updated demo

app.py

@@ -105,6 +105,10 @@ def process_input(image_path, prompt_names: list, disease_name: str, descriptors
 
     return output
 
+with open("article.md", "r") as f:
+    article = f.read()
+with open("description.md", "r") as f:
+    description = f.read()
 
 # Define the Gradio interface
 iface = gr.Interface(
@@ -118,14 +122,16 @@ iface = gr.Interface(
                  'Custom'],
         default=['Enlarged Cardiomediastinum', 'Cardiomegaly', 'Lung Opacity', 'Lung Lesion', 'Edema', 'Consolidation', 'Pneumonia',
                  'Atelectasis', 'Pneumothorax', 'Pleural Effusion', 'Pleural Other', 'Fracture', 'Support Devices'],
-        label='Selct to use predefined disease descriptors. Select "Custom" to define your own observations.'),
+        label='Select to use predefined disease descriptors. Select "Custom" to define your own observations.'),
             gr.inputs.Textbox(lines=2, placeholder="Name of pathology for which you want to define custom observations", label='Pathology:'),
             gr.inputs.Textbox(lines=2, placeholder="Add your custom (positive) observations separated by a new line"
                                                    "\n Note: Each descriptor will automatically be embedded into our prompt format: There is/are (no) <observation> indicating <pathology>"
                                                    "\n Example:\n\n Opacity\nPleural Effusion\nConsolidation"
                               , label='Custom Observations:')],
+    article=article,
+    description=description,
     outputs=gr.outputs.Image(type="filepath")
 )
 
 # Launch the interface
-iface.launch()
+iface.launch()

article.md

@@ -0,0 +1,31 @@
+
+We propose a new way of explainability for zero-shot diagnosis prediction in the clinical domain. Instead of directly predicting a diagnosis, we prompt the model to classify the existence of descriptive observations, which a radiologist would look for on an X-Ray scan, and use the descriptor probabilities to estimate the likelihood of a diagnosis, making our model explainable by design. For this we leverage BioVil, a pretrained CLIP model for X-rays and apply contrastive observation-based prompting. We evaluate Xplainer on two chest X-ray
+datasets, CheXpert and ChestX-ray14, and demonstrate its effectiveness
+in improving the performance and explainability of zero-shot diagnosis.
+**Authors**: [Chantal Pellegrini][cp], [Matthias Keicher][mk], [Ege Özsoy][eo], [Petra Jiraskova][pj], [Rickmer Braren][rb], [Nassir Navab][nn]
+
+[cp]:https://www.cs.cit.tum.de/camp/members/chantal-pellegrini/
+[eo]:https://www.cs.cit.tum.de/camp/members/ege-oezsoy/
+[mk]:https://www.cs.cit.tum.de/camp/members/matthias-keicher/
+[pj]:https://campus.tum.de/tumonline/ee/ui/ca2/app/desktop/#/pl/ui/$ctx/visitenkarte.show_vcard?$ctx=design=ca2;header=max;lang=de&pPersonenGruppe=3&pPersonenId=46F3A857F258DEE6
+[rb]:https://radiologie.mri.tum.de/de/person/prof-dr-rickmer-f-braren
+[nn]:https://www.cs.cit.tum.de/camp/members/cv-nassir-navab/nassir-navab/
+
+**License**: MIT
+
+**Where to send questions or comments about the model**: Open an issue on [`Xplainer`](https://github.com/ChantalMP/Xplainer) repo.
+
+**Intended Use**: This model is intended to be used solely for (I) future research on visual-language processing and (II) reproducibility of the experimental results reported in the reference paper.
+
+**Primary intended uses/users**: Vision-Language and CAD researchers 
+
+
+## Citation
+```bib
+@article{pellegrini2023xplainer,
+  title={Xplainer: From X-Ray Observations to Explainable Zero-Shot Diagnosis},
+  author={Pellegrini, Chantal and Keicher, Matthias and {\"O}zsoy, Ege and Jiraskova, Petra and Braren, Rickmer and Navab, Nassir},
+  journal={arXiv preprint arXiv:2303.13391},
+  year={2023}
+}
+```

description.md

@@ -0,0 +1,3 @@
+This demo provides a playground for the testing the model of our paper "Xplainer: From X-Ray Observations to Explainable Zero-Shot Diagnosis", which was accepted for publication at MICCAI 2023. You can test our pre-defined prompts and define your own prompts and diseases.
+
+**Paper**: [arxiv](https://arxiv.org/pdf/2303.13391.pdf), **Code**: [Github](https://github.com/ChantalMP/Xplainer)