融合CatBoost和SHAP的乳腺癌预测及特征分析

doi:10.3969/j.issn.1006-2475.2023.10.005

摘要/Abstract

摘要： 针对当前乳腺癌预测模型存在性能不足和可解释性差的问题，提出一种融合CatBoost和SHAP的乳腺癌预测及特征分析模型。首先，对原始乳腺癌数据集进行异常值处理和数据归一化处理等工作，以提高数据的质量。然后，基于CatBoost建立乳腺癌预测的模型，并进行泛化能力分析。最后，将预测模型结合SHAP进行可解释分析，以探索影响乳腺癌的关键因素。使用威斯康星大学的Breast Cancer Wisconsin （Diagnostic）数据集验证该模型，结果表明：Accuracy值为99.30%，Precision值为99.50%，Recall值为98.91%，F1值为99.19%，均优于现有文献。其中Accuracy指标提升1.12~6.90个百分点，Precision指标提升2.00~7.50个百分点，Recall指标提升2.41~6.91个百分点，F1值提升2.19~7.19个百分点，以此验证本文模型的优越性。此外，SHAP模型得出影响乳腺癌的核心因素有concave points_worst（乳腺组织细胞核凹点极值）、perimeter_worst（乳腺组织细胞核周长极值）、area_worst（乳腺组织细胞核面积极值）等，这为医生诊断提供原理性支撑。

关键词: 关键词：CatBoost算法, 可解释, 乳腺癌, 疾病预测, 特征分析, 机器学习

Abstract: To address the problems of insufficient performance and poor interpretability of current breast cancer prediction models， this paper proposes a breast cancer prediction and feature analysis model incorporating CatBoost and SHAP. First， the original breast cancer dataset is processed with outliers and data normalization to improve the quality of the data. Then， a model for breast cancer prediction based on CatBoost is built and generalization ability analysis is performed. Finally， the prediction model is combined with SHAP for interpretable analysis to explore the key factors affecting breast cancer. The model is validated using the Breast Cancer Wisconsin （Diagnostic） dataset from the University of Wisconsin， and the results show that the Accuracy value of 99.30%， Precision value of 99.50%， Recall value of 98.91%， and F1 value of 99.19% are better than the existing literature. The superiority of this model is verified by the fact that the Accuracy index improved by 1.12-6.90 percentage points， the Precision index improved by 2.00-7.50 percentage points， the Recall index improved by 2.41-6.91 percentage points， and the F1 value improved by 2.19-7.19 percentage points. In addition， the SHAP model yields the core factors affecting breast cancer， such as concave points_worst， perimeter_worst， and area_worst， which provide the principle support for doctors’ diagnosis.

Key words: Key words: CatBoost algorithm, interpretable, breast cancer, disease prediction, feature analysis, machine learning

中图分类号:

TP391

贾潇瑶, . 融合CatBoost和SHAP的乳腺癌预测及特征分析[J]. 计算机与现代化, 2023, 0(10): 32-38.

JIA Xiao-yao, . Breast Cancer Prediction and Feature Analysis Model Based on CatBoost and SHAP[J]. Computer and Modernization, 2023, 0(10): 32-38.

参考文献［28］

［1］	欧阳平，李小溪，冷芬，等. 机器学习算法在体检人群糖尿病风险预测中的应用［J］. 中华疾病控制杂志， 2021，25（7）:849-853.
［2］	李仪，林建君，朱习军. 基于改进DNN的糖尿病预测模型设计［J］. 计算机工程与设计， 2021，42（5）:1418-1424.
［3］	李军，胡晓娟，屠立平，等. 基于舌象参数与多指标特征联合的2型糖尿病风险预测模型［J］. 中国中医基础医学杂志， 2021，27（3）:451-456.
［4］	孟辉，张加宏，李敏，等. 基于IPSO-BP神经网络与BCG信号的冠心病预测分类研究［J］. 传感技术学报， 2020，33（10）:1379-1385.
［5］	蒋林甫，袁贞明，张邢炜，等. 基于PCHD-TabNet的十年冠心病预测［J］. 数据分析与知识发现， 2023，7（5）:133-144.
［6］	LIAO B， JIA X Y， ZHANG T， et al. DHDIP: An interpretable model for hypertension and hyperlipidemia prediction based on EMR data［J］. Computer Methods and Programs in Biomedicine， 2022，226. DOI: 10.1016/j.cmpb.2022.107088.
［7］	谢爽，范会敏. 基于Word2vec和卷积神经网络特征提取的双高疾病预测［J］. 计算机应用与软件， 2021，38（2）:93-96.
［8］	董章功，宋波，孟友新. 基于SEIR-ARIMA混合模型的新冠肺炎预测［J］. 计算机与现代化， 2022（2）:1-6.
［9］	朱岩，张利，王煜. 基于RoBERTa-WWM的中文电子病历命名实体识别［J］. 计算机与现代化， 2021（2）:51-55.
［10］	徐子晨，范子祥，王文倩，等. 人参皂苷抗乳腺癌分子机制的研究进展［J］. 中草药， 2022，53（20）:6601-6610.
［11］	SUNG H， FERLAY J， SIEGEL R L， et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries［J］. CA: A Cancer Journal for Clinicians， 2021，71（3）:209-249.
［12］	ZHANG X T， ZHANG Y， GAO H R， et al. A wrapper feature selection algorithm based on brain storm optimization［C］// Proceedings of the 2018 International Conference on Bio-inspired Computing: Theories and Applications. 2018:308-315.
［13］	ABDEL-BASSET M， EL-SHAHAT D， EL-HENAWY I， et al. A new fusion of grey wolf optimizer algorithm with a two-phase mutation for feature selection［J］. Expert Systems with Applications， 2020，139. DOI: 10.1016/j.eswa.
	2019.112824.
［14］	WANG H F， ZHENG B C， YOON S W， et al. A support vector machine-based ensemble algorithm for breast cancer diagnosis［J］. European Journal of Operational Research， 2018，267（2）:687-699.
［15］	RAO H D， SHI X Z， RODRIGUE A K， et al. Feature selection based on artificial bee colony and gradient boosting decision tree［J］. Applied Soft Computing， 2019，74:634-642.
［16］	NAJI M A， EL FILALI S， AARIKA K， et al. Machine learning algorithms for breast cancer prediction and diagnosis［J］. Procedia Computer Science， 2021，191:487-492.
［17］	MISHRA A K， ROY P， BANDYOPADHYAY S. Binary particle swarm optimization based feature selection （BPSO-FS） for improving breast cancer prediction［C］// Proceedings of the 2021 International Conference on Artificial Intelligence and Applications. 2021:373-374.
［18］	AGARAP A F M. Deep learning using rectified linear units （ReLU）［J］. arXiv preprint arXiv:1803.08375， 2018.
［19］	LIU N， QI E S， XU M， et al. A novel intelligent classification model for breast cancer diagnosis［J］. Information Processing & Management， 2019，56（3）:609-623.
［20］	SINGH D， SINGH B， KAUR M. Simultaneous feature weighting and parameter determination of neural networks using ant lion optimization for the classification of breast cancer［J］. Biocybernetics and Biomedical Engineering， 2020，40（1）:337-351.
［21］	甘丹. 面向多模态数据的医疗与健康决策支持研究［D］. 天津:天津大学， 2020.
［22］	WOLBERG W， MANGASARIAN O， STREET N， et al. Breast cancer Wisconsin （diagnostic）［EB/OL］. （1995-10-31）［2022-10-12］. https://archive.ics.uci.edu/dataset/17/breast+cancer+wisconsin+diagnostic.
［23］	WOLBERG W. Breast cancer Wisconsin （original）［EB/OL］. （1992-07-14）［2022-10-12］. http://archive.ics.uci.edu/dataset/15/breast+cancer+wisconsin+original.
［24］	PATRICIO M， PEREIRA J， CRISOSTOMO J， et al. Using resistin， glucose， age and BMI to predict the presence of breast cancer［J］. BMC Cancer， 2018，18（1）. DOI: 10.1186
	/s12885-017-3877-1.
［25］	PROKHORENKOVA L， GUSEV G， VOROBEV A， et al. CatBoost: Unbiased boosting with categorical features［J］. arXiv preprint arXiv:1706.09516， 2017.
［26］	DOROGUSH A V， ERSHOV V， GULIN A. CatBoost: Gradient boosting with categorical features support［J］. arXiv preprint arXiv:1810.11363， 2018.
［27］	FRIEDMAN J H. Greedy function approximation: A gradient boosting machine［J］. The Annals of Statistics， 2001，29（5）:1189-1232.
［28］	LUNDBERG S， LEE S I. A unified approach to interpreting model predictions［J］. arXiv preprint arXiv:1705.07874，2017.

[1]	郭凯, 艾菊梅. 基于FLANN改进的KNN医疗分类算法[J]. 计算机与现代化, 2022, 0(08): 25-29.
[2]	王钰蓥, 王勇. 基于知识图谱和Bi-LSTM的推荐算法[J]. 计算机与现代化, 2021, 0(09): 90-98.
[3]	宋紫华1,2,郭春1,2,蒋朝惠1,2. 一种基于网络流量分析的快速木马检测方法[J]. 计算机与现代化, 2019, 0(06): 9-.