Using Vision Transformer for high robustness and generalization in predicting EGFR mutation status in lung adenocarcinoma

• Luoqi Weng ^[1] ; Yuhan Chen ^[1] ; Chengshui Chen ^[1] ; Yilun Xu ^[2] ; Qinqing Qian ^[3] ; Jie Pan ^[4] ; Huang Su ^[4]
  1. [1] Department of Pulmonary and Critical Care Medicine, The First Afliated Hospital of Wenzhou Medical University, Wenzhou 325000, Zhejiang, China
  2. [2] Department of Gastroenterology, The First Afliated Hospital of Wenzhou Medical University, Wenzhou 325000, Zhejiang, China
  3. [3] Department of Respiratory Medicine, Shaoxing People’s Hospital, Shaoxing 312000, Zhejiang, China
  4. [4] Department of Gastroenterology, Wenzhou Central Hospital, Wenzhou 325000, Zhejiang, China

  Mostrar afiliaciones +
• Localización: Clinical & translational oncology, ISSN 1699-048X, Vol. 26, Nº. 6, 2024, págs. 1438-1445
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Background Lung adenocarcinoma is a common cause of cancer-related deaths worldwide, and accurate EGFR genotyping is crucial for optimal treatment outcomes. Conventional methods for identifying the EGFR genotype have several limitations. Therefore, we proposed a deep learning model using non-invasive CT images to predict EGFR mutation status with robustness and generalizability.

    Methods A total of 525 patients were enrolled at the local hospital to serve as the internal data set for model training and validation. In addition, a cohort of 30 patients from the publicly available Cancer Imaging Archive Data Set was selected for external testing. All patients underwent plain chest CT, and their EGFR mutation status labels were categorized as either mutant or wild type. The CT images were analyzed using a self-attention-based ViT-B/16 model to predict the EGFR mutation status, and the model’s performance was evaluated. To produce an attention map indicating the suspicious locations of EGFR mutations, Grad-CAM was utilized.

    Results The ViT deep learning model achieved impressive results, with an accuracy of 0.848, an AUC of 0.868, a sensitivity of 0.924, and a specifcity of 0.718 on the validation cohort. Furthermore, in the external test cohort, the model achieved comparable performances, with an accuracy of 0.833, an AUC of 0.885, a sensitivity of 0.900, and a specifcity of 0.800.

    Conclusions The ViT model demonstrates a high level of accuracy in predicting the EGFR mutation status of lung adenocarcinoma patients. Moreover, with the aid of attention maps, the model can assist clinicians in making informed clinical decisions.