单一与比较Fβ评分的统计推断与功效分析统一框架
针对F1与Fβ评分缺乏严谨统计推断与功效分析的问题,提出psF1统一框架,支持区间估计、假设检验及功效与样本量计算;模拟验证其准确稳健,并展示于真实生物医学分类任务,提供可靠不确定性量化与样本量规划。
要点
- psF1可对单一及比较F1/Fβ评分做区间估计、假设检验与功效分析。
- 框架利用精确分布与大样本近似,模拟显示跨敏感度与精确度场景准确稳健。
- psF1已用于真实生物医学分类任务,免费开源供分类器评估与比较。
核心结论
- psF1为单一与比较F1和Fβ评分提供统一的统计推断与功效分析框架,支持可靠的不确定性量化与样本量规划。
研究设计与样本
- 方法学研究:提出psF1框架,利用精确概率分布与大样本近似实现区间估计、假设检验及功效与样本量计算。
- 通过广泛模拟评估psF1在不同敏感度、精确度与样本量场景下的准确性与稳健性。
- 以真实生物医学分类任务展示psF1的实用价值。
主要发现
- 模拟显示psF1在多种敏感度、精确度与样本量场景下具有准确性与稳健性。
- psF1在真实生物医学分类任务中展示了实用效用。
临床怎么用
- 可使用psF1对分类器的F1或Fβ评分进行区间估计与假设检验,并开展功效与样本量规划。
- psF1免费开源(http://github.com/cyhsuTN/psF1),便于在生物医学分类研究中应用。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
Machine learning and artificial intelligence are increasingly applied to medical diagnostics and clinical decision-making. To evaluate model performance, the F 1 $$ {F}_1 $$ score and its generalized form, the F β $$ {F}_{\beta } $$ score, are widely used as they balance precision and sensitivity. However, rigorous statistical inference and power analysis for the F 1 $$ {F}_1 $$ and F β $$ {F}_{\beta } $$ scores remain limited. In this study, we propose psF1, a unified and comprehensive framework for interval estimation, hypothesis testing, and power and sample size calculation for both single and comparative F 1 $$ {F}_1 $$ and F β $$ {F}_{\beta } $$ scores. psF1 leverages exact probability distributions as well as approximations for large sample sizes to provide valid statistical inference and power analyses. Extensive simulations demonstrate the accuracy and robustness of psF1 across a range of sensitivity, precision, and sample size scenarios. We further showcase its practical utility through real-world biomedical classification tasks. This framework enables principled evaluation and comparison of classifiers using F 1 $$ {F}_1 $$ and F β $$ {F}_{\beta } $$ scores with reliable uncertainty quantification and informed sample size planning. psF1 is freely available at http://github.com/cyhsuTN/psF1.
引用
- Hsu CY, Liu Q, Shyr Y, et al. A Unified Framework for Statistical Inference and Power Analysis of Single and Comparative F β $$ {F}_{\beta } $$ Scores. Stat Med. 2026. PMID: 42050845. doi:10.1002/sim.70557
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。