"Foundation model for cancer imaging biomarkers"
"Foundation models are generally trained using self-supervised learning and excel in reducing the demand for training samples in downstream applications. This is especially important in medicine, where large labelled datasets are often scarce. Here, we developed a foundation model for cancer imaging biomarker discovery by training a convolutional encoder through self-supervised learning using a comprehensive dataset of 11,467 radiographic lesions. The foundation model was evaluated in distinct and clinically relevant applications of cancer imaging-based biomarkers."
"We compared simple auto-encoder pretraining and several stateof-the-art self-supervised pretraining approaches—namely SimCLR5 , SwAV15 and NNCLR16—against the modified version of SimCLR developed in our study (Methods). We evaluated pretraining strategies on the technical validation use case of lesion anatomical site classification by comparing linear classifiers trained on top of features extracted from each of the chosen strategies. We observed that our modified SimCLR pretraining surpassed all others (P?<?0.001) in balanced accuracy (Fig. 2a) and mean average precision (mAP) (Fig. 2b), achieving a balanced accuracy of 0.779 (95% confidence interval (CI) 0.750–0.810) and mAP?=?0.847 (95% CI 0.750–0.810)."