协议(科学)
乙型肝炎病毒
假阳性率
计算机科学
医学
外部质量评估
统计
算法
人工智能
数学
病理
免疫学
病毒
替代医学
作者
Tingting Li,Jiamin Li,Shunwang Cao,Yi Wang,Hongmei Wang,Cheng Zhang,Peifeng Ke,Xianzhang Huang
标识
DOI:10.1515/cclm-2022-0338
摘要
Abstract Objectives Patient-based real-time quality control (PBRTQC) has gained increasing attention in the field of laboratory quality management in recent years. However, PBRTQC has not been reported for use in molecular diagnostics. This study introduces PBRTQC to quantitative hepatitis B virus (HBV) DNA test using moving rate (MR) of positive and negative patient results. Methods In contrast to the MR protocols described in other literature, MR protocol for HBV-DNA test has an additional logarithmic transformation and binary conversion steps before using a common statistical process control algorithm, such as the MR. We used all patient test results of HBV-DNA assay from August 2018 to August 2021 at the Second Affiliated Hospital of Guangzhou University of Chinese Medicine, for parameters setting, optimization, and performance validation. The false rejection rate, error detection curves and validation charts were used to assess the MR protocols. Results The false rejection rates of two MR protocols were both <0.7%. The optimal block sizes for positive and negative errors in each cut-off value were not the same, so we first proposed a combined protocol that used different block size to detect negative and positive errors. It turned out that the combined protocols outperformed the simple protocols for each cut-off value, especially detecting positive errors. Conclusions The performances of MR protocols using positive or negative patient results to detect constant errors of HBV-DNA test could meet laboratory requirements. Therefore, we have provided an effective alternative tool for internal quality control in the field of molecular diagnostics.
科研通智能强力驱动
Strongly Powered by AbleSci AI