Upload 3 files

app.py

@@ -0,0 +1,101 @@
+import streamlit as st
+import tensorflow as tf
+from PIL import Image
+import numpy as np
+from tensorflow.keras.applications.efficientnet import EfficientNetB7 as PretrainedModel, preprocess_input
+from tensorflow.keras.layers import GlobalAveragePooling2D, Flatten, Dense
+from tensorflow.keras.models import Model
+from skimage.filters import threshold_otsu
+import cv2
+import time
+
+X = Y = 224
+import os
+os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
+
+def preprocess_image(uploaded_image, predicted_label):
+    # Read image bytes from uploaded file
+    file_bytes = np.asarray(bytearray(uploaded_image.read()), dtype=np.uint8)
+    img = cv2.imdecode(file_bytes, cv2.IMREAD_COLOR)
+    
+    if img is None:
+        raise ValueError("Failed to load the uploaded image.")
+
+    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+    img_gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
+    thresh = threshold_otsu(img_gray)
+    img_otsu = img_gray < thresh
+    total_area = img_otsu.size
+    black_area = np.count_nonzero(img_otsu == 0)
+    white_area = np.count_nonzero(img_otsu == 1)
+
+    if predicted_label != 'lung_n':
+        if white_area >= 300000:
+            level = 3
+        elif 200000 <= white_area < 3000000:
+            level = 2
+        elif 100000 <= white_area < 200000:
+            level = 1
+        elif white_area < 100000:
+            level = 0
+
+        if level == 3:
+            st.error("Cancer type: " + predicted_label)
+            st.error("Level of Cancer: 3")
+        elif level == 2:
+            st.warning("Cancer type: " + predicted_label)
+            st.warning("Level of Cancer: 2")
+        elif level == 1:
+            st.info("Cancer type: " + predicted_label)
+            st.info("Level of Cancer: 1")
+        elif level == 0:
+            st.success("Cancer type: " + predicted_label)
+            st.success("Level of Cancer: 0")
+    else:
+        st.success("Predicted as Lung with Benign Tumor")
+
+
+# Create a Streamlit application
+def main():
+    st.title('Lung Cancer Detection & Severity Level using Deep Learning')
+    input_shape = (X, Y, 3)
+    K = 3
+    start_time = time.time()
+    ptm = PretrainedModel(
+        input_shape=(X, Y, 3),
+        weights='imagenet',
+        include_top=False
+    )
+    ptm.trainable = False
+    x = GlobalAveragePooling2D()(ptm.output)
+    x = Flatten()(x)
+    x = Dense(128, activation='relu')(x)
+    x = Dense(64, activation='relu')(x)
+    y = Dense(K, activation='softmax')(x)
+    model = Model(inputs=ptm.input, outputs=y)
+    model.load_weights('lungs_weights.h5')
+    # model.save('lungs_model.h5')
+
+    label_map = {0: 'lung_Adenocarcinoma', 1: 'lung_benign', 2: 'lung_squamous cell carcinoma'}
+
+    uploaded_image = st.file_uploader('Upload an image')
+
+    if uploaded_image is not None:
+        image = Image.open(uploaded_image)
+        image = image.resize((224, 224))
+        image_array = np.array(image)
+        image_array = preprocess_input(image_array)
+        image_array = np.expand_dims(image_array, axis=0)
+        predictions = model.predict(image_array)
+        predicted_class = np.argmax(predictions)
+        predicted_label = label_map[predicted_class]
+        preprocess_image(uploaded_image, predicted_label)
+        end_time = time.time()
+        elapsed_time = end_time - start_time
+        st.write('Time taken to predict is approximately:', round(elapsed_time, 2), 'seconds')
+        st.image(image, caption='Uploaded Image', width=300)
+    else:
+        st.write('Please upload an image')
+
+if __name__ == '__main__':
+    main()

lungs_weights.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db1a9bd087cd0aac54215314a5b8d3516eefbd63b0be2b62cf377b20d7771eb4
+size 259503352

requirements.txt

@@ -0,0 +1,8 @@
+numpy
+Pillow
+keras
+tensorflow
+streamlit
+opencv-python
+scikit-image
+