医学影像组学特征选择与分类模型的工程化实现 ——基于 feature engineering特征工程的 AI平台设计实践

李 昕
华超神控（上海）科技有限公司

摘要


本文以feature engineering软件为实践载体，探讨医学影像组学特征选择与分类模型的工程化实现路径。研
究围绕高维小样本医学影像数据，从需求约束、系统模块与工程落地三方面展开，构建统一数据底座、可组合特征
筛选链路、标准化模型训练接口，并配套日志追踪与一键式自动化流程。结果表明，影像组学平台的核心价值在于
流程稳定、数据可追溯、结果可复核，而非单纯堆叠算法。该研究为医学AI科研平台研发、工程化落地及产品化迭
代提供可复用的设计思路与技术参考。

关键词


影像组学；特征选择；分类模型；软件工程；feature engineering

全文:

PDF


参考


[1]WOZNICKI P, SARRAZIN O, MUIR J, et al.

AutoRadiomics: A framework for reproducible radiomics

research[J]. Frontiers in Radiology, 2022, 2:919133.

[2]DEWI D E O, SUNOQROT M R S, NKETIAH

G A, et al. The impact of pre-processing and disease

characteristics on reproducibility of T2-weighted MRI

radiomics features[J]. Magnetic Resonance Materials in

Physics, Biology and Medicine, 2023, 36:945-956.

[3]SHOEMAKER K, GER R, COURT L E, et al.

Bayesian feature selection for radiomics using reliability

metrics[J]. Frontiers in Genetics, 2023, 14:1112914.

[4]HANIFF N S M, NG K H, KAMAL I, et al.

Systematic review and meta-analysis on the classification

metrics of machine learning algorithm based radiomics

in hepatocellular carcinoma diagnosis[J]. Heliyon, 2024,

10(16):e36313.

[5]MORADMAND H, MOLITORIS J, LING X, et al.

Graph feature selection for enhancing radiomic stability and

reproducibility across multiple institutions in head and neck

cancer[J]. Scientific Reports, 2025, 15:27995.


Refbacks

  • 当前没有refback。