Commit 22ad3efb authored by Alessia Marcolini's avatar Alessia Marcolini
Browse files

Add pretrained weights

parent d346320a
# Multimodal learning for prognosis prediction in oncology by combined deep and radiomics features
**Authors**: Andrea Bizzego, Nicole Bussola, Alessia Marcolini, Valerio Maggio, Marco Chierici, Damiana Salvalai, Giuseppe Jurman, Cesare Furlanello
This repository contains the source code and trained models for the RADLER pipeline for deep and radiomics features integration for the [Head-Neck-PET-CT](https://wiki.cancerimagingarchive.net/display/Public/Head-Neck-PET-CT) multimodal dataset.
Repository attached to the paper "Multimodal learning for prognosis prediction in oncology by combined deep and radiomics features".
In this study, we introduce a machine learning framework for medical imaging that combines the current pattern recognition approach ("radiomics") with Deep Learning (DL). We apply the framework in cancer bioimaging for prognosis of locoregional recurrence in head and neck squamous cell carcinoma (N=298) from Computed Tomography (CT) and Positron Emission Tomography (PET) imaging. The DL architecture is composed of two parallel cascades of Convolutional Neural Network (CNN) layers merging in a softmax classification layer. The network is first pretrained on head and neck tumor stage diagnosis, then fine-tuned on the prognostic task by internal transfer learning. In parallel, radiomics features (e.g., shape of the tumor mass, texture and pixels intensity statistics) are derived by predefined feature extractors on the PET-CT pairs. We first compare and then mix deep learning and radiomics features into a unifying classification pipeline (RADLER), where model selection and evaluation are based on a data analysis plan developed in the MAQC initiative for reproducible biomarkers. On the multimodal PET-CT cancer dataset, the mixed deep learning/radiomics approach is more accurate than using only one feature type, or image mode. Further, RADLER significantly improves over published results on the same data.
## Pretrained models
We share the weights of the models, as described in the manuscript. See folder [pretrained_weights](pretrained_weights).
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment