Upload 15 files

app.py

@@ -0,0 +1,60 @@
+import gradio as gr
+import numpy as np
+from tensorflow.keras.preprocessing import image
+import joblib
+from tensorflow.keras.models import load_model
+from PIL import Image
+import os
+path = 'test/'
+
+trained_meta_model = joblib.load('meta_model.pkl')
+densenet_model = load_model('densenet_model.h5')
+effnet_model = load_model('effnet_model.h5')
+base_models = [densenet_model, effnet_model]
+
+class_names = ['Cataract', 'DR', 'Glaucoma', 'Normal']
+
+def preprocess_image(img):
+    if isinstance(img, np.ndarray):
+        img = Image.fromarray(img)  # Convert NumPy array to PIL Image
+    img = img.resize((224, 224))  # Resize the image to the required size
+    img_array = image.img_to_array(img)  # Convert image to array
+    img_array = np.expand_dims(img_array, axis=0)  # Add batch dimension
+    img_array = img_array / 255.0  # Normalize the image if needed
+    return img_array
+
+def get_single_image_meta_features(models, img_array):
+    meta_features = []
+    for model in models:
+        preds = model.predict(img_array)
+        meta_features.append(np.argmax(preds, axis=1))
+    return np.array(meta_features).T
+
+def classify_image(img, patient_name, patient_age, patient_gender):
+    img_array = preprocess_image(img)
+    single_image_meta_features = get_single_image_meta_features(base_models, img_array)
+    final_prediction = trained_meta_model.predict(single_image_meta_features)
+    p = int(final_prediction[0])
+    result = f"Patient Name: {patient_name}\nPatient Age: {patient_age}\nPatient Gender: {patient_gender}\nPredicted Class: {class_names[p]}"
+    return result
+
+# Dummy prediction to warm up the models
+dummy_img = np.zeros((224, 224, 3), dtype=np.uint8)
+classify_image(dummy_img, "Dummy", 0, "None")
+
+# Define the inputs
+image_input = gr.inputs.Image(shape=(224, 224), image_mode='RGB')
+name_input = gr.inputs.Textbox(label="Patient Name")
+age_input = gr.inputs.Number(label="Patient Age")
+gender_input = gr.inputs.Radio(choices=["Male", "Female", "Other"], label="Patient Gender")
+
+# Define the output
+output_text = gr.outputs.Textbox(label="Result")
+
+iface = gr.Interface(fn=classify_image, 
+                     inputs=[image_input, name_input, age_input, gender_input], 
+                     outputs=output_text, 
+                     title="Eye Disease Classification",
+                     description="Upload eye images, provide patient information, and get the predicted disease class.")  # Update with actual example paths if needed
+
+iface.launch(share=True)

densenet_model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd2fbc97abc596b32315dea9f692500c1eb465e841f55ccde69e11608d7a9af1
+size 72787216

effnet_model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a2ff10d8db181341faf53be20acf87c9e5c12962440d266338fc0905b535efd
+size 148835264

flagged/log.csv

@@ -0,0 +1,5 @@
+'img','Patient Name','Patient Age','Patient Gender','Result','flag','username','timestamp'
+'img/0.jpg','Hmal',20,'Male','Patient Name: Hmal
+Patient Age: 20.0
+Patient Gender: Male
+Predicted Class: DR','','','2024-06-24 17:51:42.587173'

meta_model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7027b0ec9ce16d50f1870004f11da0ebeee7d1e504789dd8fd8d5cc59a1b9e5d
+size 224505