LightGBM混合模型在乳腺癌诊断中的应用

doi:10.3778/j.issn.1002-8331.2211-0218

摘要/Abstract

摘要： 乳腺癌是最常见的癌症种类之一，且患病率每年仍在上升。在不进行手术活检的情况下，通过分析细胞核的各项指标来预测肿块的良性与否，可以有效地为医生提供辅助诊疗并减少患者的痛苦。为此，提出了一种基于LightGBM算法的乳腺癌诊断模型。使用边界-合成少数类过采样算法（borderline-synthetic minority oversampling technique，Borderline-SMOTE）来改善乳腺癌确诊数据不平衡的问题。在麻雀搜索算法（sparrow search algorithm，SSA）中引入PWLCM混沌映射、全新的惯性权重和纵横交叉算法对其进行改进，再运用改进后的SSA算法对LightGBM的参数进行自动寻优。由于LightGBM对噪点较为敏感，所以提出了一种OVR-Jacobian正则化方法对LightGBM进行降噪处理。使用改进后的LightGBM混合模型对乳腺癌进行诊断。实验结果表明，提出的混合模型在均方误差、决定系数和交叉验证得分这三个指标上均优于常见的模型，显示出其较好的诊断效果。

关键词: 乳腺癌预测, LightGBM, 麻雀搜索算法, Borderline-SMOTE算法, 机器学习, Jacobian正则化

Abstract: Breast cancer is one of the most common types of cancer, and its prevalence continues to rise every year. Without surgical biopsy, it can effectively provide auxiliary diagnosis and treatment for doctors and reduce the pain of patients by analyzing various indicators of the nucleus to predict whether the mass is benign or not. Therefore, a breast cancer diagnosis model based on LightGBM algorithm is proposed. Firstly, the borderline-synthetic minority oversampling technique (Borderline-SMOTE) is used to improve the problem of imbalanced breast cancer diagnosis data. Secondly, the PWLCM chaotic map, the new inertia weight and the criss-cross algorithm are introduced into the sparrow search algorithm (SSA) to improve it, and then the improved SSA algorithm is used to automatically optimize the parameters of LightGBM. Then, because LightGBM is sensitive to noise, an OVR-Jacobian regularization method is proposed to reduce the noise of LightGBM. Finally, the improved LightGBM hybrid model is used to diagnose breast cancer. The experimental results show that the proposed hybrid model is superior to the common models in the three indicators of mean square error, coefficient of determination and cross-validation score, showing its better diagnostic effect.

Key words: breast cancer prediction, LightGBM, sparrow search algorithm, Borderline-SMOTE algorithm, machine learning, Jacobian regularization

邢长征, 徐佳玉. LightGBM混合模型在乳腺癌诊断中的应用[J]. 计算机工程与应用, 2024, 60(6): 330-338.

XING Changzheng, XU Jiayu. Hybrid LightGBM Model for Breast Cancer Diagnosis[J]. Computer Engineering and Applications, 2024, 60(6): 330-338.

参考文献

[1] 陈原华, 周冬. 早期乳腺癌的危险因素及防控措施分析[J]. 中国肿瘤临床与康复, 2022, 29(9): 1070-1072.
CHEN Y H, ZHOU D. Risk factors and prevention and control measures for early stage breast cancer[J]. Chinese Journal of Clinical Oncology and Rehabilitation, 2022, 29(9): 1070-1072.
[2] BRAY F, FERLAY J, SOERJOMATARAM I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA: A Cancer Journal for Clinicians, 2018, 68(6): 394-424.
[3] CARLOS A P, MOSHE S. A fuzzy-genetic approach to breast cancer diagnosis[J]. Artificial Intelligence in Medicine, 1999, 17(2): 131-155.
[4] TRYGGVADOTTIR L, TULINIUS H, EYFJORD J E, et al. Breastfeeding and reduced risk of breast cancer in an Icelandic cohort study[J]. American Journal of Epidemiology, 2001, 154(1): 37-42.
[5] JERNSTROM H, LUNINSKI J, LYNCH H T, et al. Breast-feeding and the risk of breast cancer in BRCA1 and BRCA2 mutation carriers[J]. Journal of the National Cancer Institute, 2004, 96(14): 1094-1098.
[6] GAIL M H, BRTINTON L A, BYAR D P, et al. Projecting individualized probabilities of developing breast cancer for white females who are being examined annually[J]. Journal of the National Cancer Institute, 1989, 81(24): 1879-1886.
[7] KEMAL A. Diagnosis of breast cancer with stacked autoencoder and subspace kNN[J]. Physica A: Statistical Mechanics and its Applications, 2020, 551(4): 124591.
[8] 苗立志, 刁继尧, 娄冲, 等. 基于Spark和随机森林的乳腺癌风险预测分析[J]. 计算机技术与发展, 2019, 29(8): 142-146.
MIAO L Z, DIAO J Y, LOU C, et al. Breast cancer risk prediction analysis based on apache spark and random forest algorithm[J]. Computer Technology and Development, 2019, 29(8): 142-146.
[9] ALFIAN G, SYAFRUDIN M, FAHRURROZ I, et al. Predicting breast cancer from risk factors using SVM and extra-trees-based feature selection method[J]. Computers, 2022, 11(9): 136.
[10] 李勇, 陈思萱, 贾海, 等. 基于C-AdaBoost模型的乳腺癌预测研究[J]. 计算机工程与科学, 2020, 42(8): 1414-1422.
LI Y, CHEN S X, JIA H, et al. Prediction of breast cancer based on C-AdaBoost model[J]. Computer Engineering & Science, 2020, 42(8): 1414-1422.
[11] 袁梦绚. 基于深度学习的乳腺癌预测[J]. 计算机与数字工程, 2021, 49(11): 2353-2357.
YUAN M X. Breast cancer prediction based on deep learning[J]. Computer & Digital Engineering, 2021, 49(11): 2353-2357.
[12] ABDIKENOV B, IKLASSOV Z, SHARIPOV A, et al. Analytics of heterogeneous breast cancer data using neuroevolution[J]. IEEE Access, 2019, 7: 18050-18060.
[13] 沈倩倩, 邵峰晶, 孙仁诚. 基于XGBoost的乳腺癌预测模型[J]. 青岛大学学报 (自然科学版), 2019, 32(1): 95-100.
SHEN Q Q, SHAO F J, SUN R C. Prediction model of breast cancer based on XGBoost[J]. Journal of Qingdao University (Natural Science Edition), 2019, 32(1): 95-100.
[14] 甘富文, 武明辉, 吴亚平, 等. 多特征融合的肝细胞癌分化等级术前预测方法研究[J]. 计算机应用与软件, 2022, 39(7): 147-153.
GAN F W, WU M H, WU Y P, et al. A preoperative prediction method for differentiation grades of hepatocellular carcinoma based on multi-feature fusion[J]. Computer Applications and Software, 2022, 39(7): 147-153.
[15] 王鑫, 廖彬, 李敏, 等. 融合LightGBM与SHAP的糖尿病预测及其特征分析方法[J]. 小型微型计算机系统, 2022, 43(9): 1877-1885.
WANG X, LIAO B, LI M, et al. Combination of LightGBM and SHAP for diabetes prediction and feature analysis[J]. Journal of Chinese Computer Systems, 2022, 43(9): 1877-1885.
[16] ZHANG S, YUAN Y, YAO Z, et al. Coronary artery disease detection model based on class balancing methods and LightGBM algorithm[J]. Electronics, 2022, 11(9): 1495.
[17] ZHANG C, LEI X, LIU L. Predicting metabolite-disease associations based on LightGBM model[J]. Frontiers in Genetics, 2021, 12: 660275.
[18] LIAO H, ZHANG X, ZHAO C, et al. LightGBM: an efficient and accurate method for predicting pregnancy diseases[J]. Journal of Obstetrics and Gynaecology, 2021, 42(4): 620-629.
[19] HAN H, WANG W Y, MAO B H. Borderline-SMOTE: a new over-sampling method in imbalanced data sets learning[C]//Proceedings of the International Conference on Advances in Intelligent Computing, 2005: 878-887.
[20] XUE J K, SHEN B. A novel swarm intelligence optimization approach: sparrow search algorithm[J]. Systems Science & Control Engineering, 2020, 8(1): 22-34.
[21] 柴岩, 孙笑笑, 任生. 融合多向学习的混沌麻雀搜索算法[J]. 计算机工程与应用, 2023, 59(6): 81-91.
CHAI Y, SUN X X, REN S. Chaotic sparrow search algorithm based on multi-directional learning[J]. Computer Engineering and Applications, 2023, 59(6): 81-91.
[22] ZHANG C L, DING S F. A stochastic configuration network based on chaotic sparrow search algorithm[J]. Knowledge-Based Systems, 2021, 220: 106924.
[23] 高晨峰, 陈家清, 石默涵. 融合黄金正弦和曲线自适应的多策略麻雀搜索算法[J]. 计算机应用研究, 2022, 39(2): 491-499.
GAO C F, CHEN J Q, SHI M H. Multi-strategy sparrow search algorithm integrating golden sine and curve adaptive[J]. Application Research of Computers, 2022, 39(2): 491-499.
[24] 楚哲宇, 唐秀英, 谭庆, 等. 基于逐维高斯变异的混沌麻雀优化算法[J]. 自动化应用, 2021(8): 60-63.
CHU Z Y, TANG X Y, TAN Q, et al. Chaos sparrow optimization algorithm based on dimensional Gaussian mutation[J]. Automation Application, 2021(8): 60-63.
[25] 张伟康, 刘升, 任春慧. 混合策略改进的麻雀搜索算法[J]. 计算机工程与应用, 2021, 57(24): 74-82.
ZHANG W K, LIU S, REN C H. Mixed strategy improved sparrow search algorithm[J]. Computer Engineering and Applications, 2021, 57(24): 74-82.
[26] BARANOVSKY A, DAEMS D. Design of one-dimensional chaotic maps with prescribed statistical properties[J]. International Journal of Bifurcation and Chaos, 1995, 5(6): 1585-1598.
[27] 段玉先, 刘昌云. 基于Sobol序列和纵横交叉策略的麻雀搜索算法[J]. 计算机应用, 2022, 42(1): 36-43.
DUAN Y X, LIU C Y. Sparrow search algorithm based on sobol sequence and crisscross strategy[J]. Journal of Computer Applications, 2022, 42(1): 36-43.
[28] MENG A B, CHEN Y C, YIN H, et al. Crisscross optimization algorithm and its application[J]. Knowledge-Based Systems, 2014, 67: 218-229.
[29] 王静莹. 服务互联网价值建模与优化分析方法研究[D]. 哈尔滨: 哈尔滨工业大学, 2021.
WANG J Y. Value modeling and value optimization analysis method for the internet of services[D]. Harbin: Harbin Institute of Technology, 2021.
[30] DNIEL J, RAJA G. Improving DNN robustness to adversarial attacks using Jacobian regularization[J]. arXiv:1803. 08680, 2018.
[31] Breast cancer wisconsin (diagnostic) data set[EB/OL]. http://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+%28Diagnostic%29.