Fondation Pierre Fabre

Saving lives by expediting ECG diagnosis and accelerating treatment

Main beneficiary countries:
India - Malaysia - Kenya - Indonesia - Nigeria

About the sponsor

Tricog Health Pte. Ltd.

In 2014 Dr. Charit Bhograj, a reputed interventional cardiologist, and a trio of experienced technologists (Dr. Zainul Charbiwala, Dr. Udayan Dasgupta and Mr. Abhinav Gujjar), started Tricog with the mission to saving lives by expediting cardiac diagnosis and accelerating treatment. In 2015, Tricog launched a remote electrocardiograms (ECG) interpretation service by combining skilled physicians with state-of-art IOT and AI technologies to deliver accurate ECG diagnosis anywhere in the world, within a few minutes. Going further, Tricog plans to use the same IOT+physician+AI approach to provide additional cardiac diagnostic services for holters, patch monitors and echocardiograms.

Sector: Industrial: Startups, enterprises, etc.

Country of origin: Singapore

The organisation has:

Full-Time Equivalent
Service providers

Initiative overview

Healthcare themes targeted

  • Primary healthcare
  • Cardiovascular illnesses

Stage of development:

  • Routine project/operational

Area where initiative is utilised

  • International (in several countries)

Initiative start date

  • 02/01/2015


Financing method

  • Private (private investors, crowdfunding, philanthropy, etc.)

Economic model(s)

  • Revenue generated by the beneficiaries/healthcare facilities

About the initiative

Millions of people in India and other developing countries lose their lives to cardiovascular diseases every year. Cardiovascular diseases are a growing problem since they are linked to factors like age, sedentary lifestyles, poor eating habits and high levels of stress. In spite of the large number of deaths and the growing number of heart patients, these countries lack immediate diagnosis and coordination systems. The likelihood of surviving a heart attack is over 80% if action is taken within the "golden hour". However, the average time between symptoms and treatment in India is over 6 hours. By simply reducing this, millions of lives can be saved every year.

ECGs are the primary means of diagnosing serious cardiac conditions like heart attacks. While ECG equipment is a fairly commonplace, reading ECGs require the skills of a cardiologist or an experienced physician. Given that such expertise is limited in India, misdiagnosis and delayed diagnosis are commonplace. Furthermore a delay of an hour in diagnosing heart attacks has been shown to significantly increase the risk of mortality and permanent damage to the heart.

An ECG can be used to diagnose over 200 cardiac conditions. Although higher end ECG machines come with in-built ECG analysis algorithms, such embedded algorithms have limited accuracy, especially when it comes to identifying heart attacks, possibly causing significant delays in diagnosing cardiovascular conditions and increasing risks to the patient's life.

The InstaECG service provides accurate (physician verified) ECG reports. If the patient is found to be critical, Tricog works with neighbourhood tertiary hospitals to accelerate treatment. The InstaECG service achieves accurate readings within minutes by combining a highly trained medical team reading ECGs 24/7/365 with state-of-art IOT & cloud based algorithms.

By using custom built IOT connectors, Tricog successfully converted inexpensive off-the-shelf ECG machines into cloud connected ones and allowed them to leverage advanced proprietary algorithms (THA) on the Tricog cloud. Thus each incoming ECG is first diagnosed by THA, afterward the Tricog internal medical team over-reads the diagnosis. Hence incoming ECGs are human verified and annotated in real-time, allowing the algorithms to learn and improve over time. Thus a virtuous cycle of AI helping humans and humans helping AI has been created in a real-time production environment. The Tricog data store currently has over 1.5M ECGs that can be used for training the THA models, and parts of it have recently been submitted for FDA approval. The THA platform first employs advanced signal processing techniques to make a large number of measurements on the ECG graphs, followed by a variety of classifiers for the 200 odd conditions that it diagnoses. Broadly speaking, THA incorporates the following three types of classifiers:

  • Deep Learning Based: for conditions/ECG morphologies that are frequently occurring and hence have a large presence in the Tricog data store.   
  • Machine learning/Shallow Neural Net Based: for conditions/ECG morphologies that have moderate frequency of occurrence and use ECG features that can be robustly estimated.
  • Expert System Based: These expert systems leverage medical domain knowledge are used for conditions which occur infrequently.

Coupling the algorithms with a carefully designed ECG viewing system, has driven the Tricog medical team’s average ECG diagnosis time to less than 30 seconds, the Tricog internal medical team simply need to verify, and at times, correct the diagnosis.

For a low monthly fee, Tricog provides its customers (clinics, small hospitals and diagnostic centres) with a clinical grade 12-lead ECG machine coupled with a proprietary cloud connecting device (Tricog communicator), and the Tricog Android/iOs App. Each time an ECG is taken at such remote centers, the ECG is pushed to the cloud, where it is analysed and a Tricog physician verified report is sent back via SMS, email and app to the remote center within ten minutes. In addition, Tricog coordinates with neighborhood tertiary hospitals to expedite care for critical patients discovered at these remote centres.

The service is currently deployed in over 1500 centers in India and being deployed in South East Asia and Africa. By combining the best of technology with medical domain expertise, the InstaECG service has been designed to be affordable, accurate and accessible for all sections of society.

Fields of application:

Telemedicine (remote diagnosis and consultations)

Target audience

  • Healthcare professionals and structures (hospitals, healthcare centres/clinics, health networks)
  • Entire population

Initiative objectives

  • Decreased mortality
  • Improved treatment

Key figures

1500000 Number of beneficiaries since launch

100000 Number of users per Month

Materials used

  • Cellular (mobile) phone
  • Smartphone
  • Tablet
  • Connected objects
  • Clinical grade 12-lead ECG machines

Technologies used

  • Mobile telecommunications (without data connection)
  • Internet
  • Geolocation
  • Mobile app (Android, iOS, Windows Phone, HTML5, etc.)
  • Artificial Intelligence

Offline use


Open Source


Open Data


Independent evaluation




Industrial: Startups, enterprises, etc.