import os
import re
import cv2
import time
import bcrypt
import numpy as np
import streamlit as st
import tensorflow as tf
from pymongo import MongoClient
def load_data():
client = MongoClient(os.environ.get("MONGODB_URL"))
db = client['medical-app-auth']
collection = db['login-data']
brain_tumor_model = tf.keras.models.load_model('models/brain_tumor.h5')
alzheimer_model = tf.keras.models.load_model('models/alzheimer.h5')
return collection, brain_tumor_model, alzheimer_model
collection, brain_tumor_model, alzheimer_model = load_data()
if 'current_page' not in st.session_state:
st.session_state.current_user = None
st.session_state.current_page = 'login'
def clear_cache():
keys = list(st.session_state.keys())
for key in keys:
st.session_state.pop(key)
def login():
st.set_page_config(layout='centered', page_title="Brain MRI", page_icon="mri_of_brain.jpg")
col1, col2 = st.columns([5,1])
with col2:
if st.button("Register", use_container_width=True):
st.session_state.current_page = 'register'
st.rerun()
def reset_passowrd_input():
st.session_state.password = ""
def reset_username_inputs():
st.session_state.username = ""
reset_passowrd_input()
with st.form(key='login', clear_on_submit=True):
st.subheader("Login")
username = st.text_input("Username", placeholder="Enter Username")
password = st.text_input("Enter Password", type="password")
username = username.lower().strip()
submit = st.form_submit_button("Login")
if submit:
with st.spinner('Checking credentials...'):
user = collection.find_one({"username":username})
if user==None:
st.warning("Username is does not exits, please register")
reset_username_inputs()
elif bcrypt.checkpw(password.encode('utf-8'), user.get("password")):
st.warning("**Credential Matched**: Redirecting...")
st.session_state.current_user = username
st.session_state.current_page = 'medical'
st.rerun()
else:
st.warning("Password is incorrect")
reset_passowrd_input()
def register():
st.set_page_config(layout='centered', page_title="Brain MRI", page_icon="mri_of_brain.jpg")
col1, col2 = st.columns([5,1])
with col2:
if st.button("Login", use_container_width=True):
st.session_state.current_page = 'login'
st.rerun()
document = {}
pattern = r'^[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}$'
def reset_password_inputs():
st.session_state.password1 = ""
st.session_state.password2 = ""
def reset_username_inputs():
st.session_state.username = ""
reset_password_inputs()
def reset_email_inputs():
st.session_state.email = ""
reset_username_inputs()
with st.form(key='register', clear_on_submit=True):
st.subheader("Register")
email = st.text_input("Email", placeholder="enter@your.email", max_chars=64)
username = st.text_input("Username", placeholder="Enter Username", max_chars=17)
name = st.text_input("Name", placeholder="Enter Your Full Name")
password1 = st.text_input("Enter Password", type="password", max_chars=17)
password2 = st.text_input("Confirm Password", type="password", max_chars=17)
email = email.lower().strip()
username = username.lower().strip()
name = name.title().strip()
submit = st.form_submit_button("Register")
if submit:
with st.spinner('Checking credentials...'):
if re.match(pattern, email)==None:
st.warning(":red[Please provide a valid email.]", icon="⚠️")
reset_email_inputs()
elif collection.find_one({"email":email}):
st.warning("Email already exits, please Login!")
reset_email_inputs()
elif not name.replace(" ", "").isalpha():
st.warning("Don't use number or special characters for name")
st.session_state.name = ""
elif len(username)<5:
st.warning("Username must atleast be of 5 characters")
reset_username_inputs()
elif len(password1)<6:
st.warning("Password must atleast be of 6 characters")
reset_password_inputs()
elif collection.find_one({"username": username}):
st.warning("Username already exits, please try a different one.")
reset_username_inputs()
elif password1 != password2:
st.warning(":red[Passwords do not match. Please try again.]", icon="⚠️")
reset_password_inputs()
elif not (email and name and username and password1 and password2):
st.warning(":red[Please complete all the fields above.]", icon="⚠️")
else:
salt = bcrypt.gensalt(rounds=13)
hashed_password = bcrypt.hashpw(password1.encode('utf-8'), salt)
salt = bcrypt.gensalt(rounds=13)
hashed_password = bcrypt.hashpw(password1.encode('utf-8'), salt)
document = {"name":name, "username":username, "email":email, "password":hashed_password, "salt":salt}
collection.insert_one(document)
st.warning("**Successfully Registered**: Redirecting...")
st.session_state.current_user = username
st.session_state.current_page = 'medical'
st.rerun()
def medical_page():
def alzheimer():
col1, col2, col3 = st.columns([6, 6, 1])
with col1:
if st.button(f"Welcome! {st.session_state.current_user}"):
st.rerun()
with col3:
st.button("Logout", use_container_width=True, on_click=clear_cache)
st.markdown("***")
st.subheader("Here's your Alzheimer's Scan")
uploaded_file = st.file_uploader("Upload MRI scan image for detecting alzheimer's", type=['png', 'jpg'])
if uploaded_file is not None:
file_bytes = np.asarray(bytearray(uploaded_file.read()), dtype=np.uint8)
opencv_image = cv2.imdecode(file_bytes, cv2.IMREAD_COLOR)
resized_img = cv2.resize(opencv_image, (208, 176))
resized_img = resized_img[np.newaxis, ...]
class_names = ['ModerateDemented', 'NonDemented', 'VeryMildDemented', 'MildDemented']
pred = class_names[np.argmax(alzheimer_model.predict(resized_img))]
st.text(pred)
def brain_tumor():
col1, col2, col3 = st.columns([6, 6, 1])
with col1:
if st.button(f"Welcome! {st.session_state.current_user}"):
st.rerun()
with col3:
st.button("Logout", use_container_width=True, on_click=clear_cache)
st.markdown("***")
st.subheader("Here's your Brain Tumor Scan")
uploaded_file = st.file_uploader("Upload MRI scan image for detecting Brain Tumor", type=['png', 'jpg'])
if uploaded_file is not None:
file_bytes = np.asarray(bytearray(uploaded_file.read()), dtype=np.uint8)
opencv_image = cv2.imdecode(file_bytes, cv2.IMREAD_COLOR)
resized_img = cv2.resize(opencv_image, (168, 150))
resized_img = resized_img[np.newaxis, ...]
class_names = ['Pituitary', 'No-Tumor', 'Meningioma', 'Glioma']
pred = class_names[np.argmax(brain_tumor_model.predict(resized_img))]
st.text(pred)
def abscesses():
col1, col2, col3 = st.columns([6, 6, 1])
with col1:
if st.button(f"Welcome! {st.session_state.current_user}"):
st.rerun()
with col3:
st.button("Logout", use_container_width=True, on_click=clear_cache)
st.markdown("***")
st.subheader("Here's your Abscesses Scan")
uploaded_file = st.file_uploader("Upload MRI scan image for detecting Abscesses", type=['png', 'jpg'], disabled=True)
st.write("Feature currently unavailable")
if uploaded_file is not None:
file_bytes = np.asarray(bytearray(uploaded_file.read()), dtype=np.uint8)
opencv_image = cv2.imdecode(file_bytes, cv2.IMREAD_COLOR)
resized_img = cv2.resize(opencv_image, (208, 176))
resized_img = resized_img[np.newaxis, ...]
class_names = ['ModerateDemented', 'NonDemented', 'VeryMildDemented', 'MildDemented']
pred = class_names[np.argmax(abscesses_model.predict(resized_img))]
st.text(pred)
st.set_page_config(layout='wide', page_title="Brain MRI", page_icon="mri_of_brain.jpg")
st.sidebar.image("mri_of_brain.jpg")
st.sidebar.title("Navigation")
page_options = ["Alzheimer", "Brain Tumor", "Abscesses"]
selected_page = st.sidebar.selectbox("Select a Scan", page_options)
if selected_page == "Alzheimer":
alzheimer()
elif selected_page == "Brain Tumor":
brain_tumor()
elif selected_page == "Abscesses":
abscesses()
def main():
if st.session_state.current_page == 'login':
login()
elif st.session_state.current_page == 'register':
register()
elif st.session_state.current_page == 'medical':
medical_page()
if __name__ == "__main__":
main()