Pub Date : 2024-08-17DOI: 10.1515/labmed-2024-0062
Atiqah Mokhsin, Poonaresi Subramaniam, Sivasooriar Sivaneson, Nelson Nheu, Gobhy Ramaloo, Azana S. Hanifah, Sumitha B. Mahathevan, Mohanaraja Nadarajah, Gayathiri Sampasivam, Aletza Mohd Ismail, Thuhairah Abdul Rahman
Objectives Our study aimed to assess the stability of 26 biochemistry analytes in serum or plasma samples separated from blood samples centrifuged at different time intervals after collection, simulating sample transport via despatch delivery systems. Methods Blood from forty-one volunteers were collected using five serum separator tubes (SST) and five fluoride oxalate tubes (FOT) for each volunteer following written informed consent. Each of the five tubes in both groups of SST and FOT was centrifuged at one of the time intervals: 0.5 h, 4 h, 8 h, 12 and 24 h after collection. These samples were left standing prior to centrifugation at room temperature. We calculated the percentage difference for each analyte between the 0.5 h and other time intervals to assess analyte stability. The percentage difference was compared to the desirable specification for bias and reference change value (RCV). Results Mean concentration of serum potassium showed a significant increase in the percentage RCV after 8 h, while CKMB showed an increase after 12 h of delayed centrifugation compared to the baseline (0.5 h). There were no significant percentage RCV for the other analytes at all timelines. Conclusions Serum potassium and CKMB were stable up to 8 and 12 h of delayed centrifugation respectively whilst all other analytes appear stable up to 24 h, suggesting that sample transport delay of up to 8 h, with the condition that room temperature is maintained, may not have a significant impact on accuracy of the biochemistry/immunochemistry test results.
{"title":"Assessing the stability of uncentrifuged serum and plasma analytes at various post-collection intervals","authors":"Atiqah Mokhsin, Poonaresi Subramaniam, Sivasooriar Sivaneson, Nelson Nheu, Gobhy Ramaloo, Azana S. Hanifah, Sumitha B. Mahathevan, Mohanaraja Nadarajah, Gayathiri Sampasivam, Aletza Mohd Ismail, Thuhairah Abdul Rahman","doi":"10.1515/labmed-2024-0062","DOIUrl":"https://doi.org/10.1515/labmed-2024-0062","url":null,"abstract":"Objectives Our study aimed to assess the stability of 26 biochemistry analytes in serum or plasma samples separated from blood samples centrifuged at different time intervals after collection, simulating sample transport via despatch delivery systems. Methods Blood from forty-one volunteers were collected using five serum separator tubes (SST) and five fluoride oxalate tubes (FOT) for each volunteer following written informed consent. Each of the five tubes in both groups of SST and FOT was centrifuged at one of the time intervals: 0.5 h, 4 h, 8 h, 12 and 24 h after collection. These samples were left standing prior to centrifugation at room temperature. We calculated the percentage difference for each analyte between the 0.5 h and other time intervals to assess analyte stability. The percentage difference was compared to the desirable specification for bias and reference change value (RCV). Results Mean concentration of serum potassium showed a significant increase in the percentage RCV after 8 h, while CKMB showed an increase after 12 h of delayed centrifugation compared to the baseline (0.5 h). There were no significant percentage RCV for the other analytes at all timelines. Conclusions Serum potassium and CKMB were stable up to 8 and 12 h of delayed centrifugation respectively whilst all other analytes appear stable up to 24 h, suggesting that sample transport delay of up to 8 h, with the condition that room temperature is maintained, may not have a significant impact on accuracy of the biochemistry/immunochemistry test results.","PeriodicalId":55986,"journal":{"name":"Journal of Laboratory Medicine","volume":null,"pages":null},"PeriodicalIF":1.2,"publicationDate":"2024-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142220144","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-16DOI: 10.1515/labmed-2024-0118
Georg Hoffmann, Nina Allmeier, Modupe Kuti, Stefan Holdenrieder, Inga Trulson
Objectives Although there are several indirect methods that can be used to verify reference limits, they have a common weakness in that they assume a low proportion of pathological values. This paper investigates whether a Gaussian decomposition algorithm can identify the non-pathological fraction even if it is not the main subset of mixed data. Methods All investigations are carried out in the R programming environment. The mclust package is used for Gaussian mixture modelling via the expectation maximization (EM) algorithm. For right-skewed distributions, logarithms of the original values are taken to approximate the Gaussian model. We use the Bayesian information criterion (BIC) for evaluation of the results. The reflimR and refineR packages serve as comparison procedures. Results We generate synthetic data mixtures with known normal distributions to demonstrate the feasibility and reliability of our approach. Application of the algorithm to real data from a Nigerian and a German population produces results, which help to interpret reference intervals of reflimR and refineR that are obviously too wide. In the first example, the mclust analysis of hemoglobin in Nigerian women supports the medical hypothesis that an anemia rate of more than 50 % leads to falsely low reference limits. Our algorithm proposes various scenarios based on the BIC values, one of which suggests reference limits that are close to published data for Nigeria but significantly lower than those established for the Caucasian population. In the second example, the standard statistical analysis of creatine kinase in German patients with predominantly cardiac diseases yields a reference interval that is clearly too wide. With mclust we identify overlapping fractions that explain this false result. Conclusions Gaussian mixture modelling does not replace standard methods for reference interval estimation but is a valuable adjunct when these methods produce discrepant or implausible results.
{"title":"How Gaussian mixture modelling can help to verify reference intervals from laboratory data with a high proportion of pathological values","authors":"Georg Hoffmann, Nina Allmeier, Modupe Kuti, Stefan Holdenrieder, Inga Trulson","doi":"10.1515/labmed-2024-0118","DOIUrl":"https://doi.org/10.1515/labmed-2024-0118","url":null,"abstract":"Objectives Although there are several indirect methods that can be used to verify reference limits, they have a common weakness in that they assume a low proportion of pathological values. This paper investigates whether a Gaussian decomposition algorithm can identify the non-pathological fraction even if it is not the main subset of mixed data. Methods All investigations are carried out in the R programming environment. The mclust package is used for Gaussian mixture modelling via the expectation maximization (EM) algorithm. For right-skewed distributions, logarithms of the original values are taken to approximate the Gaussian model. We use the Bayesian information criterion (BIC) for evaluation of the results. The reflimR and refineR packages serve as comparison procedures. Results We generate synthetic data mixtures with known normal distributions to demonstrate the feasibility and reliability of our approach. Application of the algorithm to real data from a Nigerian and a German population produces results, which help to interpret reference intervals of reflimR and refineR that are obviously too wide. In the first example, the mclust analysis of hemoglobin in Nigerian women supports the medical hypothesis that an anemia rate of more than 50 % leads to falsely low reference limits. Our algorithm proposes various scenarios based on the BIC values, one of which suggests reference limits that are close to published data for Nigeria but significantly lower than those established for the Caucasian population. In the second example, the standard statistical analysis of creatine kinase in German patients with predominantly cardiac diseases yields a reference interval that is clearly too wide. With mclust we identify overlapping fractions that explain this false result. Conclusions Gaussian mixture modelling does not replace standard methods for reference interval estimation but is a valuable adjunct when these methods produce discrepant or implausible results.","PeriodicalId":55986,"journal":{"name":"Journal of Laboratory Medicine","volume":null,"pages":null},"PeriodicalIF":1.2,"publicationDate":"2024-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142220103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-12DOI: 10.1515/labmed-2024-0100
Inga Trulson, Stefan Holdenrieder, Georg Hoffmann
Objectives The study aims to acquaint readers with six widely used machine learning (ML) techniques (Principal Component Analysis (PCA), Uniform Manifold Approximation and Projection (UMAP), k-means, hierarchical clustering and the decision tree models (rpart and random forest)) that might be useful for the analysis of laboratory data. Methods Utilizing a recently validated data set from lung cancer diagnostics, we investigate how ML can support the search for a suitable tumor marker panel for the differentiation of small cell (SCLC) and non-small cell lung cancer (NSCLC). Results The ML techniques used here effectively helped to gain a quick overview of the data structures and provide initial answers to the clinical questions. Dimensionality reduction techniques such as PCA and UMAP offered insightful visualization and impression of the data structure, suggesting the existence of two tumor groups with a large overlap of largely inconspicuous values. This impression was confirmed by a cluster analysis with the k-means algorithm, indicative of unsupervised learning. For supervised learning, decision tree models like rpart or random forest demonstrated their utility in differential diagnosis of the two tumor types. The rpart model, which constructs binary decision trees based on the recursive partitioning algorithm, suggests a tree involving four serum tumor markers (STMs), which were confirmed by the random forest approach. Both highlighted pro-gastrin-releasing peptide (ProGRP), neuron specific enolase (NSE), cytokeratin-19 fragment (CYFRA 21-1) and cancer antigen (CA) 72-4 as key tumor markers, aligning with the outcomes of the initial statistical analysis. Cross-validation of the two proposals showed a higher area under the receiver operating characteristic (AUROC) curve of 0.95 with a 95 % confidence interval (CI) of 0.92–0.97 for the random forest model compared to an AUROC curve of 0.88 (95 % CI: 0.83–0.93). Conclusions ML can provide a useful overview of inherent medical data structures and distinguish significant from less pertinent features. While by no means replacing human medical and statistical expertise, ML can significantly accelerate the evaluation of medical data, supporting a more informed diagnostic dialogue between physicians and statisticians.
目的 本研究旨在让读者了解六种广泛使用的机器学习（ML）技术（主成分分析（PCA）、统一表层逼近和投影（UMAP）、k-均值、分层聚类和决策树模型（rpart 和随机森林）），这些技术可能对实验室数据分析有用。方法 利用最近验证的肺癌诊断数据集，我们研究了 ML 如何支持寻找合适的肿瘤标记物面板，以区分小细胞肺癌（SCLC）和非小细胞肺癌（NSCLC）。结果 这里使用的 ML 技术有效地帮助我们快速了解了数据结构，并为临床问题提供了初步答案。PCA 和 UMAP 等降维技术为数据结构提供了深入的可视化和印象，表明存在两个肿瘤组，其中有大量基本不明显的值重叠。使用 k-means 算法进行的聚类分析证实了这一印象，表明这是无监督学习。在监督学习方面，rpart 或随机森林等决策树模型在两种肿瘤类型的鉴别诊断中发挥了作用。基于递归分割算法构建二元决策树的 rpart 模型提出了一种涉及四种血清肿瘤标志物（STMs）的决策树，随机森林方法证实了这一点。这两种方法都强调促胃泌素释放肽（ProGRP）、神经元特异性烯醇化酶（NSE）、细胞角蛋白-19片段（CYFRA 21-1）和癌抗原（CA）72-4是关键的肿瘤标志物，与初步统计分析的结果一致。两种方案的交叉验证结果显示，随机森林模型的接收者操作特征曲线下面积（AUROC）为 0.95，95 % 置信区间（CI）为 0.92-0.97，而随机森林模型的接收者操作特征曲线下面积（AUROC）为 0.88（95 % 置信区间：0.83-0.93）。结论 ML 可以提供对固有医疗数据结构的有用概述，并区分重要特征和不太相关的特征。虽然 ML 无法取代人类的医学和统计专业知识，但它能大大加快医疗数据的评估速度，支持医生和统计学家之间进行更明智的诊断对话。
{"title":"Using machine learning techniques for exploration and classification of laboratory data","authors":"Inga Trulson, Stefan Holdenrieder, Georg Hoffmann","doi":"10.1515/labmed-2024-0100","DOIUrl":"https://doi.org/10.1515/labmed-2024-0100","url":null,"abstract":"Objectives The study aims to acquaint readers with six widely used machine learning (ML) techniques (Principal Component Analysis (PCA), Uniform Manifold Approximation and Projection (UMAP), k-means, hierarchical clustering and the decision tree models (rpart and random forest)) that might be useful for the analysis of laboratory data. Methods Utilizing a recently validated data set from lung cancer diagnostics, we investigate how ML can support the search for a suitable tumor marker panel for the differentiation of small cell (SCLC) and non-small cell lung cancer (NSCLC). Results The ML techniques used here effectively helped to gain a quick overview of the data structures and provide initial answers to the clinical questions. Dimensionality reduction techniques such as PCA and UMAP offered insightful visualization and impression of the data structure, suggesting the existence of two tumor groups with a large overlap of largely inconspicuous values. This impression was confirmed by a cluster analysis with the k-means algorithm, indicative of unsupervised learning. For supervised learning, decision tree models like rpart or random forest demonstrated their utility in differential diagnosis of the two tumor types. The rpart model, which constructs binary decision trees based on the recursive partitioning algorithm, suggests a tree involving four serum tumor markers (STMs), which were confirmed by the random forest approach. Both highlighted pro-gastrin-releasing peptide (ProGRP), neuron specific enolase (NSE), cytokeratin-19 fragment (CYFRA 21-1) and cancer antigen (CA) 72-4 as key tumor markers, aligning with the outcomes of the initial statistical analysis. Cross-validation of the two proposals showed a higher area under the receiver operating characteristic (AUROC) curve of 0.95 with a 95 % confidence interval (CI) of 0.92–0.97 for the random forest model compared to an AUROC curve of 0.88 (95 % CI: 0.83–0.93). Conclusions ML can provide a useful overview of inherent medical data structures and distinguish significant from less pertinent features. While by no means replacing human medical and statistical expertise, ML can significantly accelerate the evaluation of medical data, supporting a more informed diagnostic dialogue between physicians and statisticians.","PeriodicalId":55986,"journal":{"name":"Journal of Laboratory Medicine","volume":null,"pages":null},"PeriodicalIF":1.2,"publicationDate":"2024-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142220140","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-05DOI: 10.1515/labmed-2024-0083
Sandra Klawitter, Johannes Böhm, Alexander Tolios, Julian E. Gebauer
Objectives Reference intervals (RI) play a decisive role in the interpretation of medical laboratory results. An important step in the determination of RI is age- and sex specific partitioning, which is usually based on an empirical approach by graphical representation. In this study, we evaluate an automated machine learning approach. Methods This study uses pediatric data from the CALIPER RI (Canadian laboratory initiative on pediatric reference intervals) study. The calculation of potential partitions is carried out using a regression tree model included in the rpart package of the statistical programming language R. The Harris & Boyd method is used to compare the corresponding partitions suggested by rpart and CALIPER. For better comparability, the reference ranges of the partitions of both approaches are then calculated using reflimR. Results Most of the partitions suggested by rpart or CALIPER show sufficient heterogeneity among themselves to justify age- and/or sex-specific RI partitioning. With only few individual exceptions, both methods yield comparable results. The partitions of both approaches for albumin and γ-glutamyltransferase are very similar to each other. For creatinine rpart suggests a slightly earlier distinction between the sexes. Alkaline phosphatase shows the most pronounced differences. In addition to a considerable earlier sex split, rpart suggests different age intervals for both sexes, resulting in three partitions for females and four partitions for males. Conclusions Our findings indicate that the automated analysis provided by rpart yields results that comparable to traditional methods. Nevertheless, the medical plausibility of the automatic suggestions needs to be validated by human experts.
{"title":"Automated sex and age partitioning for the estimation of reference intervals using a regression tree model","authors":"Sandra Klawitter, Johannes Böhm, Alexander Tolios, Julian E. Gebauer","doi":"10.1515/labmed-2024-0083","DOIUrl":"https://doi.org/10.1515/labmed-2024-0083","url":null,"abstract":"Objectives Reference intervals (RI) play a decisive role in the interpretation of medical laboratory results. An important step in the determination of RI is age- and sex specific partitioning, which is usually based on an empirical approach by graphical representation. In this study, we evaluate an automated machine learning approach. Methods This study uses pediatric data from the CALIPER RI (Canadian laboratory initiative on pediatric reference intervals) study. The calculation of potential partitions is carried out using a regression tree model included in the <jats:monospace>rpart</jats:monospace> package of the statistical programming language R. The Harris & Boyd method is used to compare the corresponding partitions suggested by <jats:monospace>rpart</jats:monospace> and CALIPER. For better comparability, the reference ranges of the partitions of both approaches are then calculated using <jats:monospace>reflimR</jats:monospace>. Results Most of the partitions suggested by <jats:monospace>rpart</jats:monospace> or CALIPER show sufficient heterogeneity among themselves to justify age- and/or sex-specific RI partitioning. With only few individual exceptions, both methods yield comparable results. The partitions of both approaches for albumin and <jats:italic>γ</jats:italic>-glutamyltransferase are very similar to each other. For creatinine <jats:monospace>rpart</jats:monospace> suggests a slightly earlier distinction between the sexes. Alkaline phosphatase shows the most pronounced differences. In addition to a considerable earlier sex split, <jats:monospace>rpart</jats:monospace> suggests different age intervals for both sexes, resulting in three partitions for females and four partitions for males. Conclusions Our findings indicate that the automated analysis provided by <jats:monospace>rpart</jats:monospace> yields results that comparable to traditional methods. Nevertheless, the medical plausibility of the automatic suggestions needs to be validated by human experts.","PeriodicalId":55986,"journal":{"name":"Journal of Laboratory Medicine","volume":null,"pages":null},"PeriodicalIF":1.2,"publicationDate":"2024-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141937835","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-04DOI: 10.1515/labmed-2024-0067
Jun Lv, Yu Wan, Fei Jiang, Fei Fan
Objectives To contrast the level of lactate dehydrogenase (LDH) and its isoenzymes between general mycoplasma pneumoniae pneumonia (GMPP) and refractory mycoplasma pneumoniae pneumonia (RMPP) groups and to investigate their predictive value for RMPP in children. Methods A total of 160 children with GMPP and 100 children with RMPP were enrolled from August 2022 to April 2023 in our hospital. Serum LDH and its isoenzymes levels were assessed between the two groups. LDH and its isoenzymes were entered into multivariate logistic regression analysis to identify risk factors for RMPP, and variables with significance were used to analyze their diagnostic values for RMPP. ROC curves were drawn, and the AUC was calculated, with sensitivity and specificity obtained. Results Children with RMPP displayed more blatant inflammatory responses as well as more alarming imaging findings compared to those with GMPP. The levels of serum LDH and its isoenzymes in children with RMPP were significantly higher than those in children with GMPP. In the multivariate logistic regression analysis, LDH (OR=1.02, p<0.001), LDH2 (OR=1.05, p=0.010) and LDH5 (OR=1.04, p˂0.001) showed statistically significant differences. When the cut-off values were 372.5, 97.46, and 49.29 U/L respectively, the AUCs of LDH (sensitivity=0.80, specificity=0.89), LDH2 (sensitivity=0.83, specificity=0.71), and LDH5 (sensitivity=0.82, specificity=0.72) predicting RMPP were 0.91, 0.81, and 0.82, respectively. The AUC of [LDH + LDH5] (0.92) was the highest. Conclusions Serum LDH, LDH2, and LDH5 have good diagnostic values for RMPP and possess the potential to be biological markers in children with RMPP. And the predictive value is higher when used in combination.
{"title":"Serum LDH and its isoenzymes (LDH2 and LDH5) associated with predictive value for refractory mycoplasma pneumoniae pneumonia in children","authors":"Jun Lv, Yu Wan, Fei Jiang, Fei Fan","doi":"10.1515/labmed-2024-0067","DOIUrl":"https://doi.org/10.1515/labmed-2024-0067","url":null,"abstract":"Objectives To contrast the level of lactate dehydrogenase (LDH) and its isoenzymes between general mycoplasma pneumoniae pneumonia (GMPP) and refractory mycoplasma pneumoniae pneumonia (RMPP) groups and to investigate their predictive value for RMPP in children. Methods A total of 160 children with GMPP and 100 children with RMPP were enrolled from August 2022 to April 2023 in our hospital. Serum LDH and its isoenzymes levels were assessed between the two groups. LDH and its isoenzymes were entered into multivariate logistic regression analysis to identify risk factors for RMPP, and variables with significance were used to analyze their diagnostic values for RMPP. ROC curves were drawn, and the AUC was calculated, with sensitivity and specificity obtained. Results Children with RMPP displayed more blatant inflammatory responses as well as more alarming imaging findings compared to those with GMPP. The levels of serum LDH and its isoenzymes in children with RMPP were significantly higher than those in children with GMPP. In the multivariate logistic regression analysis, LDH (OR=1.02, p<0.001), LDH2 (OR=1.05, p=0.010) and LDH5 (OR=1.04, p˂0.001) showed statistically significant differences. When the cut-off values were 372.5, 97.46, and 49.29 U/L respectively, the AUCs of LDH (sensitivity=0.80, specificity=0.89), LDH2 (sensitivity=0.83, specificity=0.71), and LDH5 (sensitivity=0.82, specificity=0.72) predicting RMPP were 0.91, 0.81, and 0.82, respectively. The AUC of [LDH + LDH5] (0.92) was the highest. Conclusions Serum LDH, LDH2, and LDH5 have good diagnostic values for RMPP and possess the potential to be biological markers in children with RMPP. And the predictive value is higher when used in combination.","PeriodicalId":55986,"journal":{"name":"Journal of Laboratory Medicine","volume":null,"pages":null},"PeriodicalIF":1.2,"publicationDate":"2024-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141937816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-02DOI: 10.1515/labmed-2024-0052
Noel Stierlin, Andreas Hemmerle, Karin Jung, Jörg Thumfart, Martin Risch, Lorenz Risch
Objectives This study systematically compared the performance and comparability of two medical laboratory analytical instruments, the conventional wet chemistry analyzer (cobas) and the dry slide technology (Vitros), across various clinical chemistry assays. Methods The evaluation focused on assessing imprecision, inaccuracy, recovery, and method comparison using leftover patient serum samples. Results The results indicated good to very good agreement for most clinical chemistry analytes, with larger differences observed for comparison of serum patient samples on albumin and protein. Conclusions Understanding and acknowledging method-specific variations, are crucial for accurate result interpretation in clinical laboratories. This study contributes valuable insights to ongoing discussions on method standardization.
{"title":"Comparison of two different technologies measuring the same analytes in view of the In Vitro Diagnostica Regulation (IVDR)","authors":"Noel Stierlin, Andreas Hemmerle, Karin Jung, Jörg Thumfart, Martin Risch, Lorenz Risch","doi":"10.1515/labmed-2024-0052","DOIUrl":"https://doi.org/10.1515/labmed-2024-0052","url":null,"abstract":"Objectives This study systematically compared the performance and comparability of two medical laboratory analytical instruments, the conventional wet chemistry analyzer (cobas) and the dry slide technology (Vitros), across various clinical chemistry assays. Methods The evaluation focused on assessing imprecision, inaccuracy, recovery, and method comparison using leftover patient serum samples. Results The results indicated good to very good agreement for most clinical chemistry analytes, with larger differences observed for comparison of serum patient samples on albumin and protein. Conclusions Understanding and acknowledging method-specific variations, are crucial for accurate result interpretation in clinical laboratories. This study contributes valuable insights to ongoing discussions on method standardization.","PeriodicalId":55986,"journal":{"name":"Journal of Laboratory Medicine","volume":null,"pages":null},"PeriodicalIF":1.2,"publicationDate":"2024-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141882462","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-01DOI: 10.1515/labmed-2024-0076
Tobias Ueli Blatter, Christos Theodoros Nakas, Alexander Benedikt Leichtle
Objectives Reference intervals for the general clinical practice are expected to cover non-pathological values, but also reflect the underlying biological variation present in age- and gender-specific patient populations. Reference intervals can be inferred from routine patient data measured in high capacity using parametric approaches. Stratified reference distributions are obtained which may be transformed to normality via e.g. a Yeo-Johnson transformation. The estimation of the optimal transformation parameter for Yeo-Johnson through maximum likelihood can be highly influenced by the presence of outlying observations, resulting in biased reference interval estimates. Methods To reduce the influence of outlying observations on parametric reference interval estimation, a reweighted M-estimator approach for the Yeo-Johnson (YJ) transformation was utilised to achieve central normality in stratified reference populations for a variety of laboratory test results. The reweighted M-estimator for the YJ transformation offers a robust parametric approach to infer relevant reference intervals. Results The proposed method showcases robustness up to 15 % of outliers present in routine patient data, highlighting the applicability of the reweighted M-estimator in laboratory medicine. Furthermore, reference intervals are personalised based on the patients’ age and gender for a variety of analytes from routine patient data collected in a tertiary hospital, robustly reducing the dimensionality of the data for more data-driven approaches. Conclusions The method shows the advantages for estimating reference intervals directly and parametrically from routine patient data in order to provide expected reference ranges. This approach to locally inferred reference intervals allows a more nuanced comparison of patients’ test results.
目标 一般临床实践的参考区间应涵盖非病理值，但也要反映特定年龄和性别患者群体中存在的潜在生物变异。参考区间可通过参数方法从高容量测量的常规患者数据中推断出来。分层参考分布可通过杨-约翰逊（Yeo-Johnson）变换等方法转化为正态分布。通过最大似然法估计杨-约翰逊的最佳变换参数时，可能会受到离群观测数据的严重影响，导致参考区间估计值出现偏差。方法 为了减少离群观测数据对参数参考区间估计的影响，我们采用了一种针对杨-约翰逊（YJ）转换的再加权 M-估计方法，以实现各种实验室检验结果的分层参考人群的中心正态性。YJ 转换的再加权 M 估计器为推断相关参考区间提供了一种稳健的参数方法。结果 所提出的方法对常规患者数据中 15% 的异常值具有稳健性，突出了重加权 M 估计器在检验医学中的适用性。此外，根据患者的年龄和性别，对一家三甲医院收集的常规患者数据中的各种分析物进行了参考区间个性化处理，为更多数据驱动型方法降低了数据维度。结论 该方法显示了从常规患者数据中直接估算参考区间和参数化估算参考区间以提供预期参考范围的优势。这种局部推断参考区间的方法可以对患者的检测结果进行更细致的比较。
{"title":"Direct, age- and gender-specific reference intervals: applying a modified M-estimator of the Yeo-Johnson transformation to clinical real-world data","authors":"Tobias Ueli Blatter, Christos Theodoros Nakas, Alexander Benedikt Leichtle","doi":"10.1515/labmed-2024-0076","DOIUrl":"https://doi.org/10.1515/labmed-2024-0076","url":null,"abstract":"Objectives Reference intervals for the general clinical practice are expected to cover non-pathological values, but also reflect the underlying biological variation present in age- and gender-specific patient populations. Reference intervals can be inferred from routine patient data measured in high capacity using parametric approaches. Stratified reference distributions are obtained which may be transformed to normality via e.g. a Yeo-Johnson transformation. The estimation of the optimal transformation parameter for Yeo-Johnson through maximum likelihood can be highly influenced by the presence of outlying observations, resulting in biased reference interval estimates. Methods To reduce the influence of outlying observations on parametric reference interval estimation, a reweighted M-estimator approach for the Yeo-Johnson (YJ) transformation was utilised to achieve central normality in stratified reference populations for a variety of laboratory test results. The reweighted M-estimator for the YJ transformation offers a robust parametric approach to infer relevant reference intervals. Results The proposed method showcases robustness up to 15 % of outliers present in routine patient data, highlighting the applicability of the reweighted M-estimator in laboratory medicine. Furthermore, reference intervals are personalised based on the patients’ age and gender for a variety of analytes from routine patient data collected in a tertiary hospital, robustly reducing the dimensionality of the data for more data-driven approaches. Conclusions The method shows the advantages for estimating reference intervals directly and parametrically from routine patient data in order to provide expected reference ranges. This approach to locally inferred reference intervals allows a more nuanced comparison of patients’ test results.","PeriodicalId":55986,"journal":{"name":"Journal of Laboratory Medicine","volume":null,"pages":null},"PeriodicalIF":1.2,"publicationDate":"2024-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141882467","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-27DOI: 10.1515/labmed-2024-0051
Amani Al-Mekhlafi, Sandra Klawitter, Frank Klawonn
Objectives In the context of exploratory data analysis and machine learning, standardization of laboratory results is an important pre-processing step. Variable proportions of pathological results in routine datasets lead to changes of the mean (µ) and standard deviation (σ), and thus cause problems in the classical z-score transformation. Therefore, this study investigates whether the zlog transformation compensates these disadvantages and makes the results more meaningful from a medical perspective. Methods The results presented here were obtained with the statistical software environment R, and the underlying data set was obtained from the UC Irvine Machine Learning Repository. We compare the differences of the zlog and z-score transformation for five different dimension reduction methods, hierarchical clustering and four supervised classification methods. Results With the zlog transformation, we obtain better results in this study than with the z-score transformation for dimension reduction, clustering and classification methods. By compensating the disadvantages of the z-score transformation, the zlog transformation allows more meaningful medical conclusions. Conclusions We recommend using the zlog transformation of laboratory results for pre-processing when exploratory data analysis and machine learning techniques are applied.
{"title":"Standardization with zlog values improves exploratory data analysis and machine learning for laboratory data","authors":"Amani Al-Mekhlafi, Sandra Klawitter, Frank Klawonn","doi":"10.1515/labmed-2024-0051","DOIUrl":"https://doi.org/10.1515/labmed-2024-0051","url":null,"abstract":"Objectives In the context of exploratory data analysis and machine learning, standardization of laboratory results is an important pre-processing step. Variable proportions of pathological results in routine datasets lead to changes of the mean (<jats:italic>µ</jats:italic>) and standard deviation (<jats:italic>σ</jats:italic>), and thus cause problems in the classical z-score transformation. Therefore, this study investigates whether the zlog transformation compensates these disadvantages and makes the results more meaningful from a medical perspective. Methods The results presented here were obtained with the statistical software environment R, and the underlying data set was obtained from the UC Irvine Machine Learning Repository. We compare the differences of the zlog and z-score transformation for five different dimension reduction methods, hierarchical clustering and four supervised classification methods. Results With the zlog transformation, we obtain better results in this study than with the z-score transformation for dimension reduction, clustering and classification methods. By compensating the disadvantages of the z-score transformation, the zlog transformation allows more meaningful medical conclusions. Conclusions We recommend using the zlog transformation of laboratory results for pre-processing when exploratory data analysis and machine learning techniques are applied.","PeriodicalId":55986,"journal":{"name":"Journal of Laboratory Medicine","volume":null,"pages":null},"PeriodicalIF":1.2,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141504945","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-19DOI: 10.1515/labmed-2024-0037
Xueying Lin, Qiaofang Yan, Yuanyuan Du, Jianbing Wang, Di Huang, Jun Yan, Min Zhan, Pengwei Zhang, Jingyu Cheng, Qiaoxuan Zhang, Xianzhang Huang, Liqiao Han
Abstract Objectives The accuracy of blood glucose measurement in clinical laboratories is vital for diabetes diagnosis. Trueness Verification Plan was carried out and analyzed for evaluating the standardization of serum glucose among clinical laboratories. Methods Trueness verification samples were distributed to clinical laboratories for three days measurement, and their target values were assigned by two certified reference laboratories. The relative bias, coefficient of variation (CV), and total error (TE) for each clinical laboratory were calculated and analyzed. Moreover, the Six Sigma metrics and Quality Goal Index were utilized to reflect the measurement quality of the clinical laboratories. Results The pass rates evaluated by bias, CV, and TE ranged from 45.2 % to 64.8 %, 96.8 %–98.9 %, and 83.9 %–97.1 % over the six years. The matched systems used in clinical laboratories demonstrated better accuracy than the un-matched systems. The pass rate by bias of hexokinase method is 53.1 %–78.6 %, while the glucose oxidase method is 29.2 %–52.2 %. Overall, 74.2 %–85.7 % of clinical laboratories achieved an acceptable level (both σ>3), and 35.2 %–61.4 % of laboratories reached a “world-class” level (both σ>6). Conclusions The quality for serum glucose measurement has been greatly improved. However, standardization among clinical systems still needs to be further promoted.
{"title":"Evaluation for serum glucose standardization in clinical laboratories of Southern China by consecutive 6 years proficiency testing based on JCTLM-recommended reference methods","authors":"Xueying Lin, Qiaofang Yan, Yuanyuan Du, Jianbing Wang, Di Huang, Jun Yan, Min Zhan, Pengwei Zhang, Jingyu Cheng, Qiaoxuan Zhang, Xianzhang Huang, Liqiao Han","doi":"10.1515/labmed-2024-0037","DOIUrl":"https://doi.org/10.1515/labmed-2024-0037","url":null,"abstract":"Abstract Objectives The accuracy of blood glucose measurement in clinical laboratories is vital for diabetes diagnosis. Trueness Verification Plan was carried out and analyzed for evaluating the standardization of serum glucose among clinical laboratories. Methods Trueness verification samples were distributed to clinical laboratories for three days measurement, and their target values were assigned by two certified reference laboratories. The relative bias, coefficient of variation (CV), and total error (TE) for each clinical laboratory were calculated and analyzed. Moreover, the Six Sigma metrics and Quality Goal Index were utilized to reflect the measurement quality of the clinical laboratories. Results The pass rates evaluated by bias, CV, and TE ranged from 45.2 % to 64.8 %, 96.8 %–98.9 %, and 83.9 %–97.1 % over the six years. The matched systems used in clinical laboratories demonstrated better accuracy than the un-matched systems. The pass rate by bias of hexokinase method is 53.1 %–78.6 %, while the glucose oxidase method is 29.2 %–52.2 %. Overall, 74.2 %–85.7 % of clinical laboratories achieved an acceptable level (both σ>3), and 35.2 %–61.4 % of laboratories reached a “world-class” level (both σ>6). Conclusions The quality for serum glucose measurement has been greatly improved. However, standardization among clinical systems still needs to be further promoted.","PeriodicalId":55986,"journal":{"name":"Journal of Laboratory Medicine","volume":null,"pages":null},"PeriodicalIF":1.2,"publicationDate":"2024-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141334825","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-18DOI: 10.1515/labmed-2024-0007
Shu-Qian Cai, Tingting Xia, Xiaoping Xu
Abstract Objectives To explore the usefulness of neutrophil-to-lymphocyte count ratio (NLR), procalcitonin (PCT), and interleukin-6 (IL-6) for the severity assessment of bacterial sepsis. Methods This study enrolled 100 patients with bacterial sepsis (disease group) who presented to Jinhua Central Hospital between March 2022 and March 2023 and 90 healthy individuals (control group). The patients were categorized into sepsis (64 cases), severe sepsis (18 cases), and septic shock (18 cases) groups according to the disease severity. The groups were compared in terms of the NLR, PCT, and IL-6, as well as the usefulness of these parameters, both alone and in combination, for the severity assessment of bacterial sepsis. Results The NLR, PCT, and IL-6 levels were significantly different among the three groups, with increasing values corresponding with disease aggravation. The area under the curve (AUC) values of the combinations of NLR, PCT, and IL-6 levels were higher than those of single markers. The sensitivity and AUC value of the combination of PCT and IL-6 levels were the highest (0.87), with a similar AUC value of the combination of NLR, PCT, and IL-6 (0.865); however, the specificity was significantly improved with the latter (0.938 vs. 0.859). Conclusions NLR, PCT, and IL-6 levels are significantly increased in bacterial sepsis, and the combination of PCT, and IL-6 levels can improve the sensitivity of the evaluation ability for severe sepsis, and is more economical.
{"title":"Usefulness of neutrophil-to-lymphocyte count ratio, procalcitonin, and interleukin-6 for severity assessment of bacterial sepsis","authors":"Shu-Qian Cai, Tingting Xia, Xiaoping Xu","doi":"10.1515/labmed-2024-0007","DOIUrl":"https://doi.org/10.1515/labmed-2024-0007","url":null,"abstract":"Abstract Objectives To explore the usefulness of neutrophil-to-lymphocyte count ratio (NLR), procalcitonin (PCT), and interleukin-6 (IL-6) for the severity assessment of bacterial sepsis. Methods This study enrolled 100 patients with bacterial sepsis (disease group) who presented to Jinhua Central Hospital between March 2022 and March 2023 and 90 healthy individuals (control group). The patients were categorized into sepsis (64 cases), severe sepsis (18 cases), and septic shock (18 cases) groups according to the disease severity. The groups were compared in terms of the NLR, PCT, and IL-6, as well as the usefulness of these parameters, both alone and in combination, for the severity assessment of bacterial sepsis. Results The NLR, PCT, and IL-6 levels were significantly different among the three groups, with increasing values corresponding with disease aggravation. The area under the curve (AUC) values of the combinations of NLR, PCT, and IL-6 levels were higher than those of single markers. The sensitivity and AUC value of the combination of PCT and IL-6 levels were the highest (0.87), with a similar AUC value of the combination of NLR, PCT, and IL-6 (0.865); however, the specificity was significantly improved with the latter (0.938 vs. 0.859). Conclusions NLR, PCT, and IL-6 levels are significantly increased in bacterial sepsis, and the combination of PCT, and IL-6 levels can improve the sensitivity of the evaluation ability for severe sepsis, and is more economical.","PeriodicalId":55986,"journal":{"name":"Journal of Laboratory Medicine","volume":null,"pages":null},"PeriodicalIF":1.2,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141334919","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：