-
-
High level Design + User flow
-
CAMERA
-
SMS ALERT
-
COMPUTER VISION DETECTION
Problem Statement
Epilepsy is a chronic neurological disorder that affects about 10 million people in Africa, of which 40% are children aged 5-151. These children face significant barriers in accessing timely and effective assistance during seizures, due to the limited availability and affordability of trained caregivers, specialized equipment, and quality healthcare services in the region. This results in increased risks of injury, disability, death, and social stigma for these children, as well as higher healthcare expenditures and productivity losses for their families and communities. To address this critical gap, we propose to develop a low-cost, user-friendly, and context-appropriate device that can monitor the body activity of epileptic children, and alert their caregivers and health providers in case of a seizure.
What it does
The envisioned application seeks to tackle this issue by offering a real-time monitoring and alert system. It utilizes a camera and cloud-based machine learning algorithms to identify potential signs of Seizure. Upon detection, the app promptly notifies designated emergency contacts, including healthcare professionals, providing them with information about the user's condition and collected personal data. We are optimistic that this app holds the potential to transform the monitoring and protection of vulnerable individuals, creating a safer and more secure environment within designated institutions.
How we built it
Prior to development, we made Figma and a High-level user flow which enable to see to core functionality we need to implement within the 4 days. Then used Flutter to build the mobile App to register emergency contact. For the cloud-based machine learning algorithms, Our approach leverages Python's Computer Vision, Open CV to capture patient movement and detect the different angles of the patient’s body based on landmarks. To finish we used a programmable communication tool for customized messaging. Enables bulk messaging and automated calling.
Challenges we ran into
One of the most significant challenges has been seamlessly integrating various technologies, including live streaming and machine learning algorithms, ensuring their cohesive functionality.
Accomplishments we are proud of
We are greatly proud of successfully detecting human images in different positions. While Python is not a foreign language, it was a first for us to use the OpenCV library, and successfully integrating it into our project is a great accomplishment. Successfully integrating Twilio API and customizing our messaging for the target caregivers Developing a potentially beneficial health-tech product within 3 days.
What we learnt
Epilepsy and the great challenge epileptic patients have across the continent. The significance of computer vision and image processing in enhancing medical access The gaps existing in African health and the potential for improvement through medical innovations.
What is next?
Exploring IOT as an enhancement for our prototype. In this case, portable wristbands that can alleviate any financial concerns from our target audience. Developing a potential financial model for Safetycall that would make it cost effective in the villages and rural areas in Africa where there are a lot of unattended epilepsy patients.
Log in or sign up for Devpost to join the conversation.