Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,98 @@
|
|
1 |
-
---
|
2 |
-
license: apache-2.0
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
datasets:
|
4 |
+
- SilpaCS/Augmented_alzheimer
|
5 |
+
language:
|
6 |
+
- en
|
7 |
+
base_model:
|
8 |
+
- google/siglip2-base-patch16-224
|
9 |
+
pipeline_tag: image-classification
|
10 |
+
library_name: transformers
|
11 |
+
tags:
|
12 |
+
- Alzheimer
|
13 |
+
- Stage-Classifier
|
14 |
+
- SigLIP2
|
15 |
+
---
|
16 |
+
|
17 |
+
|
18 |
+
# **Alzheimer-Stage-Classifier**
|
19 |
+
|
20 |
+
> **Alzheimer-Stage-Classifier** is a multi-class image classification model based on `google/siglip2-base-patch16-224`, designed to identify stages of Alzheimer’s disease from medical imaging data. This tool can assist in **clinical decision support**, **early diagnosis**, and **disease progression tracking**.
|
21 |
+
|
22 |
+
---
|
23 |
+
|
24 |
+
## **Label Classes**
|
25 |
+
|
26 |
+
The model classifies input images into the following stages of Alzheimer’s disease:
|
27 |
+
|
28 |
+
```
|
29 |
+
0: MildDemented
|
30 |
+
1: ModerateDemented
|
31 |
+
2: NonDemented
|
32 |
+
3: VeryMildDemented
|
33 |
+
```
|
34 |
+
|
35 |
+
---
|
36 |
+
|
37 |
+
## **Installation**
|
38 |
+
|
39 |
+
```bash
|
40 |
+
pip install transformers torch pillow gradio
|
41 |
+
```
|
42 |
+
|
43 |
+
---
|
44 |
+
|
45 |
+
## **Example Inference Code**
|
46 |
+
|
47 |
+
```python
|
48 |
+
import gradio as gr
|
49 |
+
from transformers import AutoImageProcessor, SiglipForImageClassification
|
50 |
+
from PIL import Image
|
51 |
+
import torch
|
52 |
+
|
53 |
+
# Load model and processor
|
54 |
+
model_name = "prithivMLmods/Alzheimer-Stage-Classifier"
|
55 |
+
model = SiglipForImageClassification.from_pretrained(model_name)
|
56 |
+
processor = AutoImageProcessor.from_pretrained(model_name)
|
57 |
+
|
58 |
+
# ID to label mapping
|
59 |
+
id2label = {
|
60 |
+
"0": "MildDemented",
|
61 |
+
"1": "ModerateDemented",
|
62 |
+
"2": "NonDemented",
|
63 |
+
"3": "VeryMildDemented"
|
64 |
+
}
|
65 |
+
|
66 |
+
def classify_alzheimer_stage(image):
|
67 |
+
image = Image.fromarray(image).convert("RGB")
|
68 |
+
inputs = processor(images=image, return_tensors="pt")
|
69 |
+
|
70 |
+
with torch.no_grad():
|
71 |
+
outputs = model(**inputs)
|
72 |
+
logits = outputs.logits
|
73 |
+
probs = torch.nn.functional.softmax(logits, dim=1).squeeze().tolist()
|
74 |
+
|
75 |
+
prediction = {id2label[str(i)]: round(probs[i], 3) for i in range(len(probs))}
|
76 |
+
return prediction
|
77 |
+
|
78 |
+
# Gradio Interface
|
79 |
+
iface = gr.Interface(
|
80 |
+
fn=classify_alzheimer_stage,
|
81 |
+
inputs=gr.Image(type="numpy"),
|
82 |
+
outputs=gr.Label(num_top_classes=4, label="Alzheimer Stage"),
|
83 |
+
title="Alzheimer-Stage-Classifier",
|
84 |
+
description="Upload a brain scan image to classify the stage of Alzheimer's: NonDemented, VeryMildDemented, MildDemented, or ModerateDemented."
|
85 |
+
)
|
86 |
+
|
87 |
+
if __name__ == "__main__":
|
88 |
+
iface.launch()
|
89 |
+
```
|
90 |
+
|
91 |
+
---
|
92 |
+
|
93 |
+
## **Applications**
|
94 |
+
|
95 |
+
* **Early Alzheimer’s Screening**
|
96 |
+
* **Clinical Diagnosis Support**
|
97 |
+
* **Longitudinal Study & Disease Monitoring**
|
98 |
+
* **Research on Cognitive Decline**
|