Research progress of artificial intelligence in the image recognition of cystic lesions of the jawbone
WANG Jianhao
SUN Rui
LONG Shuqin
SUN Yuanyuan
Abstract：As one of the common diseases in the oral and maxillofacial region,cystic lesions of the jaw are characterized by strong insidiousness,a tendency to cause local bone destruction,and functional impairment.Although traditional imaging methods such as panoramic radiographs and cone-beam CT can provide information on the morphology and range of the lesion,they have limitations in the differential diagnosis of different lesion types,especially for lesions with similar imaging features such as odontogenic keratocysts and ameloblastomas,the misdiagnosis rate is relatively high.By introducing artificial intelligence(AI)technology into the imaging diagnosis of jaw cystic lesions in recent years,current studies focus on three main directions:object detection,image classification,and image segmentation.The YOLO(You Only Look Once)series models have demonstrated high accuracy in lesion detection and early screening.Convolutional neural networks(CNNs)and their variants have achieved levels of performance in lesion classification that are close to or even better than those of manual diagnosis.Meanwhile,U-Net and its derivative models have shown significant advantages in lesion segmentation and quantification of lesion extent.Additionally,emerging architectures such as visual Transformers offer new ideas for further enhancing diagnostic performance.
Keywords：jaw cystsartificial intelligencedental radiographyYOLO modelU-Net
Publication Date：2025-11-20
Online Publishing Date：2025-12-02(First online date of this platform, not the publication date of the document)
Pages：5( 465-469 )