OncoTUM models

<h1>OncoTUM models</h1>
<p>
This repository hosts pretrained neural network models for <a href="https://github.com/masud-src/OncoTUM" target="_blank">OncoTUM</a>,
a key software package within the umbrella project <strong>Onco*</strong> for modelling and numerical simulations of tumours. OncoTUM is designed to facilitate tumour segmentations from medical images, leveraging state-of-the-art deep learning techniques.
</p>

<h2>Purpose</h2>
<p>
The pretrained models in this repository are required for using the <strong>inference function</strong> of OncoTUM. These models have been trained on relevant datasets (<a href="https://www.kaggle.com/datasets/awsaf49/brats20-dataset-training-validation" target="_blank">BraTS 2020</a>) to ensure
high accuracy and performance in tumor segmentation and its classification.
</p>

<h2>Usage</h2>
<p>
To utilise the inference functionality of OncoTUM, download the appropriate pretrained models from this repository and ensure they are correctly linked to the OncoTUM software package.
Detailed instructions for setup and integration can be found in the
<a href="https://github.com/masud-src/OncoTUM" target="_blank">OncoTUM documentation</a>.
</p>

<h2>Content</h2>
<p> In order to remain with most possible flexibility, the modality agnostic implementation allows to perform segmentation with a subset of the gold standard modalities.
</p>
<p> <strong>Brain tumour segmentation</strong></p>
<ul>
<li><strong>Full modal model:</strong> trained to all gold standard modalities (t1, t1gd, t2, flair).</li>
<li><strong>Single modality model:</strong> trained to single modalities of the gold standard.</li>
<li><strong>Null image:</strong> Empty image for training.</li>
</ul>