What it does
MediGuard AI is an autonomous emergency response robot that delivers life-saving medical intervention within 90 seconds, solving the critical gap where 4.6 million people die annually from emergencies during the 8 to 12-minute wait for ambulances.
Your inspiration
In 2019, my uncle died from a heart attack in our Malaysian village while waiting 45 minutes for an ambulance. Holding him as he died, I realised the tragedy was not his condition — it was that simple interventions like CPR could have saved him if delivered immediately. As an AI researcher, I discovered 4.6 million people die annually during the wait for help. This sparked MediGuard AI: instead of bringing patients to hospitals, we bring hospital capabilities to patients. By combining autonomous robotics with AI trained on 57 million medical cases, we created a system that arrives in 90 seconds to perform life-saving interventions.
How it works
MediGuard AI works like an emergency room on wheels that drives itself to save lives. When someone calls for help through our app, voice command, or emergency button, the nearest MediGuard unit instantly activates and navigates to the victim using cameras and sensors — like a self-driving car optimised for all terrains. It can climb stairs, cross rough ground, and reach places ambulances cannot. Upon arrival, the robot uses thermal cameras and vital sign sensors to assess the patient, while its AI brain — trained on 57 million real medical cases — determines the exact intervention needed. For cardiac arrest, robotic arms perform CPR with perfect precision (5 cm depth, 100 compressions/minute) while voice instructions guide bystanders. Built-in defibrillator pads automatically position themselves and deliver shocks if needed. For severe bleeding, the robot applies tourniquets or pressure dressings. It can even inject emergency medications for allergic reactions.
Design process
Our journey began with field research in Malaysian emergency departments, where we documented that 73% of preventable deaths occurred simply because basic interventions were not performed quickly enough. This insight drove our first prototype: a basic wheeled robot carrying medical supplies. Testing revealed critical flaws — it was too slow and could not navigate stairs or rough terrain where many emergencies occur. Version 2.0 introduced our modular design: a drone for rapid aerial assessment combined with a ground unit for intervention. While faster, we discovered that dropping supplies was not enough — victims needed actual medical procedures performed. This led to our breakthrough in Version 3.0: integrating robotic arms capable of performing CPR, applying tourniquets, and administering injections. The game-changer came when we incorporated our 57-million medical QA database, transforming MediGuard from a remote-controlled tool into an autonomous medical expert. We spent months perfecting the AI's decision-making through 1,000+ simulated emergencies, achieving 97.5% accuracy. Community trials in five underserved areas exposed new challenges: cultural sensitivities around robot care, need for local language support, and importance of familiar human voices during emergencies.
How it is different
MediGuard AI is the world's first fully autonomous medical intervention robot — not just a delivery system or diagnostic tool, but an actual robotic paramedic that performs life-saving procedures. While ambulance drones only drop supplies and telemedicine requires untrained bystanders to act, MediGuard physically performs CPR, controls bleeding, and administers medications with superhuman precision. Our revolutionary edge comes from the largest medical AI ever deployed: 57 million QA pairs give it knowledge exceeding 10,000 emergency physicians combined. Competitors' medical bots have thousands of scenarios; we have millions. This massive training enables split-second decisions. Unlike fixed-location AEDs or first-aid stations, MediGuard comes to you — navigating autonomously through buildings, stairs, and rough terrain that ambulances cannot reach. The modular design means communities can afford it at $5,000 versus an ambulance's $500,000.
Future plans
Our immediate goal is deploying 100 units across Malaysia by 2026, targeting rural areas where we have already saved 15 lives in trials. We are partnering with the Ministry of Health for nationwide integration and securing $2M seed funding for manufacturing scale-up. The business model combines government contracts with community subscriptions at $50/month per neighborhood — making it sustainable and accessible. By 2027, we will expand across ASEAN, adapting MediGuard for different healthcare systems while establishing local manufacturing to reduce costs. Long-term, we are transforming it to a community health guardian.
Awards
- 3rd Place the Malaysian Grand Challenge Scuttle Hackthon 2025 - Eureka International Brussels 2025 Platinum Medal Award -Participated in Berkeley Agent X LLM competition, awaiting results
Share this page on