Clinical chemistry and laboratory medicine最新文献

筛选
英文 中文
Validation of (self-collected) capillary blood using a topper collection system as alternative for venous sampling for 15 common clinical chemistry analytes. 验证(自行采集)毛细管血使用顶部采集系统作为15种常见临床化学分析的静脉采样的替代。
IF 3.8 2区 医学
Clinical chemistry and laboratory medicine Pub Date : 2025-06-02 DOI: 10.1515/cclm-2025-0148
Rümeysa Geujar, Dilana Treffers, Maaike Roelofs, Anna van Dongen, Kalpana Ramkisoensing, Rixt Even, Huub H van Rossum
{"title":"Validation of (self-collected) capillary blood using a topper collection system as alternative for venous sampling for 15 common clinical chemistry analytes.","authors":"Rümeysa Geujar, Dilana Treffers, Maaike Roelofs, Anna van Dongen, Kalpana Ramkisoensing, Rixt Even, Huub H van Rossum","doi":"10.1515/cclm-2025-0148","DOIUrl":"https://doi.org/10.1515/cclm-2025-0148","url":null,"abstract":"<p><strong>Objectives: </strong>Home blood sampling for clinical purposes has gained much interest. This study validates the recently developed Topper-based capillary blood collection procedure for 15 commonly used chemistry analytes.</p><p><strong>Methods: </strong>A total of 120 study participants (21 healthy volunteers and 99 patients) were enrolled. A venous sample was obtained and then participants were asked to self-collect blood by the Topper system (SelfSafeSure Blood Collection Devices). Collected sera were analyzed for 15 common clinical chemistry analytes and the serum indices on a Cobas Pro (Roche) system. Spearman correlations, Passing-Bablok regression analysis, and Bland-Altman difference analysis were performed. Comparability was determined using allowable bias criteria based on biological variation (EFLM database).</p><p><strong>Results: </strong>In 113 out of 120 (94 %) self-collections a sample was obtained that allowed for the analysis of at least one analyte. Bland-Altman difference analysis showed that glucose and uncorrected AST did not meet the minimum bias criterion, creatinine and albumin were within the minimum bias criterion and urea and calcium were within the desirable bias criterion. ALP, corrected AST, ALT, total- HDL- and LDL-cholesterol, CRP, GGT, total protein and triglycerides were all within the optimal bias criterion.</p><p><strong>Conclusions: </strong>Our study demonstrates that self-collected capillary blood can be used as a reliable alternative to venous sampling for most, if not all, analytes studied. Based on the analyte stability prior to sample processing, home sampling appears to be a reliable sampling option for a selection of these analytes.</p>","PeriodicalId":10390,"journal":{"name":"Clinical chemistry and laboratory medicine","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144207846","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Arivale is dead ‒ Hooke is alive. 阿里维尔死了,胡克还活着。
IF 3.8 2区 医学
Clinical chemistry and laboratory medicine Pub Date : 2025-05-30 DOI: 10.1515/cclm-2025-0443
Miyo K Chatanaka, Eleftherios P Diamandis
{"title":"Arivale is dead ‒ Hooke is alive.","authors":"Miyo K Chatanaka, Eleftherios P Diamandis","doi":"10.1515/cclm-2025-0443","DOIUrl":"https://doi.org/10.1515/cclm-2025-0443","url":null,"abstract":"","PeriodicalId":10390,"journal":{"name":"Clinical chemistry and laboratory medicine","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144186723","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
No fault or negligence after an adverse analytical finding due to a contaminated supplement: mission impossible. Two examples involving trimetazidine. 没有过错或疏忽后,不利的分析发现,由于污染的补充:不可能的任务。两个涉及曲美他嗪的例子。
IF 3.8 2区 医学
Clinical chemistry and laboratory medicine Pub Date : 2025-05-30 DOI: 10.1515/cclm-2025-0549
Pascal Kintz
{"title":"No fault or negligence after an adverse analytical finding due to a contaminated supplement: mission impossible. Two examples involving trimetazidine.","authors":"Pascal Kintz","doi":"10.1515/cclm-2025-0549","DOIUrl":"https://doi.org/10.1515/cclm-2025-0549","url":null,"abstract":"<p><p>Since several years, sports authorities, including national anti-doping organisations and the World Anti-Doping Agency (WADA) have published that the consumption of food supplements can be at risk for athletes due to potential contamination by prohibited substances. Despite these warnings, supplements are largely used by elite athletes and doping violations involving supplements are weekly reported in the media. Anabolic steroids, selective androgen receptor modulators (SARMs), metabolic modulators, stimulants and diuretics are among the most frequently detected substances in contaminated supplements. The author was involved in 2 cases where trimetazidine was identified as the source of the doping violation but the sport authorities sentenced both athletes. Case 1: trimetazidine in urine at 0.1 ng/mL; trimetazidine negative in 3 × 2 cm hair segments (LOQ at 1 ng/g [pg/mg]); trimetazidine at 4 ng/tablet in the supplement. Case 2: trimetazidine in urine at 0.1 and 1.6 ng/mL on 2 occasions; trimetazidine negative in 6 × 1 cm hair segments (LOQ at 1 ng/g [pg/mg]); trimetazidine at 7.2 μg/tablet in the supplement. Despite all pharmacological parameters were consistent with minute amounts of trimetazidine intake during a scenario of proven contamination, the presence of trimetazidine in the urine samples was recognized as an anti-doping rule violation and the athletes were sanctioned with a period of ineligibility of 6 months. From a forensic perspective, contamination is not doping. To avoid these conflicting issues with contaminations, the debate should move to the interest of using hair test results and the need of implementing a threshold for reporting the presence of a drug in urine at very low concentration.</p>","PeriodicalId":10390,"journal":{"name":"Clinical chemistry and laboratory medicine","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144207845","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Immunochemical measurement of urinary free light chains and Bence Jones proteinuria. 尿游离轻链和本·琼斯蛋白尿的免疫化学测定。
IF 3.8 2区 医学
Clinical chemistry and laboratory medicine Pub Date : 2025-05-30 DOI: 10.1515/cclm-2025-0265
Laura Michetti, Rudi Ravasio, Roberto Marozzi, Ester Antelmi, Arianna Ghirardi, Greta Bolzoni
{"title":"Immunochemical measurement of urinary free light chains and Bence Jones proteinuria.","authors":"Laura Michetti, Rudi Ravasio, Roberto Marozzi, Ester Antelmi, Arianna Ghirardi, Greta Bolzoni","doi":"10.1515/cclm-2025-0265","DOIUrl":"https://doi.org/10.1515/cclm-2025-0265","url":null,"abstract":"<p><strong>Objectives: </strong>To evaluate the characteristics of an immunochemical urinary free light chains (uFLC) test in screening and quantifying Bence Jones proteinuria (BJP) and its potential application in laboratory practice as an alternative to the gold standard method, urine immunofixation (UIFE) and densitometric quantification on 24 h urine sample.</p><p><strong>Methods: </strong>A total of 300 subjects were divided into five groups: controls, patients with BJP Kappa and Lambda present in trace amounts not densitometrically quantifiable, and patients with BJP Kappa and Lambda present in densitometrically measurable quantities. The Wilcoxon-Mann-Whitney test was applied to compare uFLC in controls and BJP trace samples. The correlation between immunochemical and densitometric measurements was assessed using Spearman's correlation coefficient, and agreement was evaluated with Bland-Altman plots. Samples were also stratified by estimated glomerular filtration rate (eGFR) and total urinary proteins (TUP).</p><p><strong>Results: </strong>Despite significant differences between the median values of uFLC measures in controls and BJP in trace sample groups, using the uFLC upper reference ranges would have led to over 50 % false negative results. Although a strong correlation existed between the two methods, turbidimetry consistently overestimated BJP levels.</p><p><strong>Conclusions: </strong>uFLC turbidimetric measurements cannot accurately differentiate negative samples from those containing trace BJP, lacking the sensitivity required for clinical use. UIFE has shown greater sensitivity compared to turbidimetry. In monitoring, the systematic overestimation in the quantification of BJP, aggravated in cases of reduced renal function or high TUP concentration, makes it challenging for clinicians to evaluate therapeutic efficacy since the decision thresholds outlined in clinical guidelines are based on densitometric measurements.</p>","PeriodicalId":10390,"journal":{"name":"Clinical chemistry and laboratory medicine","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144207844","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Error in prostate-specific antigen levels after prostate cancer treatment with radical prostatectomy. 前列腺癌根治性前列腺切除术后前列腺特异性抗原水平的误差。
IF 3.8 2区 医学
Clinical chemistry and laboratory medicine Pub Date : 2025-05-29 DOI: 10.1515/cclm-2025-0430
Remy J H Martens, Bart E P B Ballieux, Ronald Bos, Floris Helmich
{"title":"Error in prostate-specific antigen levels after prostate cancer treatment with radical prostatectomy.","authors":"Remy J H Martens, Bart E P B Ballieux, Ronald Bos, Floris Helmich","doi":"10.1515/cclm-2025-0430","DOIUrl":"https://doi.org/10.1515/cclm-2025-0430","url":null,"abstract":"","PeriodicalId":10390,"journal":{"name":"Clinical chemistry and laboratory medicine","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144181327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Machine learning algorithms with body fluid parameters: an interpretable framework for malignant cell screening in cerebrospinal fluid. 具有体液参数的机器学习算法:脑脊液中恶性细胞筛选的可解释框架。
IF 3.8 2区 医学
Clinical chemistry and laboratory medicine Pub Date : 2025-05-28 DOI: 10.1515/cclm-2025-0302
Xianfei Ye, Xinfeng Zhao, Yinyu Lou, Hanqi Pan, Yunying Chen
{"title":"Machine learning algorithms with body fluid parameters: an interpretable framework for malignant cell screening in cerebrospinal fluid.","authors":"Xianfei Ye, Xinfeng Zhao, Yinyu Lou, Hanqi Pan, Yunying Chen","doi":"10.1515/cclm-2025-0302","DOIUrl":"https://doi.org/10.1515/cclm-2025-0302","url":null,"abstract":"<p><strong>Objectives: </strong>This study aimed to develop and validate a machine learning (ML) model utilizing cerebrospinal fluid (CSF) body fluid parameters from hematology analyzers to screen for malignant cells.</p><p><strong>Methods: </strong>We analyzed 643 consecutive CSF samples from patients with central nervous system symptoms, with 191 samples classified as positive for malignant cells based on cytological examination, for model derivation. Body fluid parameters were measured using the body fluid mode of a hematology analyzer. Least Absolute Shrinkage and Selection Operator (LASSO) regression was applied to identify predictive biomarkers, followed by performance evaluations of six ML algorithms. Model interpretability was assessed using SHapley Additive exPlanations (SHAP). The selected model was also externally validated with an additional 136 CSF samples.</p><p><strong>Results: </strong>The median leukocyte (WBC) and total nucleated cell (TNC) counts in the cytology-positive samples were significantly lower than those in the cytology-negative samples (5.4 vs. 31.8 and 7.4 vs. 32.6, respectively, p<0.001). The support vector machine (SVM) model achieved the highest area under the curve (AUC) of 0.899 (SD: 0.035) and the highest sensitivity of 0.827 (SD: 0.059) in internal validation. SHAP analysis identified the percentage of high fluorescence cells and monocytes as the two most significant predictors, both positively correlated with malignant cell outcomes. External validation demonstrated a comparable AUC and sensitivity, confirming the model's generalizability.</p><p><strong>Conclusions: </strong>We developed an ML model that predicts cytological outcomes in CSF using routinely available body fluid parameters. The model demonstrated consistent performance during external validation.</p>","PeriodicalId":10390,"journal":{"name":"Clinical chemistry and laboratory medicine","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144180782","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Commutability evaluation of glycated albumin candidate EQA materials. 糖化白蛋白候选EQA材料的交换性评价。
IF 3.8 2区 医学
Clinical chemistry and laboratory medicine Pub Date : 2025-05-28 DOI: 10.1515/cclm-2025-0507
Rui Wu, Huijuan Zhou, Xiaerbanu Nizhamnuding, Anqi Pan, Hao Zheng, Jiangtao Zhang, Haijian Zhao, Jing Wang, Tianjiao Zhang, Chuanbao Zhang
{"title":"Commutability evaluation of glycated albumin candidate EQA materials.","authors":"Rui Wu, Huijuan Zhou, Xiaerbanu Nizhamnuding, Anqi Pan, Hao Zheng, Jiangtao Zhang, Haijian Zhao, Jing Wang, Tianjiao Zhang, Chuanbao Zhang","doi":"10.1515/cclm-2025-0507","DOIUrl":"https://doi.org/10.1515/cclm-2025-0507","url":null,"abstract":"<p><strong>Objectives: </strong>In clinical practice, glycated albumin (GA), as a key biomarker reflecting the level of blood glucose control in the short and medium term, has been widely used in the diagnosis and efficacy monitoring of diabetes. However, the consistency of GA results between different detection systems remains one of the current challenges in laboratory medicine. we evaluated the commutability of 10 evaluated samples to identify candidate EQA materials for GA measurement.</p><p><strong>Methods: </strong>According to the Clinical and Laboratory Standards Institute (CLSI) document EP14-A3 and the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) commutability evaluation program, we used the established isotope dilution liquid chromatography-tandem mass spectrometry (ID LC-MS/MS) method for serum glycated albumin determination as the comparative method, and six different commercial kits used enzyme assay as the evaluating method. A total of 40 clinical sample serums, 9 pregnancy sample serums, and 10 evaluated samples were analyzed.</p><p><strong>Results: </strong>For the CLSI approach, pooled serum samples (cRM1-3) at medium concentrations (cRM2) are commutable in 5/6 kits. For the IFCC approach, none of samples commutable.</p><p><strong>Conclusions: </strong>The commutability evaluation results of the two approaches were different, and the cRM2 has commutability in most enzymatic kits. It can be a commutable material for the EQA project.</p>","PeriodicalId":10390,"journal":{"name":"Clinical chemistry and laboratory medicine","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144180871","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Understanding the circulating forms of cardiac troponin: insights for clinical practice. 了解心脏肌钙蛋白的循环形式:对临床实践的见解。
IF 3.8 2区 医学
Clinical chemistry and laboratory medicine Pub Date : 2025-05-28 DOI: 10.1515/cclm-2024-1236
Panpan Lv, Peng Zhang, Yuxiang Dai, Junbo Ge
{"title":"Understanding the circulating forms of cardiac troponin: insights for clinical practice.","authors":"Panpan Lv, Peng Zhang, Yuxiang Dai, Junbo Ge","doi":"10.1515/cclm-2024-1236","DOIUrl":"https://doi.org/10.1515/cclm-2024-1236","url":null,"abstract":"<p><p>Ischemic heart disease remains the leading cause of death globally, with acute coronary syndrome (ACS) being the main reason for emergency hospital admissions and thus representing a significant health care issue worldwide. Cardiac troponin I (cTnI) and cardiac troponin T (cTnT) are widely recognized biomarkers of cardiomyocyte injury and the gold-standard biomarkers for diagnosing myocardial infarction (MI). High-sensitivity cardiac troponin (hs-cTn) assays have the ability to accurately detect low cTn concentrations and document minor increases. However, in addition to MI, various other pathophysiological states can trigger elevated cardiac troponin levels, thus creating potential challenges in the diagnostic process. As cTn released into the bloodstream exists in heterogeneous forms, improving our understanding and accurately characterizing these forms across various etiologies might hold clinical significance. In this review, we add to the field by offering an overview of research on possible circulating forms of cTn, the mechanisms of cTn elevation, and the clinical significance of cTn following conditions such as MI, endurance exercise, and chronic kidney disease, thus highlighting the importance and challenge of understanding of the circulating forms of cTn and possible strategies for cTn immunodetection.</p>","PeriodicalId":10390,"journal":{"name":"Clinical chemistry and laboratory medicine","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144172886","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Regulating the future of laboratory medicine: European regulatory landscape of AI-driven medical device software in laboratory medicine. 规范实验室医学的未来:欧洲实验室医学中人工智能驱动的医疗设备软件的监管格局。
IF 3.8 2区 医学
Clinical chemistry and laboratory medicine Pub Date : 2025-05-28 DOI: 10.1515/cclm-2025-0482
Hikmet Can Çubukçu, Guilaine Boursier, Solveig Linko, Francisco A Bernabeu-Andreu, Pika Meško Brguljan, Katerina Tosheska-Trajkovska, Duilio Brugnoni, Neda Milinkovic, Andrea Padoan, Marc Thelen
{"title":"Regulating the future of laboratory medicine: European regulatory landscape of AI-driven medical device software in laboratory medicine.","authors":"Hikmet Can Çubukçu, Guilaine Boursier, Solveig Linko, Francisco A Bernabeu-Andreu, Pika Meško Brguljan, Katerina Tosheska-Trajkovska, Duilio Brugnoni, Neda Milinkovic, Andrea Padoan, Marc Thelen","doi":"10.1515/cclm-2025-0482","DOIUrl":"https://doi.org/10.1515/cclm-2025-0482","url":null,"abstract":"<p><p>Artificial intelligence (AI) is rapidly transforming laboratory medicine, impacting medical devices and healthcare practices. Despite these advancements, AI-based medical device software (MDSW) introduces a new layer of complexity in regulatory compliance. This paper outlines the regulatory landscape for MDSW and AI-driven MDSW, clarifying the responsibilities of laboratory professionals and manufacturers under the <i>In Vitro</i> Diagnostic Regulation (IVDR), ISO 15189:2022, and the Artificial Intelligence Act. An analysis of 89 MDSWs approved under the IVDR, derived from the European Database on Medical Devices (EUDAMED) reveals a diverse landscape of applications, ranging from digital pathology and molecular diagnostics to laboratory automation and clinical decision support. While Germany currently dominates the EU market for these devices, and the majority of approved MDSW remain non-AI driven and classified as low-risk, the increasing presence of AI-powered Class C devices underscores the growing potential of software in complex diagnostic scenarios. However, realizing the full potential of AI in laboratory medicine requires careful navigation of the evolving regulatory landscape. Key challenges persist, including defining intended use, ensuring robust clinical evidence, mitigating data bias, and establishing rigorous post-market surveillance. Balancing regulatory oversight with innovation is critical to fostering the development of trustworthy AI systems without stifling progress. As regulatory frameworks continue to evolve, establishing clear validation methodologies and transparent compliance pathways will be essential to unlocking the full potential of AI in laboratory medicine while ensuring the highest standards of safety and clinical effectiveness.</p>","PeriodicalId":10390,"journal":{"name":"Clinical chemistry and laboratory medicine","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144179774","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Impact of analytical bias on machine learning models for sepsis prediction using laboratory data. 分析偏差对脓毒症实验室数据预测机器学习模型的影响。
IF 3.8 2区 医学
Clinical chemistry and laboratory medicine Pub Date : 2025-05-28 DOI: 10.1515/cclm-2025-0491
Meryem Rumeysa Yesil, Ilaria Talli, Michela Pelloso, Chiara Cosma, Elisa Pangrazzi, Mario Plebani, Yasemin Ustundag, Andrea Padoan
{"title":"Impact of analytical bias on machine learning models for sepsis prediction using laboratory data.","authors":"Meryem Rumeysa Yesil, Ilaria Talli, Michela Pelloso, Chiara Cosma, Elisa Pangrazzi, Mario Plebani, Yasemin Ustundag, Andrea Padoan","doi":"10.1515/cclm-2025-0491","DOIUrl":"https://doi.org/10.1515/cclm-2025-0491","url":null,"abstract":"<p><strong>Objectives: </strong>Machine learning (ML) models, using laboratory data, support early sepsis prediction. However, analytical bias in laboratory measurements can compromise their performance and validity in real-world settings. We aimed to evaluate how analytically acceptable bias may affect the validity and generalizability of ML models trained on laboratory data.</p><p><strong>Methods: </strong>A support vector machine model (SVM) for sepsis prediction was developed using complete blood count and erythrocyte sedimentation rate data from outpatients (CS, n=104) and patients from acute inflammatory status wards (SS, n=107). Twenty-six combinations were derived by white blood cells (WBC), platelets (PLT), and erythrocyte sedimentation rate (ESR) biases from analytical performance specifications (APS). The diagnostic performances of the 26 conditions tested were compared to the original dataset.</p><p><strong>Results: </strong>SVM performance of the original dataset was AUC 90.6 % [95 %CI: 80.6-98.7 %]. Minimum, desirable and optimum acceptable biases for WBC were 7.7 , 5.1 and 2.6 %, respectively, for PLT were 6.7 , 4.5 and 2.2 %, respectively and for ESR were 31.6 , 21.1 and 10.5 %, respectively. Across all conditions, AUC varied from 89.8 % [95 %CI: 79.0-97.7 %] (for PLT bias -6.7 %), to 89.5 % [95 %CI: 79.1-98.0 %] (for ESR Bias +31.6 %) to 90.4 % [95 %CI: 79.3-98.4 %] (for WBC Bias -5.1 %). Using a combination of biases, the lowest AUC was 87.8 % [95 %CI: 75.9-96.6 %]. No statistically significant differences were observed for AUC (p>0.05).</p><p><strong>Conclusions: </strong>Bias can influence model performance depending on the parameters and their combinations. Developing new validation strategies to assess the impact of analytical bias on laboratory data in ML models could improve their reliability.</p>","PeriodicalId":10390,"journal":{"name":"Clinical chemistry and laboratory medicine","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144180978","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信