Deep learning for predicting epidermal growth factor receptor mutations of non-small cell lung cancer on PET/CT images

Abstract

Background: Predicting the mutation status of the epidermal growth factor receptor (EGFR) gene based on an integrated positron emission tomography/computed tomography (PET/CT) image of non-small cell lung cancer (NSCLC) is a noninvasive, low-cost method which is valuable for targeted therapy. Although deep learning has been very successful in robotic vision, it is still challenging to predict gene mutations in PET/CT-derived studies because of the small amount of medical data and the different parameters of PET/ CT devices.Methods: We used the advanced EfficientNet-V2 model to predict the EGFR mutation based on fused PET/CT images. First, we extracted 3-dimensional (3D) pulmonary nodules from PET and CT as regions of interest (ROIs). We then fused each single PET and CT image. The network model was used to predict the mutation status of lung nodules by the new data after fusion, and the model was weighted adaptively. The EfficientNet-V2 model used multiple channels to represent nodules comprehensively. Results:We trained the EfficientNet-V2 model through our PET/CT fusion algorithm using a dataset of 150 patients. The prediction accuracy of EGFR and non-EGFR mutations was 86.25% in the training dataset, and the accuracy rate was 81.92% in the validation set. Conclusions:Combined with experiments, the demonstrated PET/CT fusion algorithm outperformed radiomics methods in predicting EGFR and non-EGFR mutations in NSCLC.^ ORCID: 0000-0003-0618-6240.

Deep-Learning-Based Predictive Imaging Biomarker Model for EGFR Mutation Status in Non-Small Cell Lung Cancer from CT Imaging

“…Other similar studies have shown a similar utility of DL models with improvements in the AUC when combined with clinical parameters [ 25 , 26 , 27 , 28 ]. Further, a study on the PET/CT fusion algorithm using a dataset of 150 patients showed a prediction accuracy of EGFR and non-EGFR mutations of 86.25% in the training dataset and 81.92% in the validation set [ 29 ].…”

Prediction of oncogene mutation status in non-small cell lung cancer: A systematic review and meta-analysis with a special focus on artificial-intelligence-based methods

“…Thus, while in radiomics analysis a process of lesion segmentation and subsequent feature extraction is required, which introduces certain degree of variability and can be a high time-consuming task, DL models only required a bounding box of the lesion, greatly reducing this effect. On the other hand, DL models, and in particular end-to-end convolutional neural network (CNN) models, such those developed in most of the DL studies included in our work 37, 42, 43, 50, 58, 68, 85, 86, 90, 91 , are generally more complex in terms of the number of parameters, allowing to solve more complicated problems than traditional ML models. Considering available evidence, it seems reasonable to think that methodologic approaches should be carefully revised when validation studies are designed and conducted.…”

New research progress on 18F-FDG PET/CT radiomics for EGFR mutation prediction in lung adenocarcinoma: a review

“…Several investigators are now focusing on applying DL models in predicting EGFR mutation status, which has shown promising performance. Xiao et al (63) proposed a deep learning framework based on the EfficientNet-V2 model. First, 32 2D views are extracted from each 3D cube of lung nodules.…”

“…Xiao et al. ( 63 ) proposed a deep learning framework based on the EfficientNet-V2 model. First, 32 2D views are extracted from each 3D cube of lung nodules.…”