Predicting visual acuity with machine learning in treated ocular trauma patients

被引:0
作者
Zhi-Lu Zhou [1 ,2 ]
Yi-Fei Yan [3 ,4 ]
Jie-Min Chen [2 ]
Rui-Jue Liu [2 ]
Xiao-Ying Yu [2 ]
Meng Wang [2 ]
Hong-Xia Hao [2 ,5 ]
Dong-Mei Liu [2 ]
Qi Zhang [3 ,4 ]
Jie Wang [1 ]
Wen-Tao Xia [2 ]
机构
[1] Department of Forensic Medicine, Guizhou Medical University
[2] Shanghai Key Laboratory of Forensic Medicine, Shanghai Forensic Service Platform, Institute of Forensic Science, Ministry of Justice
[3] The SMART (Smart Medicine and AI-based Radiology Technology) Lab , Shanghai Institute for Advanced Communication and Data Science, Shanghai University
[4] School of Communication and Information Engineering, Shanghai University
[5] Basic Medical College, Jiamusi University
基金
中国国家自然科学基金; 国家重点研发计划;
关键词
D O I
暂无
中图分类号
TP181 [自动推理、机器学习]; R779.1 [眼损伤与异物];
学科分类号
081104 ; 0812 ; 0835 ; 100212 ; 1405 ;
摘要
AIM: To predict best-corrected visual acuity(BCVA) by machine learning in patients with ocular trauma who were treated for at least 6mo. METHODS: The internal dataset consisted of 850 patients with 1589 eyes and an average age of 44.29y. The initial visual acuity was 0.99 log MAR. The test dataset consisted of 60 patients with 100 eyes collected while the model was optimized. Four different machine-learning algorithms(Extreme Gradient Boosting, support vector regression, Bayesian ridge, and random forest regressor) were used to predict BCVA, and four algorithms(Extreme Gradient Boosting, support vector machine, logistic regression, and random forest classifier) were used to classify BCVA in patients with ocular trauma after treatment for 6mo or longer. Clinical features were obtained from outpatient records, and ocular parameters were extracted from optical coherence tomography images and fundus photographs. These features were put into different machine-learning models, and the obtained predicted values were compared with the actual BCVA values. The best-performing model and the best variable selected were further evaluated in the test dataset.RESULTS: There was a significant correlation between the predicted and actual values [all Pearson correlation coefficient(PCC)>0.6]. Considering only the data from the traumatic group(group A) into account, the lowest mean absolute error(MAE) and root mean square error(RMSE) were 0.30 and 0.40 log MAR, respectively. In the traumatic and healthy groups(group B), the lowest MAE and RMSE were 0.20 and 0.33 log MAR, respectively. The sensitivity was always higher than the specificity in group A, in contrast to the results in group B. The classification accuracy and precision were above 0.80 in both groups. The MAE, RMSE, and PCC of the test dataset were 0.20, 0.29, and 0.96, respectively. The sensitivity, precision, specificity, and accuracy of the test dataset were 0.83, 0.92, 0.95, and 0.90, respectively.CONCLUSION: Predicting BCVA using machine-learning models in patients with treated ocular trauma is accurate and helpful in the identification of visual dysfunction.
引用
收藏
页码:1005 / 1014
页数:10
相关论文
共 22 条
[1]   Pediatric ocular trauma with pars plana vitrectomy in Southwest of China: clinical characteristics and outcomes [J].
Zuo-Xin Qin ;
Yan He ;
Yu-Fei Xu ;
Tao Yu ;
Yong Liu ;
Nan Wu .
International Journal of Ophthalmology, 2021, 14 (09) :1321-1326
[2]   钝性眼外伤对视网膜神经纤维层的影响(英文) [J].
Kumar Aalok ;
Singh Vipin .
国际眼科杂志, 2021, 21 (02) :199-203
[3]   异常中心凹内层对特发性黄斑前膜术后视力的预测价值研究 [J].
郑玥 ;
项振扬 ;
杨友谊 ;
陈爱菊 ;
陈绮 .
中国眼耳鼻喉科杂志, 2021, 21 (01) :9-15+28
[4]  
外伤性黄斑病变光学相干断层成像的应用研究[J].眼外伤职业眼病杂志,2003(08)
[5]   Outcomes and longitudinal trend of traumatic cataract wound dehiscence in patients with blunt ocular injury [J].
Hou, Chiun-Ho ;
Lu, Yu-Chin ;
Pu, Christy ;
Chang, Yin-Hsi ;
Lin, Ken-Kuo ;
Lee, Jiahn-Shing ;
Chen, Kuan-Jen .
SCIENTIFIC REPORTS, 2021, 11 (01)
[6]  
Wei Ling,He Wenwen,Wang Jinrui,Zhang Keke,Du Yu,Qi Jiao,Meng Jiaqi,Qiu Xiaodi,Cai Lei,Fan Qi,Zhao Zhennan,Tang Yating,Ni Shuang,Guo Haike,Song Yunxiao,He Xixi,Ding Dayong,Lu Yi,Zhu Xiangjia.An Optical Coherence Tomography-Based Deep Learning Algorithm for Visual Acuity Prediction of Highly Myopic Eyes After Cataract Surgery
[J].Frontiers in Cell and Developmental Biology,2021
[7]  
Phillips Hunter H,Blegen IV Halward J,Anthony Christopher,Davies Brett W,Wedel Marissa L,Reed Donovan S.Pars Plana Vitrectomy following Traumatic Ocular Injury and Initial Globe Repair: A Retrospective Analysis of Clinical Outcomes[J].Military Medicine,2021
[8]  
Murphy Declan C,Nasrulloh Amar V,Lendrem Clare,Graziado Sara,Alberti Mark,la Cour Morten,Obara Boguslaw,Steel David H W.Predicting Postoperative Vision for Macular Hole with Automated Image Analysis[J].Ophthalmology. Retina,2020
[9]  
Sebastian M. Waldstein,Wolf-Dieter Vogl,Hrvoje Bogunovic,Amir Sadeghipour,Sophie Riedl,Ursula Schmidt-Erfurth.Characterization of Drusen and Hyperreflective Foci as Biomarkers for Disease Progression in Age-Related Macular Degeneration Using Artificial Intelligence in Optical Coherence Tomography[J].JAMA Ophthalmology,2020
[10]  
Huang Ching-Yen,Kuo Ren-Jieh,Li Cheng-Han,Ting Daniel S,Kang Eugene Yu-Chuan,Lai Chi-Chun,Tseng Hsiao-Jung,Yang Lan-Yan,Wu Wei-Chi.Prediction of visual outcomes by an artificial neural network following intravitreal injection and laser therapy for retinopathy of prematurity[J].The British journal of ophthalmology,2019(9)