of Cystocele on Dynamic Transperineal Ultrasound.

Green classification of cystocele on dynamic transperineal ultrasound (TPUS) remains operator-dependent because it requires manual frame selection and landmark-based assessment of the Valsalva maneuver. We developed a workflow-oriented AI-assisted clinical decision support system for automated urethrovesical junction localization and dynamic Green classification and prospectively evaluated its standalone and reader-support performance. This diagnostic accuracy and reader study included 881 patients from a tertiary referral hospital, comprising a retrospective development cohort (n = 688) and an independent prospective test cohort (n = 193). A nested subset of 67 prospective patients was used for a reader study involving two junior and two intermediate radiologists under unaided and AI-assisted conditions. In the complete prospective test cohort, Green-AttGRU achieved a macro-averaged AUC of 0.939 (95% CI, 0.897-0.971) and an overall accuracy of 0.902 (95% CI, 0.860-0.943). In the reader study, overall accuracy increased from 0.761 to 0.821 without AI to 0.851-0.881 with AI, while macro-F1 increased from 0.660 to 0.777 to 0.820-0.860. Overall inter-reader agreement increased from a Fleiss' κ of 0.453 to 0.786, and pooled median interpretation time decreased from 26.7 s to 9.9 s. These findings support the preliminary feasibility of the system as a workflow-oriented decision-support tool for dynamic TPUS interpretation.

Humans↗

PubMed2026

Experimental validation of an AI-driven digital healthcare platform for oral health behavior and plaque assessment among vietnamese children.

BACKGROUND: Oral health among children in developing countries, including Vietnam, remains a significant public health concern. Innovative approaches leveraging artificial intelligence AI-based digital health platforms may offer effective strategies for managing dental plaque and promoting better oral hygiene behaviors among school-aged children. This study aimed to evaluate the effectiveness of an AI-driven oral healthcare platform (Denti-i Vietnam) in improving oral hygiene and behavioral outcomes among Vietnamese primary school students. METHODS: A total of 204 primary school students aged 8-10 years in Hanoi, Vietnam, participated in this experimental study. Participants were randomly assigned to an intervention group (n = 107), which used the AI-driven oral healthcare platform, and a comparison group (n = 97), which received traditional oral health education via pamphlets. Oral health behaviors, dental plaque levels (Simplified Oral Hygiene Index; OHI-S), and caries indices (dft/DMFT) were assessed at baseline and after the intervention period. RESULTS: The intervention group demonstrated a significant reduction in the OHI-S score compared to baseline (2.49 ± 0.60 to 1.70 ± 0.76, p < 0.001), particularly in the debris component, indicating enhanced plaque control. Notable improvements were also observed in oral hygiene behaviors, including increased frequency of toothbrushing before and after breakfast (p < 0.01) and more frequent parental assistance during brushing (p = 0.03). Furthermore, parental awareness of dental caries significantly increased in the intervention group (p = 0.001). CONCLUSIONS: The AI-driven oral healthcare platform significantly improved both oral hygiene behaviors and plaque control among Vietnamese primary school children. These findings suggest that AI-driven digital health tools can serve as practical and scalable solutions for promoting oral health in developing countries.

Humans↗

PubMed2026

Evaluation of a cornea-specialized large language model for diagnostic and management accuracy in complex corneal cases.

PURPOSE: To evaluate whether a cornea-specialized large language model (LLM) enhanced with retrieval-augmented generation (RAG) improves clinicians' diagnostic and management accuracy in complex corneal cases compared to a general-purpose GPT-4o model and unaided clinician performance. METHODS: This prospective, randomized, masked evaluation study involved three cornea trainees who each independently reviewed 39 real-world corneal cases under three experimental conditions: unaided, GPT-4o-assisted, and assisted by a cornea-specialized GPT-4o model. The cornea-specialized model was constructed by embedding over 200 publicly available Wikipedia articles into GPT-4o's RAG framework. Participants provided open-ended diagnoses and selected the next-step management options (multiple choice). They were allowed up to three GPT-4o queries per case, and the AI-assisted arms were randomized to minimize bias. Accuracy for both tasks was compared against expert reference standards using McNemar's test. RESULTS: Diagnostic accuracy was 48.7%, 20.5%, and 38.5% unaided, improving to 69.2%, 46.2%, and 59.0% with general GPT-4o (p<0.04). The cornea-specialized GPT-4o further improved accuracy to 71.8%, 48.7%, and 74.4%, with improvements over unaided performance for all clinicians (p<0.01). For next-step decisions, unaided accuracy was 76.9%, 87.2%, and 59.0%. With the specialized model, Ophthalmologist 3 improved to 71.8% (p<0.05), Ophthalmologist 1 remained high at 82.1%, and Ophthalmologist 2 declined to 64.1% (p<0.05). CONCLUSIONS: A cornea-specialized LLM enhanced with RAG improved diagnostic accuracy in complex corneal cases, particularly among clinicians with lower baseline performance. Effects on management accuracy were inconsistent. Future studies should explore the use of open-ended management tasks and examine whether smaller, curated retrieval corpora yield better model performance.

Humans↗

PubMed2026

Artificial intelligence-assisted detection and optical differentiation of colorectal lesions in Lynch syndrome surveillance (CADLY2): a multicentre, open-label, randomised controlled superiority trial.

BACKGROUND: Artificial intelligence (AI)-based computer-aided detection (CADe) systems improve adenoma detection in average-risk colorectal cancer screening. Meanwhile, evidence in Lynch syndrome surveillance is sparse and inconsistent. We assessed the effect of CADe on adenoma detection during Lynch syndrome surveillance. Computer-aided optical diagnosis (CADx) performance for optical differentiation of colorectal lesions was evaluated as a secondary aim. METHODS: CADLY2 was an international, multicentre, open-label, randomised controlled superiority trial at nine specialised hereditary cancer surveillance centres in Belgium, Germany, the Netherlands, and Spain. Adults aged 18 years or older with genetically confirmed Lynch syndrome scheduled for surveillance colonoscopy were randomly assigned (1:1) to high-definition white-light (HD-WL) colonoscopy alone or to HD-WL colonoscopy with computer-aided assistance from CAD EYE (Fujifilm, Tokyo, Japan). CAD EYE was used for CADe during withdrawal and for CADx after lesion detection. Randomisation was done centrally through a secure web-based system using Pocock's minimisation algorithm with a stochastic component and was stratified by centre, sex, previous colorectal cancer, underlying pathogenic variant, and interval since previous colonoscopy. Allocation concealment was ensured through the centralised web-based system. Patients were masked to group allocation until the start of withdrawal in procedures with mild sedation, or until completion of the procedure in procedures with propofol-based sedation. Endoscopists were not masked. The primary outcome was adenoma detection rate, defined as the proportion of patients with at least one histopathologically confirmed adenoma, analysed in the full analysis set (defined as all randomly allocated patients with available data for the primary outcome). The diagnostic performance of the CADx system was evaluated as a secondary outcome. The safety analysis set comprised all randomly allocated patients who underwent a study colonoscopy. This study is registered with the German Clinical Trials Register, DRKS00030695, and is completed. FINDINGS: Between May 9, 2023, and Oct 30, 2025, 757 patients were randomly allocated to HD-WL colonoscopy (377 patients) or to AI-assisted colonoscopy (380 patients); 733 patients were included in the full analysis set (369 HD-WL and 364 AI-assisted). The median age was 49 years (IQR 38-59) in the HD-WL group and 50 years (38-59) in the AI-assisted group; 213 (58%) were female and 156 (42%) male in the HD-WL group, and 207 (57%) were female and 157 (43%) male in the AI-assisted group. The adenoma detection rate was 30·9% (114 of 369 patients) with HD-WL versus 33·8% (123 of 364 patients) with CADe assistance (odds ratio 1·14 [95% CI 0·83-1·57], p=0·41). For CADx differentiation of neoplastic versus non-neoplastic lesions in the paired lesion-level analysis, with histopathology as the reference standard and sessile serrated lesions and traditional serrated adenomas classified as non-neoplastic, CADx sensitivity was 85·9% (95% CI 82·0-89·1) and specificity was 91·4% (89·4-93·0). Three adverse events occurred in the AI-assisted group: two mild post-polypectomy bleedings and one serious pulmonary embolism or deep venous thrombosis unrelated to the procedure. No adverse events occurred in the HD-WL group. INTERPRETATION: CADe-assisted colonoscopy did not show the absolute improvement in adenoma detection rate that was assumed in the prespecified sample-size calculation. CADx did not clearly improve lesion differentiation beyond expert optical diagnosis in expert Lynch syndrome surveillance settings. FUNDING: Third-party research funding of the National Center for Hereditary Tumor Syndromes, University Hospital Bonn.

Humans↗

PubMed2026

Role of Polygenic Risk Scores in Predicting Cognitive Functioning after Mild Traumatic Brain Injury: A TRACK-TBI Study.

Patients with traumatic brain injury (TBI) and Glasgow Coma Scale scores of 13-15 (historically called mild TBI [mTBI]) commonly experience changes in cognitive functioning, including processing speed, memory, and executive functioning. In a prospective sample (N = 523) of individuals of European descent who had been treated in a U.S. level 1 trauma center for mTBI, we examined the prognostic value of four polygenic risk scores (PRS) for cognitive outcomes at 6-months postinjury. To estimate the impact of mTBI on cognition, primary cognitive outcomes were scaled as z-scores reflecting changes in performance relative to predicted preinjury performance. The PRS examined were previously developed and validated to predict cognition-related outcomes of educational attainment (Education-PRS), intelligence (Intelligence-PRS), and Alzheimer's disease (AD-mild traumatic brain injury (APOE)-PRS and AD + APOE-PRS). Both the Education-PRS and Intelligence-PRS displayed bivariate associations with all four cognitive outcomes (β = 0.19-0.32), whereas neither Alzheimer's disease PRS was significantly associated with any outcome. After controlling for other factors known to predict cognitive outcomes of TBI (e.g., sex, education, mTBI severity defined by a combination of Glasgow Coma Scale scores and the presence/absence of acute intracranial findings on clinical neuroimaging), the Education-PRS and Intelligence-PRS remained independently predictive of verbal episodic memory (β = 0.10-0.16), whereas their associations with processing speed and executive functioning were mostly nonsignificant and were mediated through educational attainment. Looking across primary z-score and secondary raw score outcomes, cognitive outcomes 6 months post-mTBI were good on average, and PRS made small independent contributions to outcome prediction. The mediation model findings may support theories of cognitive reserve, which propose that individuals with stronger preinjury cognitive processing abilities (often estimated by educational history) can better compensate for TBI. Moreover, findings indicate that PRS may contribute modestly to multivariable models predicting cognitive function after TBI.

Humans↗