In 2011, Delhi reported a total of 7,281 road accidents. About 30% of these accidents (2,065) resulted in a fatality. Half of those dying on the roads are 'vulnerable road users': Pedestrians, cyclists, and motorcyclists.

While traditional thinking would hinge on 'scoop and run' (gather the patient and rush him to the nearest hospital), that is easier said than done in Delhi. Case in point, forty-year-old Mohammed Chand has been working with the St Stephen's Hospital in Old Delhi as an ambulance driver for the past five years. He has had three people died in this vehicle last year and says "this stupid traffic is responsible for my plight," pointing a finger accusingly at the bumper to bumper traffic.It is around five in the evening and Old Delhi is bursting at the seams on a road that connects one of the largest hospitals in this part of the capital with the rest of the city.


"Now the traffic has increased so much in Delhi that we are helpless. Last year there was this 50-something man with a cardiac arrest. His son was sitting next to him and there was nothing we could do. The traffic killed him," he says.


Heart wrenching as the above is, it forces us to reconsider the fundamental way by which we think of providing care to trauma patients. On the basis of the above, we pivot this problem of getting the victim and care provider to meet within the golden hour on 2 major axes (organizational and private-citizen):

  1. It doesn't have to be in the ER - the ambulance withambulance to hospital based telemedicine systemcan be equally effective in getting the right care started without having to wait for the perilous journey back to the hospital stuck in traffic.The key challenge revolves around getting an ambulance to the patient in the least amount of time possible. From there, even if it takes a little longer to get the patient to the hospital, it is a better alternative to a delay in getting an ambulance to the victim in the first place.

  2. Beyond the traditional ambulance model, there is also a need to generate awareness about basic first-aid procedure and technique among the general public who can often contribute to save a life. Especially so in certain core trauma scenarios like having a stroke, heart attack or a head injury. A simple SMS based solution can help here.

Our solution works as follows:


From the organizational perspective, we identify accident prone hot-spots that are traditionally difficult to get to due to traffic – this information is obtained from observed traffic data and previous accident reports. Additionally, various environmental factors such as the weather and any major events are taken into consideration and fed into a Machine Learning Predictive Analytics model to arrive at a risk score for major points in the city.

Subsequently, a few businesses are identified around these areas who are willing to spare a parking spot to act as an ambulance bay during rush-hours. This ensures that during periods of peak risk, there exists an ambulance close enough to reach a high risk zone in the shortest amount of time possible, while still allowing for traffic.

Finally, from a citizen volunteer perspective, doctors and citizens can opt-in to be notified if there is an accident near them where they are able to assist. Interested citizens could be given basic web-based training on proper first-aid procedures in the case of trauma events like head injuries. Alternately, training classes can be provided in basic life-saving skills by the local hospitals and NGOs.


The following scenarios kick into action when a crash is detected -by car (airbags, impact sensors etc.), call-in reports, visual confirmation through city traffic cameras etc.


For doctors and citizens who volunteered to sign up to be alerted and assist, based on their vicinity, they will get a text and the coordinates of the victim. Additionally, any one near the victim can text or call a hotline to get instructions on how to handle the trauma situation unfolding in front of them.


The nearest ambulance is alerted and the intersection is sent to the GPS unit of the ambulance - and an alternate ambulance is sent to this location to take its place. The locations of the ambulances are tracked using an integrated OpenXC connector and the volunteers are notified of the arrival and ETA of the ambulance.

The ambulance heading to that location is equipped with an ambulance to hospital based telemedicine system thatoffers not only directlive transmissionof voice and video but complete patient physiologic data, e.g. ECG, 12-lead STEMI, blood gases, ultrasound, e-PCR, EHR, blood pressure, and a lot more. Ultimately, a remote ER team can get to work right away with the help of the EMTs by using video toassess method of injury, the patient's exact state, situational awareness and much more.

We feel that with the citizen-organizational partnership model outlined above, many more lives can be saved.

Beyond the above, looking forward, India should look to enact adequate laws that address all five behavioral risk factors (speed, drink-driving, helmets, seat-belts, and child restraints) to bring down the incidence of accidents in the first place.

Check out for detailed scenarios and demonstrations.

Share this project: