File size: 3,698 Bytes
eb3d3f0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 |
# Face Detection-Based AI Automation of Lab Tests
# Streamlit App with OpenCV + rPPG + MediaPipe Integration (Deployable on Hugging Face Spaces)
import streamlit as st
import cv2
import numpy as np
import mediapipe as mp
import pandas as pd
import time
import os
# Setup Mediapipe Face Mesh
mp_face_mesh = mp.solutions.face_mesh
face_mesh = mp_face_mesh.FaceMesh(static_image_mode=False, max_num_faces=1, refine_landmarks=True, min_detection_confidence=0.5)
# Function to calculate mean green intensity (simplified rPPG)
def estimate_heart_rate(frame, landmarks):
h, w, _ = frame.shape
forehead_pts = [landmarks[10], landmarks[338], landmarks[297], landmarks[332]]
mask = np.zeros((h, w), dtype=np.uint8)
pts = np.array([[int(pt.x * w), int(pt.y * h)] for pt in forehead_pts], np.int32)
cv2.fillConvexPoly(mask, pts, 255)
green_channel = cv2.split(frame)[1]
mean_intensity = cv2.mean(green_channel, mask=mask)[0]
heart_rate = int(60 + 30 * np.sin(mean_intensity / 255.0 * np.pi)) # Simulated
return heart_rate
# Estimate SpO2 and Respiratory Rate (dummy based on heart rate)
def estimate_spo2_rr(heart_rate):
spo2 = min(100, max(90, 97 + (heart_rate % 5 - 2)))
rr = int(12 + abs(heart_rate % 5 - 2))
return spo2, rr
# Streamlit UI setup
st.set_page_config(page_title="Face-Based Lab Test Automation", layout="wide")
st.title("🧠 Face Detection-Based AI Automation of Lab Tests")
col1, col2 = st.columns([1, 2])
# Left: Webcam and Face Scan
with col1:
st.header("📷 Scan Face")
run = st.checkbox("Start Camera")
FRAME_WINDOW = st.image([])
camera = cv2.VideoCapture(0)
results = {}
while run:
ret, frame = camera.read()
if not ret:
break
frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
result = face_mesh.process(frame_rgb)
if result.multi_face_landmarks:
landmarks = result.multi_face_landmarks[0].landmark
heart_rate = estimate_heart_rate(frame_rgb, landmarks)
spo2, rr = estimate_spo2_rr(heart_rate)
results = {
"Hemoglobin": "12.3 g/dL (Estimated)",
"SpO2": f"{spo2}%",
"Heart Rate": f"{heart_rate} bpm",
"Blood Pressure": "Low",
"Respiratory Rate": f"{rr} breaths/min",
"Risk Flags": ["Anemia Mild", "Hydration Low"]
}
FRAME_WINDOW.image(frame_rgb)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
camera.release()
# Right: Health Report
with col2:
st.header("🧪 AI-Based Diagnostic Report")
if results:
with st.expander("Hematology & Blood Tests", expanded=True):
st.metric("Hemoglobin", results["Hemoglobin"], "Anemia Mild")
with st.expander("Vital Signs and Biochemical Tests", expanded=True):
st.metric("SpO2", results["SpO2"])
st.metric("Heart Rate", results["Heart Rate"])
st.metric("Blood Pressure", results["Blood Pressure"], "Low")
st.metric("Respiratory Rate", results["Respiratory Rate"], "Hydration Low")
with st.expander("Risk Flags"):
for flag in results["Risk Flags"]:
st.error(flag)
# Export Button
if st.button("📥 Export Report as CSV"):
df = pd.DataFrame([results])
df.to_csv("lab_scan_report.csv", index=False)
st.success("Report saved as lab_scan_report.csv")
else:
st.info("No face scan detected yet.")
# Footer
st.markdown("---")
st.caption("© 2025 FaceLab AI by Sathkrutha Tech Solutions. Built with Streamlit, OpenCV, MediaPipe, and rPPG techniques.")
|