Detecting Proximal Caries on Periapical Radiographs Using Convolutional Neural Networks with Different Training Strategies on Small Datasets

被引:7
|
作者
Lin, Xiujiao [1 ,2 ]
Hong, Dengwei [1 ,2 ]
Zhang, Dong [3 ]
Huang, Mingyi [3 ]
Yu, Hao [1 ,2 ,4 ]
机构
[1] Fujian Med Univ, Sch & Hosp Stomatol, Fujian Prov Engn Res Ctr Oral Biomat, Fuzhou 350005, Peoples R China
[2] Fujian Med Univ, Sch & Hosp Stomatol, Dept Prosthodont, Fuzhou 350005, Peoples R China
[3] Fuzhou Univ, Coll Comp & Data Sci, Fuzhou 350025, Peoples R China
[4] Nagasaki Univ, Grad Sch Biomed Sci, Dept Appl Prosthodont, Nagasaki 8528521, Japan
关键词
neural networks; proximal caries; training strategy; small dataset; periapical radiograph; DENTAL-CARIES; DIAGNOSTIC-ACCURACY; PERMANENT TEETH; CHINA;
D O I
10.3390/diagnostics12051047
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
The present study aimed to evaluate the performance of convolutional neural networks (CNNs) that were trained with small datasets using different strategies in the detection of proximal caries at different levels of severity on periapical radiographs. Small datasets containing 800 periapical radiographs were randomly categorized into a training and validation dataset (n = 600) and a test dataset (n = 200). A pretrained Cifar-10Net CNN was used in the present study. Different training strategies were used to train the CNN model independently; these strategies were defined as image recognition (IR), edge extraction (EE), and image segmentation (IS). Different metrics, such as sensitivity and area under the receiver operating characteristic curve (AUC), for the trained CNN and human observers were analysed to evaluate the performance in detecting proximal caries. IR, EE, and IS recognition modes and human eyes achieved AUCs of 0.805, 0.860, 0.549, and 0.767, respectively, with the EE recognition mode having the highest values (p all < 0.05). The EE recognition mode was significantly more sensitive in detecting both enamel and dentin caries than human eyes (p all < 0.05). The CNN trained with the EE strategy, the best performer in the present study, showed potential utility in detecting proximal caries on periapical radiographs when using small datasets.
引用
收藏
页数:13
相关论文
共 50 条
  • [11] Semantic segmentation on small datasets of satellite images using convolutional neural networks
    Younis, Mohammed Chachan
    Keedwell, Edward
    [J]. JOURNAL OF APPLIED REMOTE SENSING, 2019, 13 (04)
  • [12] Using different training strategies for urban land-use classification based on convolutional neural networks
    Qiu, Tianqi
    He, Huagui
    Liang, Xiaojin
    Chen, Fei
    Chen, Zhaoxia
    Liu, Yang
    [J]. FRONTIERS IN ENVIRONMENTAL SCIENCE, 2022, 10
  • [13] Automatic detection and classification of nasopalatine duct cyst and periapical cyst on panoramic radiographs using deep convolutional neural networks
    Lee, Han-Sol
    Yang, Su
    Han, Ji-Yong
    Kang, Ju-Hee
    Kim, Jo-Eun
    Huh, Kyung-Hoe
    Yi, Won-Jin
    Heo, Min-Suk
    Lee, Sam-Sun
    [J]. ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY ORAL RADIOLOGY, 2024, 138 (01): : 184 - 195
  • [14] Speech Recognition Using Convolutional Neural Networks on Small Training Sets
    Poliyev, A. V.
    Korsun, O. N.
    [J]. 2019 WORKSHOP ON MATERIALS AND ENGINEERING IN AERONAUTICS, 2020, 714
  • [15] Deep learning convolutional neural network algorithms for the early detection and diagnosis of dental caries on periapical radiographs: A systematic review
    Musri, Nabilla
    Christie, Brenda
    Ichwan, Solachuddin Jauhari Arief
    Cahyanto, Arief
    [J]. IMAGING SCIENCE IN DENTISTRY, 2021, 51 (03) : 237 - 242
  • [16] Training Strategies for Convolutional Neural Networks with Transformed Input
    Khandani, Masoumeh Kalantari
    Mikhael, Wasfy B.
    [J]. 2021 IEEE INTERNATIONAL MIDWEST SYMPOSIUM ON CIRCUITS AND SYSTEMS (MWSCAS), 2021, : 1058 - 1061
  • [17] CONVOLUTIONAL NEURAL NETWORKS AND TRAINING STRATEGIES FOR SKIN DETECTION
    Kim, Yoonsik
    Hwang, Insung
    Cho, Nam Ik
    [J]. 2017 24TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2017, : 3919 - 3923
  • [18] The need for training and benchmark datasets for convolutional neural networks in flood applications
    Khouakhi, Abdou
    Zawadzka, Joanna
    Truckell, Ian
    [J]. HYDROLOGY RESEARCH, 2022, 53 (06): : 795 - 806
  • [19] Multimodal Emotion Recognition via Convolutional Neural Networks: Comparison of different strategies on two multimodal datasets
    Bilotti, U.
    Bisogni, C.
    De Marsico, M.
    Tramonte, S.
    [J]. ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2024, 130
  • [20] Pneumothorax Detection in Chest Radiographs Using Convolutional Neural Networks
    Aviel, Blumenfeld
    Eli, Konen
    Hayit, Greenspan
    [J]. MEDICAL IMAGING 2018: COMPUTER-AIDED DIAGNOSIS, 2018, 10575