import gradio as gr  

title="Thyroid Tumor Classification On Ultrasound Images"
article = "This study was made by S.Serdar Helli using HF Transformers ConvNext"

description=f'''

Thyroid nodule is one of the most common endocrine carcinomas. Due to its higher reveal ability and ability to distinguish between benign and malignant nodules in pathological features, ultrasonography has become the most widely used modality for finding and diagnosing thyroid cancer when compared to CT and MRI.

In this study, the purpose is the classification of thyroid tumors on ultrasound images with 2 different categories:

- Malign(1)
- Benign(0)

SubClasses

- 1	Normal thyroid glandle	
- 2	Effectively certainly benign	Simple cyst	- % 0 Risk of malignancy (Benign)
- 3	Very probably benigped	- %0.25 Risk of malignancy (Benign)
- 4A	Suspicious nodules; low risk of malignancy	- % 6 Risk of malignancy (Malign)
- 4B	Suspicious nodules; high risk of malignancy	One or two features of high suspicion	- %69 Risk of malignancy (Malign)
- 5	Effectively certainly malignant nodules	- % 100 Risk of malignancy (Malign)

This study was made using HF Transformers :

- [ On Google Colab](https://colab.research.google.com/drive/1ueSq8Y_NmFr7NGdtS8FStI3d2HR-43LD?usp=sharing)

- [On Github](https://github.com/SerdarHelli/The-Classification-of-Thyroid-Tumors-on-UltraSound-Images-using-Deep-Learning-Methods)

- [ Using Keras and GradCam With MultiClasses  Medium Article](https://serdarhelli.medium.com/the-basic-classification-of-thyroid-tumors-on-ultrasound-images-using-deep-learning-methods-46f812d859ea)

The Dataset:
[Colombia National University presented an open access database of thyroid ultrasound images.](http://cimalab.unal.edu.co/?lang=es&mod=program&id=5)
Ref : Pedraza, Lina & Vargas, Carlos & Narváez, Fabián & Durán, Oscar & Muñoz, Emma & Romero, Eduardo. (2015). An open access thyroid ultrasound-image Database. Progress in Biomedical Optics and Imaging — Proceedings of SPIE. 9287. 10.1117/12.2073532. 

'''



examples=["example_Benign1.png" ,"example_Benign2.png", "example_Malign2.png"]
gr.Interface.load("huggingface/SerdarHelli/ThyroidTumorClassificationModel",
                  title=title,description=description,article=article,examples=examples, cache_examples=False).launch()