ABSTRACT
Objective:To establish and evaluate a new real-time quality control method that can identify the random errors by using the backpropagation neural network (BPNN) algorithm and taking blood glucose test as an example.Methods:A total of 219 000 blood glucose results measured by Siemens advia 2 400 analytical system from January 2019 to July 2020 and derived from Laboratory Information System of Beijing Chaoyang Hospital Laboratory Department was regarded as the unbiased data of our study. Six deviations with different sizes were introduced to generate the corresponding biased data. With each biased data, BPNN and MovSD algorithms were used and tested, and then evaluated by traceability method and clinical method.Results:For BPNN algorithm, the block size was pre-set to 10 and the false-positive rate in all biases was within 0.1%. For MovSD, however, the optimal block size and exclusive limit were 150 and 10% separately and its false-positive rate in all biases was 0.38%, which was 0.28% higher than BPNN. Especially, for the least two error factors of 0.5 and 1, all the random errors were not detected by MovSD; for the error factor larger than 1, random errors could be detected by MovSD but the MNPed was higher than that of BPNN under all deviations. The difference was up to 91.67 times. 460 000 reference data were produced by traceability procedure. The uncertainty of BPNN algorithm evaluated by these reference data was only 0.078%.Conclusion:A real-time quality control method based on BPNN algorithm was successfully established to identify random errors in analytical phase, which was more efficient than MovSD method and provided a new idea and method for the identification of random errors in clinical practice.
ABSTRACT
Objective:To investigate the application value of establishing the differential diagnosis model of pulmonary tuberculosis using routine laboratory data.Methods:The retrospective study was conducted. The routine laboratory data of newly diagnosed patients with pulmonary tuberculosis and other pulmonary diseases in Beijng Jishuitan Hospital and Beijing Hepingli Hospital from May 2015 to November 2021were collected. According to the random numbers showed in the computer, all the 11516 patients were divided into training dataset and test dataset with a ratio of 9∶1. Four machine learning algorithms, Support Vector Machine, Random Forest, K-Nearest Neighbor and Logistic Regression, were used to build models and select features. The diagnostic accuracy of each model was verified by using the 10-fold cross-validation method and the performance of each model was evaluated by using the receptor operator of characteristic (ROC) curve.Results:Random Forest was selected as the optimal machine learning algorithm to build the best feature model in the study. According to importance scale of factors, the differential diagnosis model of pulmonary tuberculosis consisting of 37 non-specific test indexes. In the validation set and test set the accuracy and area under curve (AUC) of the models were 0.747 and 0.736, the sensitivity, specificity and accuracy were 68.03% and 68.75%, 70.91% and 67.90%, 70.30% and 68.12%, respectively.Conclusion:A key tool in the differential diagnosis model of pulmonary tuberculosis was established by routine laboratory data in combination with machine learning. The results of this study need to be further verified by more data from medical institutions.
ABSTRACT
Objective To prepare the trueness verification materials of C-reactive protein (CRP) and evaluate its homogeneity, stability and commutability. Methods The high and low CRP concentrations trueness verification materials were from patient leftover sera which were pooled, mixed thoroughly, filtered and aliquoted. The homogeneity, stability and commutability of these materials were evaluated according to CNAS(China National Accreditation Service for Conformity Assessment, CNAS)-GL29:2010 "Reference materials-General and statistical principles for certification (ISO Guide35:2006)"and the Clinical and Laboratory Standards Institute (CLSI) EP30A. The trueness verification materials were used to evaluate the commutability in 10 clinical CRP detection systems, using forty-five patients' leftover sera with different CRP concentration evaluated by Deming regression in EP30A of CLSI. Meanwhile, the commutability of dilution series of ERM DA-474/IFCC were evaluated using the same method. Results A total of two CRP concentration level trueness verification materials were prepared, with high and low concentration levels of 754 and 743 vials, 1 ml each, respectively. The preparation showed good homogeneity (F<F0.05(14,30);On the condition of room temperature, 2-8 ℃ and -80 ℃, these materials were stable for 7 days and 44 months respectively, the slope of the linear equation of | b1 | less than t0.95,n-2 · s(b1), there was no statistically significant difference between the slope and zero, the stability is satisfied. The materials and the dilution series of ERM-DA 474/IFCC also showed good commutability among patient sera in 10 systems. Conclusions The trueness verification materials of C-reactive protein (CRP) showed good homogeneity, stability and commutability. The dilution series ERM DA-474/IFCC also have good commutability. These provided experimental support for the value transfer and application of the trueness verification materials .
ABSTRACT
Objective To explore the CRP harmonization by calibration using commutable trueness verification materials. Methods High and low level of CRP concentrations trueness verification materials(H and L) were prepared by Beijing center for clinical laboratories. Thesetrueness verification materials were diluted to 5 calibration points(5L, 4L+1H, 3L+2H, 1L+4H, 5H) by weighing method, respectively. These 5 points were used to calibrate four different brands of CRP detection system (Diasys, Leadman, Siemens and Roche) instead of the original procedure. Sera from 21 patients and the international standard ERM DA-474/IFCC were used to compare harmonization and trueness after calibration. Each sample above was measured twice. Results After calibration, the median of CV was reduced from 19.33％ to 2.92％ among 21 patient samples, less than the optimal CV based on biological variability (CV=10.6％). Compared with Desai, the slopes were closer to 1 from 0.90-1.09 to 0.93-0.96 after calibration. Meanwhile, if ERM-DA474/IFCC was used as the trueness verification materials, the absolute bias wasreduced from 3.08-11.07 mg/L to 0.52-2.97 mg/L which was close to theuncertainty of itself (2.5 mg/L). Conclusions Afterthe calibration which contained five linear concentration points of CRP trueness verification materials by weighing method, both harmonization and trueness of CRP were improved.
ABSTRACT
Objective To value C-reactive protein ( CRP ) trueness verification materials and to perform the CRP trueness verification program in Beijing .Methods The CRP value of trueness verification materials were assigned by the international reference material ERM DA-474/IFCC, using 10 clinical routine detection systems at departments of clinical laboratory of Beijing Chaoyang and Luhe Hospital Affiliated to Capital Medical University .The calibration curves with 4 ERM DA-474/IFCC dilutions were established and used for value transfer for trueness verification materials of two levels .The uncertainty was also assessed during the process.Then, the trueness verification was performed in the EQA at Beijing Center for Clinical Laboratories ( BCCL ) among 42 clinical laboratories.The samples were distributed according to BCCL standard operating procedure .The Microsoft Excel 2007 and SPSS 17.0 were used to process the results and the function of efficiency ( En) was calculated to verify the difference between the value and the overall mean of all participating laboratories .Results The values and uncertainties of two trueness verification materials of CRP were (109.9 ±9.4) mg/L and (27.1 ±2.4) mg/L respectively.The results of trial application of two level trueness verification materials in the EQA at Beijing Center for Clinical Laboratories (BCCL) were satisfied.There were no significant difference between the transfer values from our study and the values from means of all laboratories in Beijing .The function of efficiency ( En ) was less than 1.Conclusions The valueswhich were established by using multiple detection platforms for CRP trueness verification materialswere accurate and the uncertainties were small .This method is a preferably method for CRP value assignment because there was no suitable reference method for CRP measurement till now .Thematerialswere suitable for the trueness verification program for clinical laboratories in Beijing .
ABSTRACT
Objective To establish the target measurement uncertainty(MU)of the routine coagulation assay according to the External Quality Assessment data(EQA)of routine coagulation assay. Methods Beijing Center for Clinical Laboratory(BCCL)established the target measurementuncertainty for routine coagulation assayswith the"up-down"methodon the basis of 93 clinical laboratoriesEQA datain BeijingThese assays includedActivated partial thromboplastine time(APTT), Fibrinogen(FBG), International Normalized Ratio(INR), Prothrombin time(PT), Thrombin time(TT)and D-dimer, Compared with CLIA′88,the proficiency of current coagulation assayswas observed.Results The MU of six routine coagulation assayscompared with CLIA ′88 showed that: The 90th percentile MU met the creteriain APTTof group B,FBG of group A&B&C,INR of group B and D-dimer of group B.The 75th percentile MU met the creteriainINR of group A&C,PT of group C.The medium met the creteriainAPTT of group A&C,PT of group A and INR of group D.Conclusions Target Measurement Uncertainty was establishedin routine coagulation assay by using EQA data only,whichcan simplify the procedure of determining MU and continuously update MU according to the frequency of EQA.It has good clinical practical value.However, the applicability of this method should also be considered.
ABSTRACT
Objective Through statistical analysis of specimen rejection reasons, aimed at finding the ways to reduce the failed specimen, making sure of continuous improvements in laboratory quality. Methods Department of Laboratory, Beijing ChaoYang Hospital, Capital Medical University had analyzed rejections specimen from October 2013 to September 2014 in CAP Q-TRACKS QT3 ( QT3: Laboratory Specimen Acceptability ) , and compared the laboratory rejection reasons with all laboratories of CAP. Results From October 2013 to September 2014, the total number of rejection specimen number was 2 367, in which 225 were incomplete labeled specimen/inadequate filled-out form, accounting for 9.5%.898 samples were rejected because of specimen clotted, accounting for 37.9%.The samples could not meet the requirement of specimen quantity of 254, accounting for 10.7%.Other reasons accounting for above 1%included wrong collection container, specimen hemolysis, lipemia or icteric specimen.After comparing the first five rejection reasons, Department of Laboratory, Beijing ChaoYang Hospital was found to be different with others laboratories of CAP.Conclusion The quality of the specimen is very important for the accuracy and reliability of the test results, and doing some positive statistical analysis and taking corrective measures can effectively reduce the unqualified specimen of the proportion.
ABSTRACT
Objective To investigate the internal quality control ( IQC ) on clinical chemistry , clinical immunology and clinical hematology in mutual recognition laboratories in medical institutions in Beijing.Methods By means of questionnaire survey and on -site investigation, fresh frozen serum and whole blood samples with assigned values by reference method were measured to investigate the status of IQC on clinical chemistry , clinical immunology and clinical hematology in 142 mutual recognition laboratories in medical institutions of Beijing,and results were analyzed.Results 142 copies of questionnaireson clinical chemistry, clinical immunology and clinical hematology were send out and 120, 97, and 101 laboratories returned the questionnaires respectively .The information feedback rate was 84.5%, 68.3% and 71.1%respectively .All the questionnaires were effective .Questionnaires survey results showed that more than 50%laboratories set up quality control goals and the most of the goals were probability for error detection ( Ped) 95%, probability for false rejection(Pfr)5%;About 70% laboratories usecd the same quality control plan for different tests ;The most frequently used quality control rules are 12s/13s/22s.On-site investigation showed that ,take the results of clinical chemistry for example , based on the desirable biological variation and WS/T 403 -2012 , most of the tests can't meet the quality control goalsunder the existing quality controlcondition.Conclusion Clinical laboratories should consider their actual situations , assess their own qualitylevels that they can reach , set reasonable quality standards for themselves , and make appropriateindividualized quality control plan.
ABSTRACT
Objective To investigate the distribution of critical values of adults in Beijing, to provide the evidence for the formulation of the Standardized Management Guideline in Critical Values, in order to promote the accurate management of critical values.Methods A total of 110 398 data of critical values from the tertiary and above medical institutions during January 1 to May 31 in 2015 in Beijing were collected by the way of on-site inspection, covering the disciplines of hematology, clinical chemistry, coagulation and blood gas analysis.Fristly, the selected critical values were classified by the factor of admission departments and disease types,then were analyzed by using Kruskal-Wallis test, to compare the differences in each group.Secondly,the combined groups were classified by the factor of gender then were analyzed by using Mann-Whithey U test, to compare the differences in each group.Finally, the stratification thresholds of critical values were established.Results Except for the upper limits of Ca, pH, pCO2, Hb and the lower limits of Glu, pH, the rest of thresholds of critical values had significant differences due to different admission departments and disease types and/or gender.Conclusion Depending on the different admission departmentsces disease types and/or gender, hierarchical limit values on each critical value were formulated.
ABSTRACT
In recent years , with the development of error theory and the requirements of the international accreditation bodies , the quality indicators ( QIs) have become one of the important tools in the laboratory quality management .To explore the evolution and origin of QIs , the status and the future of QIs will be helpful for us to understand and to use the QIs .
ABSTRACT
ObjectiveTo evaluate clinical significance of serum free light chains (sFLC) in diagnosis and response to the therapies of patients with multiple myeloma (MM).MethodssFLC (κ,λ and κ / λ ratio)were examed by immumoassay from 62 patients with MM at different stage. The results were analyzed associated with clinical data,and 35 cases of chronic renal failure(CRF)patients and 62 cases of healthy donors were taken as controls.ResultsMedium sFLC of normal κ value was (13.25±6.46) mg/L,λ value was (18.39±11.42) mg/L; and κ / λ ratio was (0.97±0.64) mg/L (range 0.33-1.61).sFLC κ and λ of CRF patients were (200.01±299.87) mg/L,(191.02±245.98) mg/L,significantly higher than that of the normal control group (t =-17.804,-16.894,both P ＜ 0.001),but the κ/λ ratio was at normal range (1.11±0.29).κ value range was at 16.20- 35 250 mg/L in newly diagnosed intact immunoglobulin MM patients with IgGκ,IgAκ and IgDκ type.The range of λ values was 15.70-4885 mg/L in IgGλ,IgAλ,IgDλ type,and κ/λ ratio was abnormal in 96.5 ％ (55/57) patients (＜0.5 or ＞1.5).The κ,λ value and κ/λ ratio were close to that of the normal after remmision.ConclusionsFLC ( κ,λ,and κ / λ ratio) are very good monitoring markers for MM.
ABSTRACT
Objective Ryanodine receptor (RyR) is one of the Ca2+ release channels on the intracellular Ca2+ stores. RyR induced-Ca2+ release is activated by the voltage-dependent Ca2+ entry, that is, calcium-induced calcium release (CICR). Intracellular free Ca2+ concentration (ECa2+]i) plays a key role on cochlear function. Our study is to investigate the differential expression of RyR in the developing rat cochlea, and to analyze the relationship between the expression of RyR and auditory functional development. Methods Immature SD rats, which were 1 day (P1), 5 days (P5), 10 days (P10), 14 days (P14), 28 days (P28) after parturition, and adutl rats(5 rats for each age) were included in the study. Frozen cochlea sectioning and immunofluorescence were applied to observe the differential expression of RyR. Results The RyR expression in the Corti's organ increased during the cochlear development. It's not significant that the stain was observed on the hair cells and supporting cells in the Corti's organ of P1 and P5 rats. The appearance of the Corti's organ of P10 rats trended to maturity. In P14 rats the RyR expression on hair cells located in the synaptic area against the afferent or efferent nerve, and the strain on supporting cells was extensive. There was little different between the strain on cochlea of P14 and P28 rats. In postmature rat spiral ganglion neurons (SGN), the RyR expression verged gradually from extensive whole cell soma to the synaptic area near to the plasma membrane. Conclusion The RyR expression peaked the 14th day after parturition, which was close to that in the mature cochlea. The time course of the RyR expression during the cochlear development was coincident with that of the auditory functional development. The RyR expression on both hair cells and SGNs located in the synaptic area near to the plasmolemma, implying that RyR induced-CICR was related to the auditory functions such as neurotransmission. Extensive RyR expression in the soma of SGNs at the early stage possibly involved in apoptosis of SGNs during neuron development.