Despite its growing impact, advanced liver fibrosis (ALF) remains substantially underdiagnosed in both primary and specialized care settings. Since liver cirrhosis is typically preceded by a prolonged phase of asymptomatic fibrosis, early detection of ALF, particularly in high-risk individuals, represents a public health priority. From a laboratory medicine perspective, this prolonged subclinical phase offers an interesting opportunity for early detection of ALF using blood-based noninvasive tests (NITs) that can be implemented in primary and nonhepatology care settings before overt disease develops. By critically appraising the evidence sources in the available literature, this paper provides an overview of blood-based NITs useful for the detection of ALF, with particular emphasis on the aspects and problems related to their implementation in daily laboratory practice. We explore the feasibility of different scenarios using strategies based on routine biochemical parameters and more specialized NITs that incorporate measurements of direct markers of fibrosis activity. Moreover, we highlight discrepancies existing among clinical practice guideline (CPG) recommendations that may hamper their widespread implementation in medical laboratories. Advances in understanding and increase in prevalence of ALF require earlier detection and more accurate risk assessment of this condition. Blood-based NITs may provide a widely accessible diagnostic aid, especially in primary care and resource-limited settings. Multidisciplinary collaboration focusing on their integration into clinical pathways to optimize patient evaluation and specialist referral is required. Harmonization of recommendations in international CPGs will certainly contribute to their more effective use.

Artificial intelligence (AI) augmented laboratory tests can improve quality, efficiency, cost-effectiveness, and staff satisfaction. However, the clinical success of these tests can introduce unforeseen challenges for model retraining. This study describes the discovery of an "AI label feedback loop" in a clinically implemented AI-augmented kidney stone composition test. An AI-augmented kidney stone composition test (V1) has been previously deployed for clinical kidney stone characterization. After several years of clinical use, a retrained model (V2) was developed using 6 times more data. Model performance of both V1 and V2 were evaluated across 3 datasets: a recent production validation (hold-out) set (mostly AI-influenced labels), the original V1 validation set (pre-AI, entirely human-labeled), and a subset of recent cases with exclusively human-generated or human-corrected labels. V2 demonstrated a 10% lower concordance rate than V1 when evaluated on the recent production hold-out set, despite a much larger training dataset. Performance between V1 and V2 was similar when applied to the pre-AI validation set. Notably, V2 outperformed V1 on the recent subset of cases with human-only or human-corrected labels, particularly for less-common stone types. These findings revealed an AI label feedback loop, confounding retraining and evaluation. The integration of AI into clinical practice can potentially influence reported test results, complicating the development and evaluation of future models. To mitigate AI label feedback loops, ongoing human annotation and careful validation set construction are essential. These strategies can ensure reliable performance assessment and support the safe evolution of clinical AI systems.

Primary immunodeficiencies (PIDs) are rare disorders caused by immune system defects that are commonly screened using multi-parameter flow cytometry (FCM). To counter the subjective and time-consuming manual data analysis of FCM data, we present PIDgeon, a fully automated computational pipeline based on artificial intelligence (AI) techniques. PIDgeon is designed to characterize PID immune profiles, suggest PID subtypes based on altered immune profiles, age, and immunoglobulin levels, and generate interpretable reports. The PIDgeon pipeline, including FlowSOM and extreme gradient boosting models, was trained and tested on standardized FCM data generated according to EuroFlow procedures on 74 healthy controls and 399 patients (281 lymphoid-PID patients and 118 non-PID diseased controls) collected by the Ghent University Hospital. Subsequently, multi-centric validation was performed on internal (n = 211) and external (n = 338) independent data sets collected across 4 EuroFlow centers. Validation demonstrated high accuracy in cell count enumeration, achieving correlation scores above 0.90 for the major lymphocyte subsets. Interestingly, PIDgeon showed high sensitivity (93% to 100%) in predicting PID with severe T-cell defects, such as severe combined immunodeficiency and late-onset combined immunodeficiency, and low false-negative rates (1.5% to 5.4%) for distinguishing other lymphoid-PID vs non-PID diseased controls across data sets. Additionally, PIDgeon gives a first hint toward prediction of subtypes of primary antibody deficiencies, such as common variable immunodeficiency. In summary, PIDgeon is an accessible and explainable AI-pipeline aligned with current clinical needs, aiding laboratory immunologists in early PID diagnostics and increasing data analysis efficiency.

Real-time reverse transcription quantitative PCR (RT-qPCR) is utilized in many areas of the life sciences, diagnostics, and forensics, yet concerns about methodological quality and reporting transparency persist. Diagnostic testing during the recent pandemic brought those concerns into the public domain. The Minimum Information for Publication of Quantitative PCR Experiments (MIQE) guidelines, introduced in 2009 and updated in 2025, were intended to standardize assay design and reporting, but their impact has been modest. We assessed trends in RT-qPCR methodological reporting between 2007 and 2025 using PubMed Central searches and manual evaluation of 355 full-text articles from 2019 and 2024. Parameters analyzed included RNA integrity assessment, oligonucleotide sequence disclosure, reference gene validation, PCR efficiency reporting, and MIQE citation. In addition, targeted cohorts of 50 "reference gene" and 50 "PCR efficiency" publications from 2024/25 were evaluated. Results were compared across timepoints, geographic regions, and MIQE-citing vs non-citing studies. Reporting of core parameters remained low or declined. Between 2019 and 2024, RNA integrity reporting fell (22% to 11%), reference gene validation was rare (13% to 5%), and PCR efficiency reporting collapsed (13% to 1%). MIQE-citing papers in 2024 showed better adherence (31% RNA integrity, 47% reference gene validation, and 40% PCR efficiency) but still omitted essential details. Asia now dominates RT-qPCR output by volume, while Europe contributes most MIQE-citing studies. Targeted cohorts reported more methodological information, yet many still failed to meet basic standards. These findings confirm that incomplete experimental design and reporting continue to undermine reproducibility and robustness of RT-qPCR assays.

While most pulmonary infections with Coccidioides spp. are self-limited, a subset of patients is at higher risk for severe disease and extrapulmonary dissemination, including individuals with certain ethnic backgrounds. Disseminated disease can involve soft tissues, bones, and the central nervous system with high morbidity and mortality in the absence of antifungal treatment. Diagnosis of coccidioidomycosis can be challenged by the availability of specimens for culture or histopathology, confusion with, or lack of availability of various antibody tests and overall lack of consideration of Coccidioides spp. as the etiology of disease, especially in immunocompetent hosts. We set out to characterize cases of disseminated coccidioidomycosis at our institution over a 7-year period, solely in immunocompetent hosts, to highlight diagnostic delays and determine which, if any, primary screening test might be the most useful. A total of 40 cases met our inclusion criteria, and 100% of these cases had positive immunoglobulin G antibodies on a US FDA-cleared Coccidioides spp. enzyme immunoassay. Nearly all of the cases (87%) had a delay in diagnosis and associated worsening of disease (71%). Locations of initial presentations that led to delayed recognition included primary care settings (56%), emergency departments (33%), and urgent care centers (11%), all in the region of Coccidioides spp. endemicity. These findings highlight the need for interventions to increase awareness of the risk factors for, the symptoms of, and the appropriate testing options to diagnose disseminated coccidioidomycosis.