HEALTH MONITORING SYSTEM :
04/06/2025 Faculty of Engineering and Technology 1
Problem Statement :
1. Elderly individuals and patients with chronic illnesses struggle to
track daily health changes, leading to missed early symptoms.
2. Doctors reply on patient memory for diagnosis, which often
results in incomplete or inaccurate medical assessments.
3. Traditional healthcare methods lack real-time monitoring,
making early detection of critical health issues difficult.
2
ABSTRACT:
1. The AI-Powered Vital Detection Device is designed to provide
continuous health monitoring for elderly individuals and patients
with chronic conditions.
2. This device uses AI to track vital signs such as sneezing
patterns, coughing frequency, heart rate, blood pressure, and
breathing irregularities in real time.
3
INTRODUCTION:
1. Problem in Traditional Healthcare – Doctors rely on patient
memory and periodic check-ups, leading to incomplete or delayed
diagnoses.
2. Need for Continuous Monitoring – Many health conditions
show early symptoms that go unnoticed without real-time tracking.
3. How AI Helps – AI analyze health patterns, detects abnormalities
early, and provides structured past reports for better diagnosis.
4. Device Functionality – It continuously tracks vital signs (heart
rate, blood pressure, sneezing, coughing, breathing patterns) and
generates detailed reports.
5. Impact on Doctors – Simplifies their work by offering real-time
insights and past data trends, reducing dependency on patient
memory.
4
Justification and Background Research :
Justification Background
Research
• Doctors need past
health data for • Traditional check-ups
accurate diagnosis. miss real-time
• AI tracks vitals and symptoms.
provides structured • Wearable devices lack
reports. AI-based analysis.
• Continuous monitoring • AI can detect health
reduces emergency risks from non-invasive
cases. data.
• Structured reports • Our device enhances
improve diagnosis monitoring with
efficiency. predictive reports.
5
OBJECTIVES OF THE PROJECT :
1. Assist Doctors – Provide structured past
health reports, making diagnosis easier and
more accurate.
2. Continuous Health Monitoring – Track
vital signs in real-time to detect early
symptoms of diseases.
3. improve Patient Care – Enable proactive
treatment by identifying health risks before
they become severe.
4. AI-Driven Analysis – Use machine
learning to analyze health trends and
detect abnormalities efficiently.
6
LITERATURE REVIEW :
• Importance of Continuous Monitoring
• Existing AI-Based Health Monitoring
• Gaps in Traditional Diagnosis
• Gaps in Current Research (Many AI-driven medical tools either
focus on specific diseases)
How Our Project Fills the Gap – By integrating AI with continuous,
non-invasive monitoring and structured reporting, our device aims to
provide doctors with reliable health data for better decision-making.
7
Convolutional Neural Network
(CNN) :
• How CNN Works in Our Project?
1️⃣Audio Input: Captures sneezing, coughing, and breathing
sounds.
2️⃣Spectrogram Conversion: Converts sound signals into
images.
3️⃣Feature Extraction: CNN identifies key health-related
patterns.
4️⃣Classification: Determines if the sound is Normal or
Abnormal for early diagnosis.
• Why CNN?
✅ High accuracy in detecting health conditions.
✅ Works automatically without manual feature selection.
✅ Enables real-time health monitoring
8
METHODOLOGY:
9
HARDWARE AND SOFTWARE SPECIFICATIONS :
Best Hardware Specifications:
• Sensors: MAX30102 (Heart & SpO2), BMP280 (BP), MEMS Mic
(Cough/Sneeze),Piezoelectric(Breathing).
• Processor & Connectivity: NVIDIA Jetson Nano / ESP32, Wi-Fi
(ESP8266),Bluetooth(BLE5.0).
• Power & Display: Rechargeable Li-ion Battery, OLED (1.3”) for
real-time stats.
Best Software Specifications:
Tech Stack: Python (AI), C++ (Embedded), JavaScript (Frontend).
AI & Data: TensorFlow Lite, PyTorch, Firebase (real-time), AWS (cloud
backup).
Security & UI: AES-256 encryption, OAuth 2.0, Flutter (mobile),
ReactJS (web).
10