Electronic Journal of the International Federation of Clinical Chemistry and Laboratory Medicine最新文献

Artificial intelligence (AI) is driving a profound transformation across the healthcare landscape, with the potential to enhance diagnostic accuracy, optimize clinical decision-making, improve resource allocation, and advance personalized medicine. In public health, AI is redefining infectious disease epidemiology by enabling outbreak forecasting, genomic surveillance, and data-driven policy support, even in the presence of incomplete information. Within clinical laboratories, AI plays a pivotal and expanding role. It facilitates automation of complex workflows, supports diagnostic interpretation, and contributes to analytical performance improvements. Particularly promising is its integration into point-of-care testing, enabling decentralized diagnostics and broader access to timely care, especially in resource-constrained settings. However, these advancements are not without challenges. Concerns regarding algorithmic bias, lack of data representativeness, and risks to privacy and transparency must be carefully addressed. Moreover, the ethical and societal implications of AI are increasingly central. As emphasized by Pope Francis, while AI may accelerate access to knowledge and innovation, it also risks deepening global disparities and promoting a "throwaway culture" that undermines human dignity. His appeal for a "culture of encounter" rooted in equity, justice, and inclusion aligns with the mission of public health and laboratory medicine. This paper, based on the invited lecture delivered at the Clinical Laboratories Artificial Intelligence Revolution (CLAIR) 2025 conference, explores these themes through a critical lens. International scientific societies such as the IFCC are called to foster equitable implementation of AI by promoting access to training, infrastructure, and governance frameworks thus ensuring that AI contributes meaningfully to global health solidarity and equity.

The artificial intelligence (AI) integration in the medical field benefits both clinicians and patients. One of the most exciting applications of AI in healthcare is its role in electronic health records (EHRs). Leveraging AI to enhance EHR holds incredible potential for streamlining processes and reducing errors, enhancing clinical decision support and improving interoperability, contributing to more accurate, personalized, and effective healthcare, improved patient outcomes and quality of life. This article provides an overview of the role of AI in EHR, illustrates several examples of how AI incorporation into EHRs transforms healthcare delivery, illustrates how AI-powered EHRs impact healthcare stakeholders, and highlights the challenges and barriers of AI adoption in EHR systems.

Background: Ensuring quality in the analytical phase of clinical chemistry is paramount for accurate diagnosis and treatment. Sigma metrics offer a quantitative framework to assess and enhance laboratory performance. In this study, we intend to comprehensively assess diverse biochemical parameters using three different QC databases to determine their suitability and design a tailor-made QC plan based on this assessment.

Methods: This is a retrospective study, from an NABL-accredited laboratory. The coefficient of variation (CV) % was obtained from the IQC results and the Bias % from Proficiency Testing (PT) results. The Sigma value was calculated using the TEa from three different biological variation databases (EFLM database, Westgard database, CLIA database). QGI was calculated for parameters with a Sigma value <3.

Results: Around 28-33 parameters in different instruments showed a Sigma value <3 (poor performance). However, several parameters lack TEa values in the CLIA database, preventing their inclusion in assessments of acceptability.

Conclusion: By integrating Sigma calculations with established TEa standards, this study helped in identifying areas needing improvement. This comprehensive assessment ensured the evaluation of the performance of diverse analytes, thereby ensuring higher accuracy and reliability in patient test results.

Background: Disease specific biomarkers are ideal tool to detect the presence of the disorder. Timely detection of disorders can improve the health outcome. The metabolic arrangements in pre-term (PT) and low-birth weight (LBW) newborns differ from those term-born and have normal-birth weight (NBW). Hence, it is crucial to compare the values across study groups and establish a dedicated reference values for each group considering the gestational-age and birth weight.

Methods: The prospective study was conducted on the cohort of 2860 newborns who underwent newborn screening (NBS) in dried-bloodspot samples within five days of birth. The study groups included were TERMNBW, TERMLBW, PTNBW and PTLBW.

Results: The central tendency measures and the comparison of the NBS parameters across the study groups are presented. Males recorded a higher n17-OHP (p<0.001) median(range) compared to female newborns whereas nIRT (p=0.008) and nMSUD (p<0.001) were higher in female newborns. nTSH values was higher in TERMNBW than the PTLBW group (p=0.03). n17-OHP levels in TERMNBW and TERMLBW groups were lower than PTNBW and PTLBW (<0.001) newborns. nBIOT range of 378.8U and nG6PD range of 17.1U/gHb was highest in TERMNBW. The reference value observed for nTSH, n17-OHP and nIRT were respectively, 9.2mIU/L, 48.6nmol/L, 95.0μg/dL in TERMNBW and 16.9mIU/L, 70.2nmol/L, 76μg/dL in PTLBW. nG6PD reference level were respectively 2.0 and 1.6u/gHb in TERMNBW and PTLBW groups. The nBIOT levels were 52.7U and 48.0U respectively. Reference values were nearly similar for nPKU, nGAL and nMSUD.

Conclusion: The study has provided a detailed comparison and reference levels observed in various study groups and sub-groups considering the gestational-age and birth weight of the newborns.

Artificial Intelligence (AI) has transitioned from a technological concept to a foundational component of healthcare innovation. In laboratory medicine, it is no longer a question of whether AI will play a role, but rather how it will be responsibly integrated to amplify clinical value, operational efficiency, and equitable patient care. This article explores the current and future impact of AI across diagnostic workflows, clinical decision-making, personalized prevention strategies, and hospital governance. It also highlights the ethical, legal, and sustainability considerations critical to ensuring AI supports a smarter, fairer, and more sustainable healthcare system.

"Calibration" conveys the connotative meaning of "correction." Therefore, calibration is frequently viewed as "perfect," but it is a measurement, and no measurement is error-free. This study aims to uncover the sources of calibration errors, to estimate their size, and assess their consequences in quality control. The analytical bias is the difference between the working (determined) graph and the ideal graph (how the reagents behave). The source of the calibration random component is the random error committed in the calibration. The primary source of the systematic component is the reference material value error, which cannot be reduced to the nominal value error. Even if the avoidable human errors are neglected, the reconstitution errors, including two volume measurements, are inherent. The random component was estimated by making five calibrations in repeatability conditions and calculating the coefficient of variation of the slope factors. The total calibration error was estimated by comparing the slope factors of new calibrations using the same reagent and calibrator lots (one-year data). The results confirmed the presumptions: the calibration error is bigger than the coefficient of variation measured in repeatability conditions. Smaller biases are incorrigible by calibration, and quality control rules must be designed to prevent them from being detected. Using the σ parameter in the QC graphs would result in too frequent alarms. Westgard proportionally increased the decision limits by overestimating σ with the standard deviation measured in reproducibility within laboratory conditions. A more accurate solution is to increase all decision limits to account for the incorrigible bias and design the QC graphs with the standard deviation measured in repeatability.

Background: Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) has become the leading cause of chronic liver disease globally, affects more than one-third of the adult population and includes a spectrum of conditions ranging from simple steatosis of liver to metabolic dysfunction-associated steatohepatitis (MASH), progressive fibrosis, cirrhosis and, in some cases, hepatocellular carcinoma. Early detection and accurate staging are important to prevent disease progression and studies have recently identified metabolic and apoptotic markers such as Adropin, a peptide hormone secreted by the liver that is involved in energy homeostasis; Irisin, a myokine that is linked to exercise and metabolic regulation; and CK-18, a biomarker of hepatocyte apoptosis.Methods: Using FibroScan for the diagnosis and staging of MASLD, CAP scores were used for steatosis and liver stiffness measurements for fibrosis. Quantification of serum adropin, irisin, and CK-18 was done, and independent t-tests, correlation analysis, and ROC curve analysis were used for statistical analysis to assess the diagnostic potential.

Results: Adropin levels were lower in MASLD cases than in controls and decreased further with the severity of the disease. The association was highly significant (p < 0.001), indicating a very high negative correlation between Adropin levels and hepatic dysfunction. Levels of CK-18 were greatly increased in MASLD patients and were highly positively correlated with the degrees of fibrosis and steatosis (p < 0.001), which supports the hypothesis that it is a marker of hepatocyte apoptosis.

Conclusion: The significant changes in their levels observed in MASLD patients suggest their possible application in multimarker diagnostic strategies. Nonetheless, the inconsistent behavior of Irisin in this study requires more conclusive evidence from future studies involving larger samples. Such biomarkers may help in identifying the disease at an early stage and improve the management of the disease.

Post-transplantation diabetes mellitus (PTDM) is a common and important complication after solid organ transplantation, affecting long-term outcomes. Graft rejection, decreased patient survival, infections and increased cardiovascular risk are associated with PTDM and may arise from both transplant-related and traditional risk factors. Early screening for PTDM is crucial for early detection and management. Despite clinical guidelines recommending regular screening for PTDM, screening rates remain suboptimal. This retrospective study analyzes PTDM screening rates between January 2014-January 2024 among pediatric kidney, liver, heart and lung transplant recipients at a large quaternary academic pediatric transplant center. PTDM screening rates vary by organ type, with kidney transplant patients at 19.4%, liver transplant patients at 14.6%, heart transplant patients at 34.3% and lung transplant patients at 91.7%. These lower-than-expected rates of PTDM screening among high risk pediatric- kidney, liver and heart transplant pediatric population highlight the need for improved screening protocols and provider education post-transplantation.

Background: The adoption of cloud computing and Artificial Intelligence (AI) technologies offers significant advantages for External Quality Assessment (EQA) providers, including scalability, cost efficiency, and broader accessibility. However, these benefits come with substantial cybersecurity and data privacy challenges.

Methodology: We performed a systematic literature review on cybersecurity risks in healthcare cloud computing, consulted experts in bioinformatics and cybersecurity, and analyzed real-world hacking incidents targeting EQA organizations. A risk-focused framework was developed to outline key challenges and best practice mitigation strategies.

Results: Ten key challenges were identified: 1. data breaches and unauthorized access, 2. compliance with regulations such as HIPAA and GDPR, 3. data sovereignty and jurisdictional issues, 4. shared infrastructure vulnerabilities, 5. insider threats, 6. data loss and availability concerns, 7. inadequate security measures by cloud providers, 8. application vulnerabilities, 9. limited visibility and control, and 10. the complexity of cloud security management.

Conclusion: To fully benefit from cloud computing and AI, EQA providers must implement robust security practices, ensure regulatory compliance, and continuously monitor their environments. Proactive cybersecurity strategies are essential to safeguarding sensitive laboratory data and maintaining operational continuity and accreditation.