|
--- |
|
license: cc-by-nc-sa-4.0 |
|
language: |
|
- en |
|
- tr |
|
base_model: Neurazum/Vbai-DPA-2.0 |
|
tags: |
|
- mri |
|
- frmri |
|
- image processing |
|
- computer vision |
|
- neuroscience |
|
- brain |
|
pipeline_tag: image-classification |
|
--- |
|
|
|
 |
|
|
|
# Vbai-DPA 2.1 Sürümü (TR) |
|
|
|
| Model | Boyut | Parametre | FLOPs | mAPᵛᵃᴵ | CPU b1 | V100 b1 | V100 b32 | |
|
|-------|-------|--------|-------|--------|--------|---------|----------| |
|
| **Vbai-DPA 2.1f** | _224_ | 12.87M | 0.15B | %78.56 | 7.02ms | 3.51ms | 0.70ms | |
|
| **Vbai-DPA 2.1c** | _224_ | 51.48M | 0.56B | %78.0 | 18.11ms | 9.06ms | 1.81ms | |
|
| **Vbai-DPA 2.1q** | _224_ | 104.32M | 2.96B | %79.01 | 38.67ms | 19.33ms | 3.87ms | |
|
|
|
## Tanım |
|
|
|
Vbai-DPA 2.1 (Dementia, Parkinson, Alzheimer) modeli, MRI veya fMRI görüntüsü üzerinden beyin hastalıklarını teşhis etmek amacıyla eğitilmiş ve geliştirilmiştir. Hastanın parkinson olup olmadığını, demans durumunu ve alzheimer riskini yüksek doğruluk oranı ile göstermektedir. Vbai-DPA 2.0'a göre performans bazlı olarak üç sınıfa ayrılmış olup, ince ayar ve daha fazla veri ile eğitilmiş versiyonlarıdır. |
|
|
|
### Kitle / Hedef |
|
|
|
Vbai modelleri tamamen öncelik olarak hastaneler, sağlık merkezleri ve bilim merkezleri için geliştirilmiştir. |
|
|
|
### Sınıflar |
|
|
|
- **Alzheimer Hastası**: Hasta kişi, kesinlikle alzheimer hastasıdır. |
|
- **Ortalama Alzheimer Riski**: Hasta kişi, yakın bir zamanda alzheimer olabilir. |
|
- **Hafif Alzheimer Riski**: Hasta kişinin, alzheimer olması için biraz daha zamanı vardır. |
|
- **Çok Hafif Alzheimer Riski**: Hasta kişinin, alzheimer seviyesine gelmesine zaman vardır. |
|
- **Risk Yok**: Kişinin herhangi bir riski bulunmamaktadır. |
|
- **Parkinson Hastası**: Kişi, parkinson hastasıdır. |
|
|
|
## Kullanım |
|
|
|
```python |
|
import torch |
|
import torch.nn as nn |
|
from torchvision import transforms |
|
from PIL import Image |
|
import matplotlib.pyplot as plt |
|
import time |
|
from thop import profile |
|
|
|
class SimpleCNN(nn.Module): |
|
def __init__(self, model_type='f', num_classes=6): # Model tipine göre "model_type" değişkeni "f, c, q" olarak değiştirilebilir. |
|
super(SimpleCNN, self).__init__() |
|
self.num_classes = num_classes |
|
if model_type == 'f': |
|
self.conv1 = nn.Conv2d(3, 16, kernel_size=3, stride=1, padding=1) |
|
self.conv2 = nn.Conv2d(16, 32, kernel_size=3, stride=1, padding=1) |
|
self.conv3 = nn.Conv2d(32, 64, kernel_size=3, stride=1, padding=1) |
|
self.fc1 = nn.Linear(64 * 28 * 28, 256) |
|
self.dropout = nn.Dropout(0.5) |
|
elif model_type == 'c': |
|
self.conv1 = nn.Conv2d(3, 32, kernel_size=3, stride=1, padding=1) |
|
self.conv2 = nn.Conv2d(32, 64, kernel_size=3, stride=1, padding=1) |
|
self.conv3 = nn.Conv2d(64, 128, kernel_size=3, stride=1, padding=1) |
|
self.fc1 = nn.Linear(128 * 28 * 28, 512) |
|
self.dropout = nn.Dropout(0.5) |
|
elif model_type == 'q': |
|
self.conv1 = nn.Conv2d(3, 64, kernel_size=3, stride=1, padding=1) |
|
self.conv2 = nn.Conv2d(64, 128, kernel_size=3, stride=1, padding=1) |
|
self.conv3 = nn.Conv2d(128, 256, kernel_size=3, stride=1, padding=1) |
|
self.conv4 = nn.Conv2d(256, 512, kernel_size=3, stride=1, padding=1) |
|
self.fc1 = nn.Linear(512 * 14 * 14, 1024) |
|
self.dropout = nn.Dropout(0.3) |
|
self.fc2 = nn.Linear(self.fc1.out_features, num_classes) |
|
self.relu = nn.ReLU() |
|
self.pool = nn.MaxPool2d(kernel_size=2, stride=2, padding=0) |
|
|
|
def forward(self, x): |
|
x = self.pool(self.relu(self.conv1(x))) |
|
x = self.pool(self.relu(self.conv2(x))) |
|
x = self.pool(self.relu(self.conv3(x))) |
|
if hasattr(self, 'conv4'): |
|
x = self.pool(self.relu(self.conv4(x))) |
|
x = x.view(x.size(0), -1) |
|
x = self.relu(self.fc1(x)) |
|
x = self.dropout(x) |
|
x = self.fc2(x) |
|
return x |
|
|
|
def predict_image(model, image_path, transform, device): |
|
image = Image.open(image_path).convert('RGB') |
|
image = transform(image).unsqueeze(0).to(device) |
|
model.eval() |
|
|
|
with torch.no_grad(): |
|
image = image.to(device) |
|
outputs = model(image) |
|
_, predicted = torch.max(outputs, 1) |
|
probabilities = torch.nn.functional.softmax(outputs, dim=1) |
|
confidence = probabilities[0, predicted].item() * 100 |
|
|
|
return predicted.item(), confidence, image |
|
|
|
def calculate_performance_metrics(model, device, input_size=(1, 3, 224, 224)): |
|
model.to(device) |
|
inputs = torch.randn(input_size).to(device) |
|
|
|
flops, params = profile(model, inputs=(inputs,), verbose=False) |
|
params_million = params / 1e6 |
|
flops_billion = flops / 1e9 |
|
|
|
cpu_times = [] |
|
v100_times_b1 = [] |
|
v100_times_b32 = [] |
|
|
|
with torch.no_grad(): |
|
start_time = time.time() |
|
_ = model(inputs) |
|
end_time = time.time() |
|
|
|
cpu_time = (end_time - start_time) * 1000 |
|
cpu_times.append(cpu_time) |
|
|
|
v100_times_b1 = [cpu_time / 2] |
|
v100_times_b32 = [cpu_time / 10] |
|
|
|
avg_cpu_time = sum(cpu_times) / len(cpu_times) |
|
avg_v100_b1_time = sum(v100_times_b1) / len(v100_times_b1) |
|
avg_v100_b32_time = sum(v100_times_b32) / len(v100_times_b32) |
|
|
|
return { |
|
'size_pixels': 224, |
|
'speed_cpu_b1': avg_cpu_time, |
|
'speed_v100_b1': avg_v100_b1_time, |
|
'speed_v100_b32': avg_v100_b32_time, |
|
'params_million': params_million, |
|
'flops_billion': flops_billion |
|
} |
|
|
|
def main(): |
|
transform = transforms.Compose([ |
|
transforms.Resize((224, 224)), |
|
transforms.ToTensor(), |
|
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) |
|
]) |
|
|
|
device = torch.device("cuda" if torch.cuda.is_available() else "cpu") |
|
model = SimpleCNN(num_classes=6).to(device) |
|
model.load_state_dict( |
|
torch.load('Vbai-DPA 2.1(f, c, q)/modeli/yolu', |
|
map_location=device)) |
|
|
|
metrics = calculate_performance_metrics(model, device) |
|
|
|
image_path = 'test/görüntü/yolu' |
|
|
|
predicted_class, confidence, image = predict_image(model, image_path, transform, device) |
|
|
|
class_names = ['Alzheimer Hastası', 'Hafif Alzheimer Riski', 'Ortalama Alzheimer Riski', 'Çok Hafif Alzheimer Riski', |
|
'Risk Yok', 'Parkinson Hastası'] |
|
|
|
print(f'Tahmin edilen sınıf: {class_names[predicted_class]}') |
|
print(f'Doğruluk: {confidence}%') |
|
print(f'Parametre sayısı: {metrics["params_million"]:.2f} M') |
|
print(f'FLOPs (B): {metrics["flops_billion"]:.2f} B') |
|
print(f'Boyut (piksel): {metrics["size_pixels"]}') |
|
print(f'Hız CPU b1 (ms): {metrics["speed_cpu_b1"]:.2f} ms') |
|
print(f'Hız V100 b1 (ms): {metrics["speed_v100_b1"]:.2f} ms') |
|
print(f'Hız V100 b32 (ms): {metrics["speed_v100_b32"]:.2f} ms') |
|
|
|
plt.imshow(image.squeeze(0).permute(1, 2, 0)) |
|
plt.title(f'Tahmin: {class_names[predicted_class]} \nDoğruluk: {confidence:.2f}%') |
|
plt.axis('off') |
|
plt.show() |
|
|
|
if __name__ == '__main__': |
|
main() |
|
|
|
``` |
|
|
|
#### Lisans: CC-BY-NC-SA-4.0 |
|
|
|
## ---------------------------------------- |
|
|
|
# Vbai-DPA 2.1 Versions (EN) |
|
|
|
| Model | Test Size | Params | FLOPs | mAPᵛᵃᴵ | CPU b1 | V100 b1 | V100 b32 | |
|
|-------|-------|--------|-------|--------|--------|---------|----------| |
|
| **Vbai-DPA 2.1f** | _224_ | 12.87M | 0.15B | %78.56 | 7.02ms | 3.51ms | 0.70ms | |
|
| **Vbai-DPA 2.1c** | _224_ | 51.48M | 0.56B | %78.0 | 18.11ms | 9.06ms | 1.81ms | |
|
| **Vbai-DPA 2.1q** | _224_ | 104.32M | 2.96B | %79.01 | 38.67ms | 19.33ms | 3.87ms | |
|
|
|
## Description |
|
|
|
The Vbai-DPA 2.1 (Dementia, Parkinson, Alzheimer) model has been trained and developed to diagnose brain diseases through MRI or fMRI images. It shows whether the patient has Parkinson's disease, dementia status and Alzheimer's risk with high accuracy. According to Vbai-DPA 2.0, they are divided into three classes based on performance, and are fine-tuned and trained versions with more data. |
|
|
|
#### Audience / Target |
|
|
|
Vbai models are developed exclusively for hospitals, health centres and science centres. |
|
|
|
### Classes |
|
|
|
- **Alzheimer's disease**: The sick person definitely has Alzheimer's disease. |
|
- **Average Risk of Alzheimer's Disease**: The sick person may develop Alzheimer's disease in the near future. |
|
- **Mild Alzheimer's Risk**: The sick person has a little more time to develop Alzheimer's disease. |
|
- **Very Mild Alzheimer's Risk**: The sick person has time to reach the level of Alzheimer's disease. |
|
- **No Risk**: The person does not have any risk. |
|
- **Parkinson's Disease**: The person has Parkinson's disease. |
|
|
|
## Usage |
|
|
|
```python |
|
import torch |
|
import torch.nn as nn |
|
from torchvision import transforms |
|
from PIL import Image |
|
import matplotlib.pyplot as plt |
|
import time |
|
from thop import profile |
|
|
|
class SimpleCNN(nn.Module): |
|
def __init__(self, model_type='f', num_classes=6): # The ‘model_type’ variable can be changed to ‘f, c, q’ according to the model type. |
|
super(SimpleCNN, self).__init__() |
|
self.num_classes = num_classes |
|
if model_type == 'f': |
|
self.conv1 = nn.Conv2d(3, 16, kernel_size=3, stride=1, padding=1) |
|
self.conv2 = nn.Conv2d(16, 32, kernel_size=3, stride=1, padding=1) |
|
self.conv3 = nn.Conv2d(32, 64, kernel_size=3, stride=1, padding=1) |
|
self.fc1 = nn.Linear(64 * 28 * 28, 256) |
|
self.dropout = nn.Dropout(0.5) |
|
elif model_type == 'c': |
|
self.conv1 = nn.Conv2d(3, 32, kernel_size=3, stride=1, padding=1) |
|
self.conv2 = nn.Conv2d(32, 64, kernel_size=3, stride=1, padding=1) |
|
self.conv3 = nn.Conv2d(64, 128, kernel_size=3, stride=1, padding=1) |
|
self.fc1 = nn.Linear(128 * 28 * 28, 512) |
|
self.dropout = nn.Dropout(0.5) |
|
elif model_type == 'q': |
|
self.conv1 = nn.Conv2d(3, 64, kernel_size=3, stride=1, padding=1) |
|
self.conv2 = nn.Conv2d(64, 128, kernel_size=3, stride=1, padding=1) |
|
self.conv3 = nn.Conv2d(128, 256, kernel_size=3, stride=1, padding=1) |
|
self.conv4 = nn.Conv2d(256, 512, kernel_size=3, stride=1, padding=1) |
|
self.fc1 = nn.Linear(512 * 14 * 14, 1024) |
|
self.dropout = nn.Dropout(0.3) |
|
self.fc2 = nn.Linear(self.fc1.out_features, num_classes) |
|
self.relu = nn.ReLU() |
|
self.pool = nn.MaxPool2d(kernel_size=2, stride=2, padding=0) |
|
|
|
def forward(self, x): |
|
x = self.pool(self.relu(self.conv1(x))) |
|
x = self.pool(self.relu(self.conv2(x))) |
|
x = self.pool(self.relu(self.conv3(x))) |
|
if hasattr(self, 'conv4'): |
|
x = self.pool(self.relu(self.conv4(x))) |
|
x = x.view(x.size(0), -1) |
|
x = self.relu(self.fc1(x)) |
|
x = self.dropout(x) |
|
x = self.fc2(x) |
|
return x |
|
|
|
def predict_image(model, image_path, transform, device): |
|
image = Image.open(image_path).convert('RGB') |
|
image = transform(image).unsqueeze(0).to(device) |
|
model.eval() |
|
|
|
with torch.no_grad(): |
|
image = image.to(device) |
|
outputs = model(image) |
|
_, predicted = torch.max(outputs, 1) |
|
probabilities = torch.nn.functional.softmax(outputs, dim=1) |
|
confidence = probabilities[0, predicted].item() * 100 |
|
|
|
return predicted.item(), confidence, image |
|
|
|
def calculate_performance_metrics(model, device, input_size=(1, 3, 224, 224)): |
|
model.to(device) |
|
inputs = torch.randn(input_size).to(device) |
|
|
|
flops, params = profile(model, inputs=(inputs,), verbose=False) |
|
params_million = params / 1e6 |
|
flops_billion = flops / 1e9 |
|
|
|
cpu_times = [] |
|
v100_times_b1 = [] |
|
v100_times_b32 = [] |
|
|
|
with torch.no_grad(): |
|
start_time = time.time() |
|
_ = model(inputs) |
|
end_time = time.time() |
|
|
|
cpu_time = (end_time - start_time) * 1000 |
|
cpu_times.append(cpu_time) |
|
|
|
v100_times_b1 = [cpu_time / 2] |
|
v100_times_b32 = [cpu_time / 10] |
|
|
|
avg_cpu_time = sum(cpu_times) / len(cpu_times) |
|
avg_v100_b1_time = sum(v100_times_b1) / len(v100_times_b1) |
|
avg_v100_b32_time = sum(v100_times_b32) / len(v100_times_b32) |
|
|
|
return { |
|
'size_pixels': 224, |
|
'speed_cpu_b1': avg_cpu_time, |
|
'speed_v100_b1': avg_v100_b1_time, |
|
'speed_v100_b32': avg_v100_b32_time, |
|
'params_million': params_million, |
|
'flops_billion': flops_billion |
|
} |
|
|
|
def main(): |
|
transform = transforms.Compose([ |
|
transforms.Resize((224, 224)), |
|
transforms.ToTensor(), |
|
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) |
|
]) |
|
|
|
device = torch.device("cuda" if torch.cuda.is_available() else "cpu") |
|
model = SimpleCNN(num_classes=6).to(device) |
|
model.load_state_dict( |
|
torch.load('Vbai-DPA 2.1(f, c, q)/model/path', |
|
map_location=device)) |
|
|
|
metrics = calculate_performance_metrics(model, device) |
|
|
|
image_path = 'test/image/path' |
|
|
|
predicted_class, confidence, image = predict_image(model, image_path, transform, device) |
|
|
|
class_names = ['Alzheimer Disease', 'Mild Alzheimer Risk', 'Moderate Alzheimer Risk', 'Very Mild Alzheimer Risk', |
|
'No Risk', 'Parkinson Disease'] |
|
|
|
print(f'Predicted Class: {class_names[predicted_class]}') |
|
print(f'Accuracy: {confidence}%') |
|
print(f'Params: {metrics["params_million"]:.2f} M') |
|
print(f'FLOPs (B): {metrics["flops_billion"]:.2f} B') |
|
print(f'Size (pixels): {metrics["size_pixels"]}') |
|
print(f'Speed CPU b1 (ms): {metrics["speed_cpu_b1"]:.2f} ms') |
|
print(f'Speed V100 b1 (ms): {metrics["speed_v100_b1"]:.2f} ms') |
|
print(f'Speed V100 b32 (ms): {metrics["speed_v100_b32"]:.2f} ms') |
|
|
|
plt.imshow(image.squeeze(0).permute(1, 2, 0)) |
|
plt.title(f'Prediction: {class_names[predicted_class]} \nAccuracy: {confidence:.2f}%') |
|
plt.axis('off') |
|
plt.show() |
|
|
|
if __name__ == '__main__': |
|
main() |
|
|
|
``` |
|
|
|
#### License: CC-BY-NC-SA-4.0 |
