Deep Learning Based HPV Status Prediction for Oropharyngeal Cancer Patients
This work addresses the problem of HPV status prediction for oropharyngeal cancer patients, offering an incremental improvement in diagnostic accuracy.
This paper explores deep learning models for predicting HPV status in oropharyngeal cancer patients using CT images. A 3D convolutional network pre-trained on sports video clips achieved an AUC of 0.81 on an external test set, outperforming models trained from scratch or pre-trained on ImageNet.
We investigated the ability of deep learning models for imaging based HPV status detection. To overcome the problem of small medical datasets we used a transfer learning approach. A 3D convolutional network pre-trained on sports video clips was fine tuned such that full 3D information in the CT images could be exploited. The video pre-trained model was able to differentiate HPV-positive from HPV-negative cases with an area under the receiver operating characteristic curve (AUC) of 0.81 for an external test set. In comparison to a 3D convolutional neural network (CNN) trained from scratch and a 2D architecture pre-trained on ImageNet the video pre-trained model performed best.