
ivdtools is an R package providing statistical workflows used in the evaluation of in vitro diagnostic (IVD) reagents, covering data inspection, statistical analysis, result summarization, and visualization. The current version requires R >= 4.1.0 and is distributed under the MIT license.
ivdtools 是一个用于体外诊断(IVD)试剂评价的 R 软件包,提供数据检查、统计分析、结果汇总和可视化等统计工作流程。当前版本要求 R >= 4.1.0,并采用 MIT 许可证发布。
The package provides the following analysis modules:
该软件包提供以下分析模块:
| Module · 模块 | Main entry points · 主要入口函数 | Related CLSI guideline · 相关 CLSI 指南 |
|---|---|---|
| Method comparison · 方法学比较 | mcr() |
EP09 |
| Reference-material commutability · 参考物质互换性 | commutability() |
EP14, EP30, and IFCC Part 2 |
| Precision · 精密度 | precision() |
EP05 and EP15 |
| Qualitative agreement · 定性一致性 |
raw_to_table() and others |
EP12 |
| C5/C95 estimation · C5/C95 估计 | c5_c95() |
EP12 |
| Reference interval · 参考区间 | reference_interval() |
EP28 |
| Stability · 稳定性 |
stability_*(), arrhenius(), mkt()
|
EP25 |
| Linearity · 线性 |
linearity() and linearity_*()
|
EP06 |
| Curve fitting · 曲线拟合 |
fit_equation() family |
General · 通用 |
| Interference · 干扰 | interference_*() |
EP07 |
| Dilution, spiking, and hook effect · 稀释、加标和钩状效应 |
dilution_recovery(), spike_recovery(), hook_effect()
|
EP34 |
| Measurement uncertainty · 测量不确定度 | uncertainty_*() |
EP29 |
| Reference-material bias · 参考物质偏倚 | reference_bias() |
YY/T 1789.2 |
| LoB/LoD/LoQ · 空白限、检出限、定量限 |
sensitivity_lob() / sensitivity_lod() / sensitivity_loq()
|
EP17 |
| Bottle-to-bottle ANOVA · 瓶间差方差分析 | bottle_anova() |
EP15 |
| ROC analysis · ROC 分析 |
roc() family |
EP24 |
| Quality control · 质量控制 |
qc_chart(), youden_plot()
|
Westgard rules · Westgard 规则 |
| Outliers and normality tests · 异常值与正态性检验 |
outliers_test(), normal_test()
|
General · 通用 |
| Sample size · 样本量 | sample_size_*() |
General · 通用 |
Statistical results should be interpreted in the context of a pre-specified study protocol, applicable standards, and clinical or analytical acceptance limits; they are not a substitute for professional judgment.
统计结果应结合预先规定的研究方案、适用标准以及临床或分析性能接受限进行解释;统计结果不能替代专业判断。
Detailed tutorials covering data import, analysis environment setup, and complete worked examples for every analysis module are available:
这里提供涵盖数据导入、分析环境配置以及各分析模块完整操作示例的详细教程:
English handbook · 英文手册:https://hiox-tech.github.io/ivdtools/
Chinese handbook · 中文手册:https://ivdtools.hiox-tech.cn/
Manual · 软件包手册:https://cran.r-project.org/web/packages/ivdtools/refman/ivdtools.html
For PDF manuals, cheatsheets and more resources, visit:
如需获取 PDF 手册、速查表及其他资源,请访问:
GitHub · 英文:https://github.com/hiox-tech/ivdtools
Gitee · 中文:https://gitee.com/hiox-tech/ivdtools
CRAN:https://CRAN.R-project.org/package=ivdtools
Bug reports, feature requests, and feedback:
如需报告错误、提出功能请求或反馈意见,请访问:
https://github.com/hiox-tech/ivdtools/issues
hiox-tech GeorgeBinDragon@outlook.com
Built for IVD reagent evaluation — designed for reproducibility and clarity.
为 IVD 试剂评估构建——追求清晰与可复现。