A CT-based transfer learning approach to predict NSCLC recurrence: The added-value of peritumoral region

Abstract

Non-small cell lung cancer (NSCLC) represents 85% of all new lung cancer diagnoses and presents a high recurrence rate after surgery. Thus, an accurate prediction of recurrence risk in NSCLC patients at diagnosis could be essential to designate risk patients to more aggressive medical treatments. In this manuscript, we apply a transfer learning approach to predict recurrence in NSCLC patients, exploiting only data acquired during its screening phase. Particularly, we used a public radiogenomic dataset of NSCLC patients having a primary tumor CT image and clinical information. Starting from the CT slice containing the tumor with maximum area, we considered three different dilatation sizes to identify three Regions of Interest (ROIs): CROP (without dilation), CROP 10 and CROP 20. Then, from each ROI, we extracted radiomic features by means of different pre-trained CNNs. The latter have been combined with clinical information; thus, we trained a Support Vector Machine classifier to predict the NSCLC recurrence. The classification performances of the devised models were finally evaluated on both the hold-out training and hold-out test sets, in which the original sample has been previously divided. The experimental results showed that the model obtained analyzing CROP 20 images, which are the ROIs containing more peritumoral area, achieved the best performances on both the hold-out training set, with an AUC of 0.73, an Accuracy of 0.61, a Sensitivity of 0.63, and a Specificity of 0.60, and on the hold-out test set, with an AUC value of 0.83, an Accuracy value of 0.79, a Sensitivity value of 0.80, and a Specificity value of 0.78. The proposed model represents a promising procedure for early predicting recurrence risk in NSCLC patients.

Fig 1. ROI extraction process.

After identifying the tumor segmentation with maximum area, along with the corresponding CT image, three ROIs were extracted for each patient: CROP, CROP 10 and CROP 20.

Fig 2. Schematic overview of the proposed approach.

(a) After extracting radiomic features by means of three different pre-trained CNNs from each identified crop, (b) we performed a feature selection procedure within a 10-fold cross-validation scheme over 5 rounds on the hold-out training set. (c) Then, we trained a SVM classifier on the hold-out training set exploiting both clinical data and radiomic features extracted in the previous step. Finally, we performed an external validation on the hold-out test set.