DBO-SVM与SVM预测wAMD患者抗VEGF治疗短期反应性能对比

黄 兰超, 左 慧懿, 黄敏 丽*, 黄 宝宇, 何 　剑, 张 宇杰
广西医科大学第一附属医院眼科

摘要


目的：本研究通过比较SVM模型在DBO优化算法优化前后预测wAMD患者对抗VEGF治疗短期反应的性
能，旨在提升SVM模型的性能，以帮助医生为每位患者制定个性化和有效的治疗方案。方法：回顾性收集在广西医
科大学第一附属医院眼科就诊治疗的96名接受抗VEGF治疗的wAMD患者数据，包括人口统计特征、基线及前3个
月的临床特征和OCT数据。构建SVM模型以及结合DBO优化器的改良模型DBO-SVM，按14：6：5将数据划分为
训练集和内、外部验证集，用于训练和验证模型，通过AUC、灵敏度以及预测准确率等指标比较优化前后模型预测
性能，并评估模型拟合程度及泛化能力。结果：DBO-SVM和SVM模型内部验证AUC、灵敏度和预测准确率分别
是0.76/0.75，0.82/0.82，0.76/0.72；DBO-SVM和SCM模型的外部验证AUC、灵敏度和预测准确率分别是0.89/0.77，
0.87/0.60，0.75/0.67。DBO-SVM模型内外部验证结果相似，预测性能良好且泛化能力较高。DBO-SVM模型预测
性能整体优于SVM模型。结论：DBO优化算法显著提升了SVM模型在预测wAMD患者对抗VEGF治疗反应的性能，
增强了其有效性和应用潜力，具有更高的临床和公共卫生价值。

关键词


蜣螂优化算法（DBO）；支持向量机（SVM）；湿性年龄相关性黄斑变性（wAMD）；抗VEGF治疗

全文:

PDF


参考


[1]黄叔仁.老年性黄斑变性[J].实用眼科杂志，1992，

10（7）：4.

[2]Gheorghe A, Mahdi L, Musat O. Age-related macular

degeneration [J]. Romanian journal of ophthalmology, 2015,

59(2): 74.

[3]Wong W L, Su X, Li X, et al. Global prevalence

of age-related macular degeneration and disease burden

projection for 2020 and 2040: A systematic review and metaanalysis [J]. The Lancet Global Health, 2014, 2(2): e106-e16.

[4]Xi-Yan A, Jiu-Mei S. Treatment of wet agerelated macular degeneration [J]. International Review of

Ophthalmology, 2015, 39(4): 265.

[5]Yang S, Zhao J, Sun X. Resistance to anti-vegf

therapy in neovascular age-related macular degeneration: A

comprehensive review [J]. Drug design, development and

therapy, 2016: 1857-67.

[6]Ricci F, Bandello F, Navarra P, et al. Neovascular

age-related macular degeneration: Therapeutic management

and new-upcoming approaches [J]. International journal of

molecular sciences, 2020, 21(21): 8242.

[7]Pugazhendhi A, Hubbell M, Jairam P, et al.

Neovascular macular degeneration: A review of etiology,

risk factors, and recent advances in research and therapy [J].

International Journal of Molecular Sciences, 2021, 22(3):

1170.

[8]Perepelkina T, Fulton A B. Artificial intelligence

(ai) applications for age-related macular degeneration (amd)

and other retinal dystrophies; proceedings of the Seminars in

ophthalmology, F, 2021 [C]. Taylor & Francis.

[9]Bogunović H, Waldstein S M, Schlegl T, et al.

Prediction of anti-vegf treatment requirements in neovascular

amd using a machine learning approach [J]. Investigative

ophthalmology & visual science, 2017, 58(7): 3240-8.

[10]Perkins S W, Wu A K, Singh R P. Predictors

of limited early response to anti-vascular endothelial

growth factor therapy in neovascular age-related macular

degeneration with machine learning feature importance [J].

Saudi Journal of Ophthalmology, 2022, 36(3): 315-21.

[11]Singh L K, Khanna M, Thawkar S. A novel hybrid

robust architecture for automatic screening of glaucoma

using fundus photos, built on feature selection and machine

learning‐nature driven computing [J]. Expert Systems, 2022,

39(10): e13069.

[12]Zhang H, Yang S, Xu D. An improved optimization

dung beetle algorithm; proceedings of the 2023 4th

International Symposium on Computer Engineering and

Intelligent Communications (ISCEIC), F, 2023 [C]. IEEE.

[13]Alamgeer M, Alruwais N, Alshahrani H M, et al.

Dung beetle optimization with deep feature fusion model for

lung cancer detection and classification [J]. Cancers, 2023,

15(15): 3982.

[14]Amoaku W, Chakravarthy U, Gale R, et al.

Defining response to anti-vegf therapies in neovascular amd

[J]. Eye, 2015, 29(6): 721-31.

[15]Chandra R S, Ying G-s. Evaluation of multiple

machine learning models for predicting number of anti-vegf

injections in the comparison of amd treatment trials (catt) [J].

Translational Vision Science & Technology, 2023, 12(1): 18-.

[16]Huang W-C, Lee P-L, Liu Y-T, et al. Support

vector machine prediction of obstructive sleep apnea in a large-scale chinese clinical sample [J]. Sleep, 2020, 43(7): zsz295.

[17]Vichianin Y, Khummongkol A, Chiewvit P, et

al. Accuracy of support-vector machines for diagnosis of

alzheimer’s disease, using volume of brain obtained by

structural mri at siriraj hospital [J]. Frontiers in neurology,

2021, 12: 640696.

[18]Perkins S W, Wu A K, Singh R P. Machine learning

prediction of limited early response to anti-vegf therapy in

neovascular age-related macular degeneration in routine

clinical practice [J]. Investigative Ophthalmology & Visual

Science, 2022, 63(7): 322–F0153-322–F.

[19]Kar S S, Cetin H, Lunasco L, et al. Oct-derived

radiomic features predict anti–vegf response and durability

in neovascular age-related macular degeneration [J].

Ophthalmology Science, 2022, 2(4): 100171.

[20]Khan B, Fatima H, Qureshi A, et al. Drawbacks

of artificial intelligence and their potential solutions in the

healthcare sector [J]. Biomedical Materials & Devices, 2023,

1(2): 731-8.

[21]Zhou X, Wen H, Li Z, et al. An interpretable model

for the susceptibility of rainfall-induced shallow landslides

based on shap and xgboost [J]. Geocarto International, 2022,

37(26): 13419-50.


Refbacks

  • 当前没有refback。