AI POWERED MODEL FOR
ORAL CANCER
DETECTION
Multi Disiciplinary Major Project Presentation 2024 - 2025
AI MODEL FOR ORAL CANCER DETECTION
Project Guide:
Dr. Geetha Kiran A Professor Dr. Mohana Lakshmi J
and Dean (corporate affairs) Associate Professor
Department of CSE, MCE Department of EEE, MCE
Hassan Hassan
Team Members
Name Department
Muskan S - 4MC21CS099 Computer Science & Engg
Harshini ND - 4MC21CS059 Computer Science & Engg
Pavan Kumar PK - 4MC21EE046 Electrical & Electronics Engg
Prajwal M D - 4MC21EE049 Electrical & Electronics Engg
INTRODUCTION
Early detection of oral cancer is crucial, as it greatly
improves treatment options and patient outcomes.
This project aims to develop a robotic arm capable of
predicting the presence of oral cancer using advanced
imaging and machine learning techniques.
The robotic arm is equipped with a high-resolution camera for real-time
image capture, controlled by a Raspberry Pi 4 microcontroller running
Python scripts. This setup enables seamless communication between the
hardware and software components, allowing for efficient image
processing using OpenCV and model inference with TensorFlow or
PyTorch, ensuring accurate detection of oral cancer.
Problem Statement
To Develop a image capturing and machine
learning model that detects early signs of oral
cancer from medical images using advanced
image processing and AI techniques, aiding
medical professionals in accurate diagnosis and
improving patient outcomes.
Literature Review
Software based research papers:
Hardware based research papers:
Requirements Specification
Hardware requirments:
Raspberry pi 4 Microcontroller: Central control unit for managing device operations.
Robotic Arm: Equipped with servo motors for precise movement during dental
examinations.
Wired Camera or High-Resolution Camera: For real-time visualization of the oral cavity
and capturing high-quality images.
Power Supply: To ensure efficient functioning of all components.
Sensors and Actuators: For precise movement and control of the robotic arm.
Communication Modules: IoT components for wireless communication with mobile
devices or laptops.
Methodology:
Software Requirements
Operating System
Linux-based OS (e.g., Raspbian for Raspberry Pi) for compatibility and ease of use.
Machine Learning Framework:
TF Lite Micro is one of the most popular frameworks for machine learning on edge devices
Image Processing Library:
OpenCV for real-time image processing and analysis from the camera feed.
Communication Protocols:
MQTT or HTTP for IoT communication between the robotic arm and mobile devices/laptops.
Data Storage:
SQLite or a cloud-based database for storing captured images and patient data securely.
Control Software:
Python scripts to manage the operation of the robotic arm, image capture, and
model inference.
User Interface:
Mobile app or web application for user interaction, displaying results, and controlling the
robotic arm remotely.
Error Handling and Logging
Implement logging for debugging and maintaining a record of system operations and errors
for reliability and safety.
THANK YOU
