Page 127 - 《中国药房》2025年15期
P. 127
现出了更好的适用性,由此可知,RF模型具有更好的预 enabled risk prediction of chronic obstructive pulmonary
测性能。 disease with unbalanced data[J]. Comput Methods Pro‐
综上所述,本研究基于真实世界数据,深入分析了 grams Biomed,2023,230:107340.
中重度抑郁症住院患者使用SNRI类药物后影响疗效的 [10] 谢宏宇,侯艳,李康. 基于正则化回归的组学数据变量筛
相关特征变量,建立的 RF 模型可对中重度抑郁症住院 选方法[J]. 中国卫生统计,2016,33(4):733-736.
患者使用SNRI类药物后的疗效进行预测。该模型可以 [11] 崔佳旭,杨博 . 贝叶斯优化方法和应用综述[J]. 软件学
报,2018,29(10):3068-3090.
集成到移动应用程序中,并结合电子病历数据,为患者
[12] 吴妍. 基于机器学习的糖尿病早期诊断模型及可解释分
提供个体化的治疗建议。临床医师可以将影响疗效的
析[D]. 重庆:重庆理工大学,2024.
变量输入预测模型以提前识别可能治疗无效的患者,及 [13] NUÑEZ N A,JOSEPH B,PAHWA M,et al. Augmenta‐
时对患者调整治疗策略,以避免无效的药物使用和治疗 tion strategies for treatment resistant major depression:a
延误。但是,本研究也存在一定的局限性:一是本研究 systematic review and network meta-analysis[J]. J Affect
属于单中心回顾性研究,存在一定的回忆偏倚,建议未 Disord,2022,302:385-400.
来开展更大规模的前瞻性研究并在更大的外部数据集 [14] YAN Y S,YANG X,WANG M,et al. Efficacy and accepta-
上进一步验证本研究结论。二是本研究所收集的病历 bility of second-generation antipsychotics with antidepres‐
信息不够全面,可能限制了预测模型的价值,建议未来 sants in unipolar depression augmentation:a systematic
研究纳入更全面的因素,包括药物代谢基因及环境因素 review and network meta-analysis[J]. Psychol Med,2022,
(饮食习惯和压力源等)作为预测变量,以提高预测模型 52(12):2224-2231.
的性能。 [15] 李凌江,马辛 . 中国抑郁障碍防治指南[M]. 2 版 . 北京:
中华医学电子音像出版社,2015:59-80.
参考文献
[ 1 ] 陆林 . 沈渔邨精神病学[M]. 6 版 . 北京:人民卫生出版 [16] FAVA M,RUSH A J,ALPERT J E,et al. Difference in
treatment outcome in outpatients with anxious versus non‐
社,2018:877-900.
[ 2 ] STAHL S M,GRADY M M,MORET C,et al. SNRIs: anxious depression:a STAR*D report[J]. Am J Psychia‐
try,2008,165(3):342-351.
their pharmacology,clinical efficacy,and tolerability in
[17] CUKOR D,RUE T,HEAGERTY P,et al. Anxiety and re‐
comparison with other classes of antidepressants[J]. CNS
Spectr,2005,10(9):732-747. sponse to treatment of depression in people undergoing
maintenance hemodialysis[J]. Clin J Am Soc Nephrol,
[ 3 ] LIN H S,LIN C H. Early improvement in HAMD-17 and
HAMD-6 scores predicts ultimate response and remission 2023,18(8):1075-1076.
[18] 谢焕山,黄毅,黄善情,等. 基于治疗药物监测的O-去甲
for depressed patients treated with fluoxetine or ECT[J]. J
Affect Disord,2019,245:91-97. 基文拉法辛与文拉法辛浓度比值影响因素分析[J]. 中国
临床药理学杂志,2022,38(20):2419-2422.
[ 4 ] SZEGEDI A,JANSEN W T,VAN WILLIGENBURG A P
P,et al. Early improvement in the first 2 weeks as a pre‐ [19] BUCKMAN J J,UNDERWOOD A,CLARKE K,et al.
Risk factors for relapse and recurrence of depression in
dictor of treatment outcome in patients with major depres‐
sive disorder:a meta-analysis including 6 562 patients[J]. adults and how they operate:a four-phase systematic re‐
view and meta-synthesis[J]. Clin Psychol Rev,2018,64:
J Clin Psychiatry,2009,70(3):344-353.
[ 5 ] LIAO X M,SU Y N,WANG Y,et al. Antidepressant treat‐ 13-38.
[20] DAVYDOVA T V,NEVIDIMOVA T I,VETRILE L A,
ment strategy with an early onset of action improves the
clinical outcome in patients with major depressive disor‐ et al. Correlation of antibodies to neurotransmitters in the
sera of women with alcohol dependence and depressive
der and high anxiety:a multicenter and 6-week follow-up
study[J]. Chin Med J (Engl),2020,133(6):726-728. disorders[J]. Bull Exp Biol Med,2021,171(6):704-706.
[21] NUNES E V. Alcohol and the etiology of depression[J].
[ 6 ] RUTLEDGE R B,CHEKROUD A M,HUYS Q J. Ma‐
chine learning and big data in psychiatry:toward clinical Am J Psychiatry,2023,180(3):179-181.
[22] ALSHEIKH A M,ELEMAM M O,EL-BAHNASAWI M.
applications[J]. Curr Opin Neurobiol,2019,55:152-159.
[ 7 ] 中华医学会行为医学分会,中华医学会行为医学分会认 Treatment of depression with alcohol and substance de‐
pendence:a systematic review[J]. Cureus,2020,12(10):
知应对治疗学组,王长虹,等. 抑郁症治疗与管理的专家
e11168.
推荐意见:2022年[J]. 中华行为医学与脑科学杂志,2023
[23] TANDON N,TANDON R. Machine learning in psychia‐
(3):193-202.
[ 8 ] CARPENITO T,MANJOURIDES J. MISL:multiple im‐ try:standards and guidelines[J]. Asian J Psychiatry,2019,
44:A1-A4.
putation by super learning[J]. Stat Methods Med Res,
(收稿日期:2025-02-10 修回日期:2025-05-29)
2022,31(10):1904-1915.
(编辑:胡晓霖)
[ 9 ] WANG X C,REN H,REN J H,et al. Machine learning-
中国药房 2025年第36卷第15期 China Pharmacy 2025 Vol. 36 No. 15 · 1941 ·

