基于两种机器学习算法的双相情感障碍患者自杀行为影响因素模型比较研究

doi:10.6040/j.issn.1671-7554.0.2021.0359

摘要/Abstract

摘要： 目的探索两种算法构建住院双相情感障碍患者自杀行为影响因素模型的特点,比较其分类能力,为住院双相情感障碍患者自杀行为的预防控制提供依据。方法利用2010年1月至2017年12月某精神专科医院住院双相情感障碍患者的数据,通过χ2 检验初步筛选自杀行为影响因素,采用Adaboost、二分类Logistic回归两种算法构建自杀行为影响因素模型,再用查全率、查准率和F1值比较不同模型特点。结果研究共纳入住院双相情感障碍患者7 782例,有自杀行为的患者1 661例,自杀行为率为21%。与Logistic回归模型相比,Adaboost模型分类能力较强且稳定。自杀行为影响因素中,诊断分型和既往自杀史在两模型中均占据重要地位。结论两种算法构建的双相情感障碍患者自杀行为影响因素模型,总体分类能力差别较小,需进一步挖掘潜在变量以提升模型分类能力。诊断分型为当前抑郁发作或混合发作、有既往自杀史的双相情感障碍患者是自杀行为的高危人群,应针对该特征加强自杀行为的预防工作。

关键词: 机器学习, 自杀行为, 影响模型, 双相情感障碍, Adaboost, Logistic回归

Abstract: Objective To explore the characteristics of two machine learning algorithms in the construction of suicidal behavior predictive models for bipolar disorder inpatients, and to compare their performance so as to provide a basis for the prevention and control of suicidal behavior of inpatients with bipolar disorder. Methods Clinical data of bipolar disorder inpatients treated during Jan. 2010 and Dec. 2017 were retrospectively analyzed. After the influencing factors of suicidal behavior were screened with Chi-square test analysis, Adaboost and binary Logistic regression were used to construct two suicidal behavior classification models. The characteristics of the two models were compared with recall ratio, precision ratio and F1 value. Results A total of 7,782 bipolar disorder inpatients were enrolled, among whom 1,661 had suicidal ideation or attempted suicide and the rate of suicidal behavior was 21%. Between the two models, Adaboost performed better. Diagnosis subtype and past suicide history were the two most important risk factors in both models. Conclusion There are only slight differences between the two models based on different machine learning algorithms, both having a low average performance. More factors are needed to improve the model performance. Current depression episode or mixed state and past suicide history are the most predictive traits for suicidal behavior in bipolar disorder inpatients. Preventive measures should be taken to address suicidal behavior risks accordingly.

Key words: Machine learning, Suicidal behavior, Influening model, Bipolar disorder, Adaboost, Logistic regression

中图分类号:

TP181

姜震,孙静,邹雯,王唱唱,高琦. 基于两种机器学习算法的双相情感障碍患者自杀行为影响因素模型比较研究[J]. 山东大学学报 (医学版), 2022, 60(1): 101-108.

JIANG Zhen, SUN Jing, ZOU Wen, WANG Changchang, GAO Qi. A comparison study of suicidal behavior predictive models of bipolar disorder patients based on two machine learning algorithms[J]. Journal of Shandong University (Health Sciences), 2022, 60(1): 101-108.

参考文献

[1] 余桂红, 毛越. 加拿大青少年自杀预防教育专业化发展及启示——基于对青少年自杀预防中心(Kids Help Phone)的考察 [J]. 徐州工程学院学报(社会科学版), 2019, 34(2): 100-108. YU Guihong, MAO Yue. Professional development and enlightenment of canadian kids suicide prevention education: a survey on kids help phone [J]. Journal of Xuzhou Institute of Technology(Social Sciences Edition), 2019, 34(2): 100-108.
[2] 邹亚明, 郝元涛. 我国伤害所致死亡损失生命年和经济负担分析 [J]. 中华疾病控制杂志, 2016, 20(5): 495-499. ZOU Yaming, HAO Yuantao. Analysis on years of life lost and economic burden caused by injury in China [J]. Chinese Journal of Disease Control & Prevention, 2016, 20(5): 495-499.
[3] Tietbohl-Santos B, Chiamenti P, Librenza-Garcia D, et al. Risk factors for suicidality in patients with panic disorder: a systematic review and meta-analysis [J]. Neurosci Biobehav Rev, 2019, 105: 34-38. doi: 10.1016/j.neubiorev.2019.07.022.
[4] Dong M, Lu L, Zhang L, et al. Prevalence of suicide attempts in bipolar disorder: a systematic review and meta-analysis of observational studies [J]. Epidemiol Psychiatr Sci, 2019, 29: e63. doi: 10.1017/S2045796019000593.
[5] Pompili M, Gonda X, Serafini G, et al. Epidemiology of suicide in bipolar disorders: a systematic review of the literature [J]. Bipolar Disord, 2013, 15(5): 457-490.
[6] Schaffer A, Isometsä ET, Tondo L, et al. International society for bipolar disorders task force on suicide: meta-analyses and meta-regression of correlates of suicide attempts and suicide deaths in bipolar disorder [J]. Bipolar Disord, 2015, 17(1): 1-16.
[7] Franklin JC, Ribeiro JD, Fox KR, et al. Risk factors for suicidal thoughts and behaviors: a meta-analysis of 50 years of research [J]. Psychol Bull, 2017, 143(2): 187-232.
[8] 许侨洋. 基于特征提取的ICU患者死亡风险预测研究[D]. 南京: 南京大学, 2019.
[9] 周阳. 基于机器学习的医疗文本分析挖掘技术研究[D]. 北京: 北京交通大学, 2019.
[10] 高萌, 杨仙鸿, 姜祎群. 人工智能在医学领域的研究进展[J]. 中华皮肤科杂志, 2019, 52(2): 131-134. GAO Meng, YANG Xianhong, JIANG Yiqun. Artificial intelligence in medicine [J]. Chinese Journal of Dermatology, 2019, 52(2): 131-134.
[11] 郑泓, 蒲城城, 王毅, 等. 基于脑结构像的精神分裂症机器学习分类[J]. 心理科学进展, 2020, 28(2): 252-265. ZHENG Hong, PU Chengcheng, WANG Yi, et al. The classification of schizophrenia based on brain structural features: a machine learning approach [J]. Advances in Psychological Science, 2020, 28(2): 252-265.
[12] Kessler RC, Hwang I, Hoffmire CA, et al. Developing a practical suicide risk prediction model for targeting high-risk patients in the Veterans health Administration [J]. Int J Methods Psychiatr Res, 2017, 26(3): e1575. doi: 10.1002/mpr.1575.
[13] Haroz EE, Walsh CG, Goklish N, et al. Reaching those at highest risk for suicide: development of a model using machine learning methods for use with native American communities [J]. Suicide Life Threat Behav, 2020, 50(2): 422-436.
[14] 赵北庚. 基于R语言adabag包的集成学习建模研究[J]. 信息与电脑(理论版), 2015(5): 106-107. ZHAO Beigeng. Research on ensemble learning modeling based on R language adabag package [J]. Information and Computers(Theory Edition), 2015(5): 106-107.
[15] Gonda X, Pompili M, Serafini G, et al. Suicidal behavior in bipolar disorder: epidemiology, characteristics and major risk factors [J]. J Affect Disord, 2012, 143(1-3): 16-26.
[16] Tidemalm D, Haglund A, Karanti A, et al. Attempted suicide in bipolar disorder: risk factors in a cohort of 6086 patients [J]. PLoS One, 2014, 9(4): e94097. doi: 10.1371/journal.pone.0094097.
[17] Hansson C, Joas E, Pälsson E, et al. Risk factors for suicide in bipolar disorder: a cohort study of 12 850 patients [J]. Acta Psychiatr Scand, 2018, 138(5): 456-463.
[18] Malhi GS, Outhred T, Das P, et al. Modeling suicide in bipolar disorders [J]. Bipolar Disord, 2018, 20(4): 334-348.
[19] Plans L, Barrot C, Nieto E, et al. Association between completed suicide and bipolar disorder: a systematic review of the literature [J]. J Affect Disord, 2019, 242: 111-122. doi: 10.1016/j.jad.2018.08.054.
[20] Isometsä ET. Suicides in mood disorders in psychiatric settings in nordic national register-based studies [J]. Front Psychiatry, 2020, 11: 721. doi: 10.3389/fpsyt.2020.00721.
[21] Pallaskorpi S, Suominen K, Ketokivi M, et al. Incidence and predictors of suicide attempts in bipolar I and II disorders: a 5-year follow-up study [J]. Bipolar Disord, 2017, 19(1): 13-22.
[22] Michaels MS, Balthrop T, Pulido A, et al. Is the higher number of suicide attempts in bipolar disorder vs. major depressive disorder attributable to illness severity? [J]. Arch Suicide Res, 2018, 22(1): 46-56.
[23] Persons JE, Coryell WH, Solomon DA, et al. Mixed state and suicide: Is the effect of mixed state on suicidal behavior more than the sum of its parts? [J]. Bipolar Disord, 2018, 20(1): 35-41.
[24] Tang H, Xiong T, Shi J, et al. Global and reflective rumination are related to suicide attempts among patients experiencing major depressive episodes [J]. BMC Psychiatry, 2021, 21(1): 117.
[25] Tondo L, Lepri B, Baldessarini RJ. Suicidal risks among 2826 Sardinian major affective disorder patients [J]. Acta Psychiatr Scand, 2007, 116(6): 419-428.
[26] 杨贵成, 徐记芳, 赫利寒. 双相情感障碍伴自杀行为的流行病学分析[J]. 中国临床实用医学, 2021, 12(1): 57-60. YANG Guicheng, XU Jifang, HE Lihan. Epidemiological analysis of bipolar disorder with suicidal behavior [J]. China Clinical Practical Medicine, 2021, 12(1): 57-60.
[27] 王煜, 方伟, 王亮, 等. 基于Adaboost-CART模型的动卧列车客座率预测[J]. 中国铁路, 2019, 10: 34-38. WANG Yu, FANG Wei, WANG Liang, et al. Seating Rate Prediction for EMU sleeping Train Based On Adaboost-CART Model [J]. Chinese Railways, 2019, 10: 34-38.
[28] 池亚平, 凌志婷, 王志强, 等. 基于支持向量机与Adaboost的入侵检测系统[J]. 计算机工程, 2019, 45(10): 183-188,202. CHI Yaping, LING Zhiting, WANG Zhiqiang, et al. Intrusion detection system based on support vector machine and adaboost [J]. Computer Engineering, 2019, 45(10): 183-188,202.
[29] 李振铠, 刘小杰, 郝良元, 等. 基于AdaBoost算法的高炉铁水含钒预测模型[J]. 华北理工大学学报(自然科学版), 2020, 42(2): 20-28. LI Zhenkai, LIU Xiaojie, HAO Liangyuan, et al. Prediction model of vanadium containing molten iron in blast furnace based on adaboost algorithm [J]. Journal of North China University of Science and Technology(Natural Science Edition), 2020, 42(2): 20-28.
[30] 刘子源, 王翀, 李晓飞. 基于AdaBoost算法的直升机结构制造复杂度估算[J]. 中国电子科学研究院学报, 2020, 15(10): 977-983. LIU Ziyuan, WANG Chong, LI Xiaofei. The helicopter structure manufacture complexity estimation based on adaBoost algorithm [J]. Journal of China Academy of Electronics and Information Technology, 2020, 15(10): 977-983.
[31] 费云利. 计算机逻辑回归分析[J]. 湖南工业职业技术学院学报, 2020, 20(1): 14-17. FEI Yunli. Logistic regression analysis on computer [J]. Journal of Hunan Industry Polytechnic, 2020, 20(1): 14-17.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed