Multi-input Deep Learning Model for RP Diagnosis Using FVEP and Prior Knowledge

被引:0
|
作者
Chen, Yuguang [1 ,2 ,3 ,4 ,5 ,6 ]
Shen, Mei [2 ,3 ,4 ,5 ,6 ]
Lu, Dongmei [2 ,3 ,4 ,5 ,6 ]
Lin, Jun [9 ]
Hu, Jiaoyue [2 ,3 ,4 ,5 ,6 ,7 ]
Li, Shiying [2 ,3 ,4 ,5 ,6 ]
Liu, Zuguo [1 ,2 ,3 ,4 ,5 ,6 ,7 ,8 ]
机构
[1] Xiamen Univ, Inst Artificial Intelligence, Xiamen 361102, Fujian, Peoples R China
[2] Xiamen Univ, Dept Ophthalmol, Xiangan Hosp, Xiamen 361005, Fujian, Peoples R China
[3] Fujian Prov Key Lab Ophthalmol & Visual Sci, Xiamen 361005, Fujian, Peoples R China
[4] Fujian Engn & Res Ctr Eye Regenerat Med, Xiamen 361005, Fujian, Peoples R China
[5] Xiamen Univ, Eye Inst, Xiamen 361005, Fujian, Peoples R China
[6] Xiamen Univ, Sch Med, Xiamen 361005, Fujian, Peoples R China
[7] Xiamen Univ, Xiamen Eye Ctr, Xiamen 361005, Fujian, Peoples R China
[8] Univ South China, Dept Ophthalmol, Affiliated Hosp 1, Hengyang 421001, Hunan, Peoples R China
[9] Yongchuan Dist Peoples Hosp Chongqing, Dept Ophthalmol, Chongqing 402160, Peoples R China
来源
ADVANCED INTELLIGENT COMPUTING IN BIOINFORMATICS, PT I, ICIC 2024 | 2024年 / 14881卷
基金
中国国家自然科学基金;
关键词
FVEP; Retinitis Pigmentosa; Deep learning; Out-of-distribution detection;
D O I
10.1007/978-981-97-5689-6_25
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Retinitis Pigmentosa (RP) is a hereditary disease characterized by progressive damage to the visual pathway, ultimately leading to vision loss. Flash Visual Evoked Potential (FVEP) serves as an effective tool for diagnosing RP, and automatic classification of FVEP using deep learning can alleviate the workload of doctors and improve work efficiency. This study proposed a multi input neural network for RP and other anomaly recognition: MGPResNet. One branch of the model conducts full connection on manually crafted features to integrate them, while the other branch adopts a 1D ResNet as its basic architecture, it incorporates global convolutional blocks and pyramid pooling blocks to extract features from FVEP waveforms at deeper levels and different scales. Subsequently, the features extracted by the two branches are concatenated, followed by full connection and activation layers to output the classification probabilities. The model was validated on the FVEP datasets of two hospitals. The proposed method demonstrated excellent accuracy on clinical datasets, with an accuracy of 96.80%, average precision of 96.52%, average recall of 96.47%, and average F1_score of 96.49%. It validated the significant potential of deep learning in the analysis of visual electrophysiological signals, provided an important foundation and new insights for the future use of deep learning techniques in clinical diagnosis and treatment.
引用
收藏
页码:287 / 299
页数:13
相关论文
共 50 条
  • [31] Integrating Prior Knowledge into Deep Learning
    Diligenti, Michelangelo
    Roychowdhury, Soumali
    Gori, Marco
    2017 16TH IEEE INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND APPLICATIONS (ICMLA), 2017, : 920 - 923
  • [32] THYROID CANCER COMPUTER-AIDED DIAGNOSIS SYSTEM USING MRI-BASED MULTI-INPUT CNN MODEL
    Naglah, Ahmed
    Khalifa, Fahmi
    Khaled, Reem
    Razek, Ahmed Abdel Khalek Abdel
    El-Baz, Ayman
    2021 IEEE 18TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (ISBI), 2021, : 1691 - 1694
  • [33] Multi-input square iterative learning control with input rate limits and bounds
    Driessen, BJ
    Sadegh, N
    IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2002, 32 (04): : 545 - 550
  • [34] A Hybrid Delay Model for Interconnected Multi-Input Gates
    Ferdowsi, Arman
    Fugger, Matthias
    Salzmann, Josef
    Schmid, Ulrich
    2023 26TH EUROMICRO CONFERENCE ON DIGITAL SYSTEM DESIGN, DSD 2023, 2023, : 381 - 390
  • [35] Parameter estimation for multirate multi-input systems using auxiliary model and multi-innovation
    Lili Han and Feng Ding School of IoT Engineering
    JournalofSystemsEngineeringandElectronics, 2010, 21 (06) : 1079 - 1083
  • [36] CCSNscore: A Multi-input Deep Learning Tool for Classification of Core-collapse Supernovae Using SED-machine Spectra
    Sharma, Yashvi
    Mahabal, Ashish A.
    Sollerman, Jesper
    Fremling, Christoffer
    Kulkarni, S. R.
    Rehemtulla, Nabeel
    Miller, Adam A.
    Aubert, Marie
    Chen, Tracy X.
    Coughlin, Michael W.
    Graham, Matthew J.
    Hale, David
    Kasliwal, Mansi M.
    Kim, Young-Lo
    Neill, James D.
    Purdum, Josiah N.
    Rusholme, Ben
    Singh, Avinash
    Sravan, Niharika
    PUBLICATIONS OF THE ASTRONOMICAL SOCIETY OF THE PACIFIC, 2025, 137 (03)
  • [37] A fault diagnosis model for rolling bearings based on a multi-input layer convolutional neural network
    Zan T.
    Wang H.
    Liu Z.
    Wang M.
    Gao X.
    Zhendong yu Chongji/Journal of Vibration and Shock, 2020, 39 (12): : 142 - 149and163
  • [38] A Comparative Study of Multiple Deep Learning Models Based on Multi-Input Resolution for Breast Ultrasound Images
    Wu, Huaiyu
    Ye, Xiuqin
    Jiang, Yitao
    Tian, Hongtian
    Yang, Keen
    Cui, Chen
    Shi, Siyuan
    Liu, Yan
    Huang, Sijing
    Chen, Jing
    Xu, Jinfeng
    Dong, Fajin
    FRONTIERS IN ONCOLOGY, 2022, 12
  • [39] Multi-input square iterative learning control with bounded inputs
    Driessen, BJ
    Sadegh, N
    Kwok, KS
    IEEE SOUTHEASTCON 2001: ENGINEERING THE FUTURE, PROCEEDINGS, 2001, : 62 - 64
  • [40] Multi-input square iterative learning control with bounded inputs
    Driessen, BJ
    Sadegh, N
    JOURNAL OF DYNAMIC SYSTEMS MEASUREMENT AND CONTROL-TRANSACTIONS OF THE ASME, 2002, 124 (04): : 582 - 584