CrossViT with ECAP: Enhanced deep learning for jaw lesion classification.

Journal: International journal of medical informatics
PMID: 39492085

Abstract

BACKGROUND: Radiolucent jaw lesions like ameloblastoma (AM), dentigerous cyst (DC), odontogenic keratocyst (OKC), and radicular cyst (RC) often share similar characteristics, making diagnosis challenging. In 2021, CrossViT, a novel deep learning approach using multi-scale vision transformers (ViT) with cross-attention, emerged for accurate image classification. Additionally, we introduced Extended Cropping and Padding (ECAP), a method to expand training data by iteratively cropping smaller images while preserving context. However, its application in dental radiographic classification remains unexplored. This study investigates the effectiveness of CrossViTs and ECAP against ResNets for classifying common radiolucent jaw lesions.

Authors

  • Wannakamon Panyarak
    Division of Oral and Maxillofacial Radiology, Department of Oral Biology and Diagnostic Sciences, Faculty of Dentistry, Chiang Mai University, Chiang Mai, Thailand. Electronic address: wannakamon.p@cmu.ac.th.
  • Wattanapong Suttapak
    Division of Computer Engineering, School of Information and Communication Technology, University of Phayao, Phayao, Thailand.
  • Phattaranant Mahasantipiya
    Department of Oral Biology and Diagnostic Sciences, Faculty of Dentistry, Chiang Mai University, Chiang Mai, Thailand.
  • Arnon Charuakkra
    Division of Oral and Maxillofacial Radiology, Department of Oral Biology and Diagnostic Sciences, Faculty of Dentistry, Chiang Mai University, Chiang Mai, Thailand.
  • Nattanit Boonsong
    Department of Oral Biology and Diagnostic Sciences, Faculty of Dentistry, Chiang Mai University, Suthep Road, Suthep Sub-district, Mueang Chiang Mai District, Chiang Mai 50200, Thailand. Electronic address: nattanit.b@cmu.ac.th.
  • Kittichai Wantanajittikul
    Department of Radiologic Technology, Faculty of Associated Medical Sciences, Chiang Mai University, Chiang Mai, Thailand.
  • Anak Iamaroon
    Department of Oral Biology and Diagnostic Sciences, Faculty of Dentistry, Chiang Mai University, Suthep Road, Suthep Sub-district, Mueang Chiang Mai District, Chiang Mai 50200, Thailand. Electronic address: anak.ia@cmu.ac.th.

Keywords

External Resources

Popular Topics
Recent Journals