Advances in the application of lightweight deep learning models in diagnosis and treatment of dental diseases

doi:10.19438/j.cjoms.2025.05.014

Abstract

Abstract: The lightweighting of medical deep learning models through adaptive compression techniques significantly reduces computational complexity while maintaining diagnostic accuracy, enabling deployment on edge devices and promoting the implementation of AI in clinical scenarios such as emergency care, surgical guidance, and portable monitoring. In the field of dental care, research on lightweight models has employed architectures such as MobileNet and YOLO, along with techniques like convolutional optimization and knowledge distillation, to accomplish imaging-based segmentation of anatomical structures such as teeth and jaws, and the diagnosis of diseases including caries and oral cancer. Some of these models have been integrated into edge devices such as smart toothbrushes and mobile applications. This paper reviewed the application progress of lightweight models in diagnosis and treatment of dental disease, aiming to provide references for further development and clinical transformation of related technologies.

Key words: Lightweight model, Deep learning, Diagnosis and treatment of dental disease, Edge devices, Real-time diagnosis

CLC Number:

R782
R739.8

Gao Anqi, Huang Xinrui, Zhang Xiaofan, Zhang Shaoting, Hong Yi, Chen Lei, Wang Xudong. Advances in the application of lightweight deep learning models in diagnosis and treatment of dental diseases[J]. China Journal of Oral and Maxillofacial Surgery, 2025, 23(5): 517-522.

References

[1] Rajpurkar P, Chen E, Banerjee O, et al.AI in health and medicine[J]. Nat Med, 2022, 28(1): 31-38.
[2] Lee M, Hwang EJ, Lee JH, et al.Artificial intelligence for low-dose CT lung cancer screening: comparison of utilization scenarios[J]. Am J Roentgenol, 2025, 225(1):e2532829.
[3] Yang Z, Wei T, Liang Y, et al.A foundation model for generalizable cancer diagnosis and survival prediction from histopathological images[J]. Nat Commun, 2025, 16(1): 2366.
[4] Wang Z, Liu C, Zhang S, et al.Foundation model for endoscopy video analysis via large-scale self-supervised pre-train[C]//Int Conf Med Image Comput Comput-Assisted Intervention[M]. Cham: Springer Nature Switzerland, 2023:101-111.
[5] Li X, Peng L, Wang YP, et al.Open challenges and opportunities in federated foundation models towards biomedical healthcare[J]. BioData Min, 2025, 18(1): 2.
[6] Khan W, Leem S, See KB, et al. A comprehensive survey of foundation models in medicine[J]. IEEE Rev Biomed Eng, 2025 May 6. [Epub ahead of print].
[7] Habib G, Qureshi S.Compressed lightweight deep learning models for resource-constrained Internet of things devices in the healthcare sector[J]. Expert Syst, 2025, 42(1): e13269.
[8] Goutam B, Hashmi MF, Geem ZW, et al.A comprehensive review of deep learning strategies in retinal disease diagnosis using fundus images[J]. IEEE Access, 2022, 10: 57796-57823.
[9] Chen H, Li C, Wang G, et al.GasHis-transformer: a multi-scale visual transformer approach for gastric histopathological image detection[J]. Pattern Recognit, 2022, 130: 108827.
[10] Wu R, Liu Y, Ning G, et al. Ultralight VM-UNet: parallel vision mamba significantly reduces parameters for skin lesion segmentation[J/OL]. Patterns, 2025 June 26.[Online now]. https://doi.org/10.1016/j.patter.2025.101298.
[11] Qayoom A, Xie J, Ali H.Polyp segmentation in medical imaging: challenges, approaches and future directions[J]. Artif Intell Rev, 2025, 58(6):169.
[12] Zhou R, Wang D, Zhang H, et al.Vision techniques for anatomical structures in laparoscopic surgery: a comprehensive review[J]. Front Surg, 2025, 12: 1557153.
[13] Saadatnejad S, Oveisi M, Hashemi M.LSTM-based ECG classification for continuous monitoring on personal wearable devices[J]. IEEE J Biomed Health Inform, 2020, 24(2): 515-523.
[14] Liu WH, Zhang MX, Zhang YD, et al.Real-time multilead convolutional neural network for myocardial infarction detection[J]. IEEE J Biomed Health Inform, 2018, 22(5): 1434-1444.
[15] Cho H, Song I, Jang J, et al.A lightweight deep learning network on a system-on-chip for wearable ultrasound bladder volume measurement systems: preliminary study[J]. Bioengineering(Basel), 2023, 10(5): 525.
[16] Lei Y, Chen X, Wang Y, et al.A lightweight knowledge-distillation-based model for the detection and classification of impacted mandibular third molars[J]. Appl Sci, 2023, 13(17): 9970.
[17] 曾怡峰, 姚潇, 华飞, 等. 面向全景智齿检测的内卷解耦轻量化网络[J]. 中国图象图形学报, 2023, 28(8):2491-2504.
Zeng YF, Yao X, Hua F, et al.A lightweight network—involute and decoupled for panoramic wisdom tooth detection[J]. J Image Graph, 2023, 28(8): 2491-2504.
[18] Jin C, Zhang D, Cao X, et al.Lightweight YOLOv8 for tongue teeth marks and fissures detection based on C2f_DCNv3[J]. Sci Rep, 2025, 15(1): 1560.
[19] 吴杰, 陶青川. 基于改进BiSeNet 的轻量化上颌前牙分割算法[J]. 现代计算机,2023, 29(24): 33-39.
Wu J, Tao QC.A lightweight maxillary anterior segmentation algorithm based on improved BiSeNet[J]. Mod Comput, 2023, 29(24): 33-39.
[20] Chotikkakamthorn K, Ritthipravat P, Kusakunniran W, et al. A lightweight deep learning approach to mouth segmentation in color images[J/OL]. Appl Comput Inform, 2022. https://doi.org/10.1108/ACI-08-2022-0225.
[21] Tang Y, Wang X, Guo S, et al.An improved lightweight tongue image semantic segmentation model based on DeepLabV3+[J]. Biomed Signal Process Control, 2025, 109: 107911.
[22] Lin S, Hao X, Liu Y, et al.Lightweight deep learning methods for panoramic dental X-ray image segmentation[J]. Neural Comput Appl, 2023, 35(11): 8295-8306.
[23] Peng J, Li X, Yang D, et al.Automatic tongue crack extraction for real-time diagnosis[C]//Proceedings of 2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). New Jersey: IEEE, 2020: 694-699.
[24] Khaldi A, Khaldi B, Aiadi O.LCAT-Net: lightweight context-aware deep learning approach for teeth segmentation in panoramic X-rays[J]. Int J Comput Intell Syst, 2024, 17(1): 1-17.
[25] Kong Z, Xiong F, Zhang C, et al.Automated maxillofacial segmentation in panoramic dental X-ray images using an efficient encoder-decoder network[J]. IEEE Access, 2020, 8: 207822-207833.
[26] Wen X, Liu Z, Chu Y, et al.MRCM-UCTransNet: automatic and accurate 3D tooth segmentation network from cone-beam CT images[J]. Int J Imaging Syst Technol, 2024, 34(4): e23139.
[27] Zhang L, Xu F, Li Y, et al.A lightweight convolutional neural network model with receptive field block for C-shaped root canal detection in mandibular second molars[J]. Sci Rep, 2022, 12(1): 17373.
[28] Dai Y, Wang Q, Cui S, et al.MediLite3DNet: a lightweight network for segmentation of nasopharyngeal airways[J]. Med Biol Eng Comput, 2025, 63(4): 1081-1099.
[29] Chang HC, Yu LW, Liu BY, et al.Classification of the implant-ridge relationship utilizing the MobileNet architecture[J]. J Dent Sci, 2024, 19(1): 411-418.
[30] Jin S, Han H, Huang Z, et al.Automatic three-dimensional nasal and pharyngeal airway subregions identification via Vision Transformer[J]. J Dent, 2023, 136: 104595.
[31] Li P, Liu Y, Cui Z, et al.Semantic graph attention with explicit anatomical association modeling for tooth segmentation from CBCT images[J]. IEEE Trans Med Imaging, 2022, 41(11): 3116-3127.
[32] Liu SJ, Kang CM, Huang FH, et al.Mesh segmentation for individual teeth based on two-stream GCN with self-attention[J]. IEEE Access, 2024, 12: 76735-76743.
[33] Xie Z, Hu X, Guo L, et al.A lightweight detection algorithm for tooth cracks in optical images[J]. Comput Biol Med, 2024, 182: 109153.
[34] Hua Y, Chen R, Qin HJE.YOLO-DentSeg: a lightweight real-time model for accurate detection and segmentation of oral diseases in panoramic radiographs[J]. Electronics, 2025, 14(4): 805.
[35] Fatima A, Shafi I, Afzal H, et al.Deep learning-based multiclass instance segmentation for dental lesion detection[C]//2nd International Electronic Conference on Healthcare. Basel, Switzerland: MDPI, 2023: 347.
[36] Vollmer A, Saravi B, Vollmer M, et al.Artificial intelligence-based prediction of oroantral communication after tooth extraction utilizing preoperative panoramic radiography[J]. Diagnostics(Basel), 2022, 12(6): 1406.
[37] Ma T, Zhou X, Yang J, et al.Dental lesion segmentation using an improved icnet network with attention[J]. Micromachines(Basel), 2022, 13(11): 1920.
[38] Latke V, Narawade VJI, Computing V.Detection of dental periapical lesions using retinex based image enhancement and lightweight deep learning model[J]. Image Vis Comput, 2024, 146:105016.
[39] Xu X, Chen J, Yin J. Tooth instance segmentation and disease detection with uncertainty-aware contrastive learning and cross-scale attention[J]. IEEE J Biomed Health Inform, 2025 Jan 2.
[Online ahead of print].
[40] Ma T, Qian JL, Qin X, et al.Lightweight dental lesion segmentation algorithm based on transformer and CNN[C]//2024 9th International Conference on Image and Vision Computing (ICIVC). Nanjing: IEEE, 2024: 183-187.
[41] Kondori I, Mottin RW, Laskin DM.Accuracy of dentists in the clinical diagnosis of oral lesions[J]. Quintessence Int, 2011, 42(7): 575-577.
[42] Jubair F, Al-Karadsheh O, Malamos D, et al.A novel lightweight deep convolutional neural network for early detection of oral cancer[J]. Oral Dis, 2022, 28(4): 1123-1130.
[43] Swamikannan LD, Sonawane AB, Patel JS, et al.Oral cancer detection using mobile vision technology[C]//2024 IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI). Houston: IEEE, 2024: 1-8.
[44] Devindi GAI, Dissanayake D, Liyanage SN, et al.Multimodal deep convolutional neural network pipeline for ai-assisted early detection of oral cancer[J]. IEEE Access, 2024, 12: 124375-124390.
[45] Song B, Sunny S, Li S, et al.Mobile-based oral cancer classification for point-of-care screening[J]. J Biomed Opt, 2021, 26(6): 065003.
[46] Goswami B, Bhuyan MK, Alfarhood S, et al.Classification of oral cancer into pre-cancerous stages from white light images using LightGBM algorithm[J]. IEEE Access, 2024, 12: 31626-31639.
[47] Yadav DP, Sharma B, Noonia A, et al.Explainable label guided lightweight network with axial transformer encoder for early detection of oral cancer[J]. Sci Rep, 2025, 15(1): 6391.
[48] Dwivedi K, Chugh B, Srivastava A, et al.TransEns-Network: an optimized light-weight transformer and feature fusion based approach of deep learning models for the classification of oral cancer[J]. Int J Comput Modelling Appl, 2024, 1(1): 32-44.
[49] Marzouk R, Alabdulkreem E, Dhahbi S, et al.Deep transfer learning driven oral cancer detection and classification model[J]. Comput Mater Continua, 2022, 73(2):3905-3920.
[50] Talwar V, Singh P, Mukhia N, et al.AI-assisted screening of oral potentially malignant disorders using smartphone-based photographic images[J]. Cancers(Basel), 2023, 15(16): 4120.
[51] Shabir A, ahmed KT, Mujahid M, et al. LWFDTL: lightweight fusion deep transfer learning for oral squamous cell carcinoma diagnosis using histopathological oral mucosa[J]. Multimed Tools Appl, 2025, 84: 30359-30383.
[52] Anitha VP, Harini R, Meenatchi R.Pioneering early diagnosis of oral squamous cell carcinoma: utilizing Mobilenetv3, DBN and PSo-BER optimization[C]//2025 6th International Conference on Mobile Computing and Sustainable Informatics (ICMCSI).Kathmandu: IEEE, 2025: 557-563.
[53] Kalli S, Kalaimani G, Sivaprakash S, et al.Compact deep learning architecture for oral cancer classification[C]//2024 Asian Conference on Intelligent Information and Technology (ACOIT). Beijing: IEEE, 2024: 1-6.
[54] Khuntia S, Fan SY, Juan PH, et al.Empowering portable optoelectronics with computer vision for intra-oral cavities detection[J]. IEEE Sensors J, 2024, 24(16): 25911-25919.
[55] Fahim M, Sharma V, Duong TQ.A wearable-based preventive model to promote oral health through personalized notification[C]//2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). Glasgow: IEEE, 2022: 4282-4285.
[56] Hu Z, Radmehr A, Zhang Y, et al.IOteeth: intra-oral teeth sensing system for dental occlusal diseases recognition[J]. Proc ACM Interact Mob Wearable Ubiquitous Technol, 2024, 8(1): 1-29.