Bio-Rad (United States)

<title>Abstract</title> <bold>Objective</bold> To evaluate the quality, safety, and clinical utility of AI-generated radiology impressions compared with human-authored impressions across multiple clinical stakeholder groups. <bold>Materials &amp; Methods</bold> A retrospective, blinded evaluation was conducted using 200 oncologic computed-tomography reports from a U.S. academic cancer center. Three impression types were assessed for each report: original radiologist-authored impressions, impressions generated by a custom domain-specific AI model fine-tuned on institutional data, and impressions generated by a general-purpose large language model. Original authoring radiologists, independent radiologists, and oncologists evaluated impressions using structured Likert-scale metrics assessing completeness, correctness, conciseness, clarity, clinical utility, and potential patient harm. Pairwise comparisons were performed using Wilcoxon signed-rank and two-proportion z-tests. <bold>Results</bold> Custom model AI impressions demonstrated near parity with human-authored impressions across most quality metrics. Original radiologists rated their own impressions as slightly more complete, while independent radiologists showed no significant differences between original and custom model impressions. Generic model impressions were longer, rated as more complete but significantly less concise. Patient harm ratings were uniformly low. Radiologists preferred original and custom model impressions over generic model impressions, whereas oncologists showed no significant preference. <bold>Discussion</bold> Evaluation outcomes varied by stakeholder group, highlighting differing priorities between radiologists and oncologists. Low inter-rater agreement across several quality metrics suggests that impression quality is inherently subjective and context dependent rather than defined by a single objective standard. <bold>Conclusion</bold> AI-generated radiology impressions, particularly those produced by custom domain-specific models, can achieve quality and safety comparable to human-authored impressions. These findings support the use of AI as an adaptable drafting aid that complements radiologist judgment.

BackgroundThe Regional Medical Laboratory Center (RMLC) integrates laboratory departments from hospitals with different levels to optimize resource allocation, enhance testing efficiency, and promote the sharing of professional expertise. Despite these benefits, ensuring consistent quality control (QC) across the RMLC presents a significant challenge due to diverse equipment, procedures, and staff expertise levels. To address these challenges, this study provided an information tool called iLab, developed by Shenzhen Huikang Information Technology Co, Ltd, for managing internal quality control (IQC) and external quality assessment (EQA).ObjectiveThis study aimed to develop and implement an integrated informatics platform managing IQC and EQA across RMLC.MethodsWe integrated 4 software programs to manage IQC and EQA, laboratory information system (LIS) supplied by Shanghai Tengcheng Medical Tech-Info Co, Ltd; Quality Control Box (QCBOX) and Quality Control EQA Center (QCEC) provided by Bio-Rad Laboratories (Shanghai) Co, Ltd; and the iLab electronic recording system. The platform centralizes QC data processing, with QCBOX and QCEC handling operational workflows, while iLab oversees documentation, analytics, and approval processes. It was successfully deployed in August 2023 across a pilot network comprising 5 district hospitals, 1 central laboratory, and 4 community health stations in Shenzhen's Luohu District.ResultsBy leveraging the LIS-QCBOX data flow, the platform streamlined the entire IQC process, which incorporated interlaboratory comparisons and iLab-facilitated postanalysis management. Concurrently, EQA was enhanced through electronic documentation, multirule analysis of acceptable data, regional proficiency testing via QCEC, and measurement uncertainty (MU) estimation. By September 2025, the platform supported 114 users, integrating 133 instruments and 859 active QC items, and generated 2331 monthly IQC analysis reports, 340 EQA multirule analyses, and 289 MU reports. Comparative analysis revealed marked improvements in key metrics from 2023 (preimplementation) to 2024 (postimplementation): the IQC implementation rate rose from 97.79% to 99.87% (P<.001), the report error rate decreased from 0.048% to 0.027% (P<.001), and the intralaboratory turnaround time compliance rate increased from 95.49% to 95.71% (P<.001). The annual EQA unacceptable rate dropped from 0.34% in 2023 to 0% in 2024. The number of ISO (International Organization for Standardization) 15189–accredited test items increased from 203 in 2022 to 206 in 2024.ConclusionsThe iLab system establishes a scalable framework for standardized IQC and EQA management in RMLC, demonstrating significant potential to enhance regional laboratory quality.

In the context of intensifying cyber threats, cyberspace resilience has become a strategic priority for organizations, governments, and international alliances. Cyber scenarios, in their various forms, provide a controlled environment in which response capabilities to cybersecurity incidents can be tested, trained, and validated. The aim of this paper is to analyze how these scenarios contribute to the development of an organizational culture of resilience, allowing the identification of vulnerabilities, strengthening decision-making under pressure, and proactively adapting to new types of threats. The study explores current practices adopted at the European and international level, revealing how scenario-based exercises support strategic planning, training of technical and operational teams, and interinstitutional cooperation. Moreover, the direct benefits to organizational flexibility and post-incident recovery capacity are also analyzed. Finally, the article formulates a series of recommendations for integrating scenarios into the cyber risk management cycle, highlighting their value not only as training tools but also as fundamental elements in the security architecture of a modern organization.

Salmonella contamination in poultry remains a persistent public health challenge, despite ongoing regulatory efforts to reduce its presence. Traditional culture-based enrichment methods commonly employed for detecting Salmonella have inherent limitations, such as biases in bacterial growth and inaccurate quantification of initial contamination levels. This study introduces an enrichment-free approach for quantifying Salmonella in raw poultry, utilizing the Pathotrak system for bacterial separation and concentration, followed by Droplet Digital PCR (ddPCR) for absolute quantification. Our findings highlight that whole-cell preparations within ddPCR droplets yield significantly higher detection efficiency compared to extracted DNA, likely due to the reduced loss of genetic material during processing. The method was validated by correlating theoretical inoculation levels with ddPCR measurements, resulting in a strong linear relationship (R2=0.903) and a slope of 0.99, indicating high accuracy. Residuals from the regression model were normally distributed, confirming its validity. This platform offers a robust, rapid, and precise tool for detecting low levels of Salmonella in poultry products without enrichment, making it a more accurate and efficient approach to bacterial quantification in complex food matrices.

Immune response to vaccination is assessed when adequate vaccine protection is in question or immunosuppression is imminent through measurement of antibody levels, which wane as time from vaccination increases. The serologic cutoff value for adequate response is based on thresholds derived from studies in adults, and age-appropriate thresholds for children have not been established. We sought to investigate age-specific differences in antibody levels in healthy children to guide determination of vaccine immunity status when clinically indicated. This cross-sectional study assessed clinical serology for measles, mumps, rubella (MMR), varicella, and hepatitis B (HepB) in an age-stratified cohort of 471 healthy children who were up to date for vaccination (1 to 18 years). Remnant specimens with sufficient volume were collected from July 23, 2019, to November 17, 2020, as convenience samples and chart reviewed for inclusion. While children of all ages had detectable titers to MMR, median titers for HepB and varicella waned by ages 11 to 12 and 9 to 10 years, respectively. Children had titers above adult thresholds for MMR at all measured timepoints, retrospectively resulting in 24.6% (95% CI, 21.6%-27.8%) of children having an inappropriate MMR classification when adult instead of pediatric thresholds were used. Current use of HepB and varicella serology may be inappropriate due to the rapid waning of titers. The adequacy of an individual's response to one vaccine component did not infer adequate responses to other components. Application of age-appropriate reference intervals for vaccine serologic tests will provide a foundation for improved treatment recommendations and standards of care.

Management of colorectal cancer (CRC) patients relies on the accurate determination of microsatellite instability (MSI) status. MSI status can have an influence on therapy decisions centered on immune checkpoint inhibitors. In this study a novel droplet digital PCR (ddPCR) kit for MSI status determination was validated across 2 separate CRC patient cohorts: 102 tumor samples from the ALGECOLS cohort and 129 plasma samples from the RASANC cohort. Each cohort was assessed for MSI status using the novel ddPCR kit and compared to historical and/or newly obtained results, (either immunohistochemistry analysis or PCR amplification). Concordance between ddPCR and conventional MSI determination methods for the analysis of tissue samples was 97.1% for ALGECOLS. When looking at positive ctDNA samples, a strong concordance was observed for the RASANC cohort. This study illustrates that ddPCR MSI testing represents a rapid, sensitive and accurate method with a strong correlation to established methods. Moreover, the ability of the described approach to monitor MSI status in both tumor and plasma enhances the potential for the use of MSI status in longitudinal monitoring of CRC patients.