340880 research outputs found
Sort by
EFLM checklist for the assessment of AI/ML studies in laboratory medicine:enhancing general medical AI frameworks for laboratory-specific applications
No full textThe integration of artificial intelligence (AI) and machine learning (ML) into laboratory medicine shows promise for advancing diagnostic, prognostic, and decision-support tools; however, routine clinical implementation remains limited and heterogeneous. Laboratory data presents unique methodological and semantic complexities - method dependency, analyte-specific variation, and contextual sensitivity-not adequately addressed by general-purpose AI reporting guidelines. To bridge this gap, the EFLM Committee on Digitalisation and Artificial Intelligence (C-AI) proposes an expanded checklist to support assessment of requirements and recommendations for the development of AI/ML models based on laboratory data. Building upon the widely adopted ChAMAI checklist (Checklist for assessment of medical AI), our proposal introduces six additional items, each grounded in the CRoss Industry Standard Process for Data Mining (CRISP-DM) framework and tailored to the specificities of laboratory workflows. These extensions address: (1) explicit documentation of laboratory data characteristics; (2) consideration of biological and analytical variability; (3) the role of metadata and peridata in contextualizing results; (4) analyte harmonization and standardization practices; (5) rigorous external validation with attention to dataset similarity; and (6) the implementation of FAIR data principles for transparency and reproducibility. Together, these recommendations aim to foster robust, interpretable, and generalizable AI systems that are fit for deployment in clinical laboratory settings. By incorporating these laboratory-aware considerations into model development pipelines, researchers and practitioners can enhance both the scientific rigor and practical applicability of AI tools. We advocate for the adoption of this extended checklist by developers, reviewers, and regulators to promote trustworthy and reproducible AI in laboratory medicine
Fractionation and biological evaluation of passion fruit Pectins:HG and RG-I backbone ratios are associated with TLR2–1 interaction and signaling
No full textPassion fruit mesocarp is rich in pectin, and high-temperature/pressure modification of this pectin has been shown to yield bioactive fragments with anticancer potential. To clarify the structure-function relationship of passion fruit pectins, we purified native and modified pectins using two fractionation methods. Comprehensive chemical characterization revealed molecular weight as the primary difference between fractions, along with varying proportions of homogalacturonan (HG) and rhamnogalacturonan-I (RG-I). All samples activated TLR2, such as specific agonists (Pam3CSK4, HKLM, FSL-1). Notably, only native and lower-molecular-weight fractions inhibited TLR2/1 activation by the specific agonist Pam3CSK4. Higher methyl esterification correlated with TLR2/1 inhibition at lower doses, whereas RG-I content showed a negative correlation; however, the galacturonic acid-to-rhamnose ratio positively influenced heterodimer inhibition. A highly methyl esterified galacturonic acid heptamer demonstrated a strong affinity for the TLR2/1 binding pocket, as evidenced by molecular dynamics simulations. This study elucidates how modified passion fruit pectin structures interact with TLR2, reinforcing the link between plant polysaccharides and human immune responses
PRevention Of sudden cardiac death aFter myocardial Infarction by Defibrillator im-plantation:Design and rationale of the PROFID EHRA randomized clinical trial
No full textBackground: Randomized clinical trials from over 20 years ago demonstrated that an implantable cardioverter defibrillator (ICD) improved survival for patients with severely reduced left ventricular ejection fraction (LVEF) after myocardial infarction (MI) compared with optimal medical therapy (OMT) alone. Since then advances in therapy have led to the reduction in the incidence of sudden cardiac death (SCD) in this population, whilst complication rates from ICD implantation are still substantial. Objectives: To determine whether OMT without ICD implantation is not inferior to OMT with ICD implantation with respect to all-cause mortality. Design: The PROFID EHRA trial is an investigator-driven, prospective, parallel-group, randomized, open-label, blinded outcome assessment (PROBE), multi-center, noninferiority trial without dedicated investigational medical device (Proof of Strategy Trial) with 2 groups with 1:1 randomization. PROFID-EHRA will recruit approximately 3,595 patients with documented history of MI at least 3 months prior, LVEF ≤35%, on OMT for at least 3 months, and with New York Heart Association class II or III, who will be randomized to OMT or OMT plus ICD, to collect 374 first primary outcome events within a median observation period of around 28 months from about 180 clinical sites in an estimated 13 countries. The primary outcome is time from randomization to the occurrence of all-cause death. Secondary outcomes include time from randomization to death from cardiovascular causes, to SCD, to first hospital readmission for cardiovascular causes after date of randomization, the average length of hospital stay during follow-up, and quality of life trajectories. Clinical Trial: Trials.gov</p
Preference research is underutilized in health prevention
No full textIntroduction: Preference research is crucial in shaping effective patient-centered preventive healthcare strategies. Preventive measures such as vaccinations, screenings, and proactive treatment of asymptomatic high-risk conditions aim to reduce future health risks, often in healthy individuals. Understanding patient preferences in these contexts is vital to ensure that interventions are clinically effective, acceptable, and actionable. Failing to consider patient preferences when developing preventive strategies may result in low uptake and adherence to prevention programs, ultimately limiting their impact on patient and population health
Unsupervised machine learning to explore inflammation following cardiopulmonary bypass
No full textIntroduction: Cardiac surgery with cardiopulmonary bypass (CPB) often induces systemic inflammatory reaction syndrome (SIRS), affecting postoperative outcome. We aimed to explore adaptive/maladaptive inflammation using unsupervised machine learning. Methods: We conducted a post hoc analysis of 1908 adult patients who underwent elective cardiac surgery with CPB between June 2016 and June 2020 at a single institution. Patients were assessed for SIRS 12 hours post-surgery and clustered using the partitioning around medoids (PAM) algorithm based on Gower distance. The influence of SIRS on a composite outcome comprising death, stroke/TIA, renal replacement therapy, reoperation for bleeding, mechanical circulatory support, and ICU stay >96 hours was analyzed via multivariable logistic regression. Results: SIRS occurred in 28.7% of patients (median age 69 years; 68.7% male). Clustering revealed two subgroups: maladaptive SIRS (52.9%) with higher preoperative risk and worse outcomes, and adaptive SIRS (47.1%) with favorable outcomes. Maladaptive SIRS patients had higher 30-day mortality (21.7% vs 1.6%, p < .001). Adaptive SIRS patients had outcomes similar to SIRS-negative controls. In selected clusters, SIRS was independently associated with a lower risk of the composite outcome (OR 0.44; 95% CI 0.26-0.74, p = .002). Conclusion: Unsupervised machine learning effectively identifies adaptive and maladaptive SIRS in cardiac surgery patients, providing a basis for personalized postoperative care. Several clinical and procedural factors associated with maladaptive SIRS may be modifiable, supporting future precision strategies to reduce harmful inflammation after cardiac surgery.</p
Evaluating the need for late angiography for complete obliteration of AVMs after endovascular treatment:collaborative AVM center experience and a systematic review
No full textBrain arteriovenous malformations (bAVMs) can be treated curatively by endovascular embolization. However, limited data exist on late recanalization rates. We investigated the rate of late bAVM recanalization following complete endovascular obliteration, confirmed by primary control angiography. We performed a single-center retrospective cohort study on late recurrences in adult patients with bAVMs after complete endovascular obliteration at a Radboud – Isala - MUMC+ (RIM) collaborative AVM center in the Netherlands between 2014 and 2022. Additionally, we conducted a systematic review to evaluate the rate of late recanalization following complete endovascular obliteration, confirmed on primary control angiography, in adults with bAVMs. The protocol for this review was registered in PROSPERO (CRD42024546875). Our retrospective study revealed 42 adult patients treated by endovascular means only; 90.5% (38 patients with mean age of 48.1 years) had confirmed complete obliteration. Both primary (6 months post-treatment) and secondary (more than 1 year post-treatment) angiographic control confirmed complete obliteration in 21 of 23 patients with complete follow-up. Mean follow-up was 47.8 months. Two late recurrences (9.5%) were detected at 5- and 6-years' follow-up imaging. Our systematic review included two studies encompassing a total of 19 patients with mean angiographical follow-up of 20.8 months. There were no late recurrences. Late bAVM recurrence after endovascular treatment with proven complete obliteration may be underestimated owing to limited long-term follow-up. Our findings suggest that after a 6-month angiographic confirmed obliteration, a 5-year angiographic imaging control is justified.</p
Updated LIfestyle for BRAin health (LIBRA2) Score, Genetics, and Risk of Alzheimer Disease, Vascular Dementia, and Stroke in Older Adults
No full textBackground and ObjectivesDementia and stroke are major global causes of disability, with modifiable lifestyle factors playing a significant role in their development. The updated LIfestyle for BRAin health (LIBRA2) score integrates 15 modifiable risk and protective factors to quantify lifestyle-based dementia risk, but its relationship with stroke and dementia subtypes remains unexplored. We investigated LIBRA2's association with stroke and dementia subtypes while also assessing potential interactions with genetic susceptibility.MethodsProspective data were used from the French multicenter Three-City (3C) Study, with participants aged 65 years and older followed for up to 17 years. Weighted LIBRA2 scores at baseline were constructed based on the presence of 15 modifiable risk and protective factors, with higher scores representing higher lifestyle-based dementia risk. Cox proportional hazards models were used to study the association of LIBRA2 with incident dementia (and subtypes) and stroke, adjudicated by expert neurologist panels. Genetic susceptibility to dementia and stroke was assessed using APOE epsilon 4 carriership and disease-specific genetic risk scores.ResultsAnalyses included 4,731 participants for stroke (mean age 73.8 years, 60.1% female) and 4,737 participants for dementia (mean age 73.8 years, 59.9% female). One-point increases in LIBRA2 scores (theoretical range -6.1 to +25.7) were associated with increased dementia risk (hazard ratio [HR] 1.08; 1.06-1.11), with a stronger association for vascular and mixed dementia (HR 1.13; 1.08-1.18) compared with Alzheimer disease (AD) dementia (HR 1.06; 1.03-1.09). LIBRA2 was not significantly associated with incident stroke risk (HR 1.03; 0.99-1.07). No significant interaction was found between LIBRA2 and APOE epsilon 4 carriership or disease-specific genetic risk scores in relation to dementia subtypes or stroke.DiscussionLIBRA2 serves as a valuable tool for assessing lifestyle-related dementia risk and its subtypes but showed no association with stroke, highlighting the potential for a stroke-specific risk reduction model. These associations were independent of genetic disease susceptibility, reinforcing the universal benefits of lifestyle modifications on dementia risk reduction. The stronger association between LIBRA2 (and some individual components) and vascular or mixed dementia, compared with AD dementia, highlights the pivotal role of vascular mechanisms in the relationship between lifestyle and brain health
