Using deep tech for better healthcare

Artificial (Intelligence) Neural Networks are statistical models aimed at solving complex pattern recognition problems.

Hyderabad-based startup CognitiveCare has developed two models Onco and MIHIC, using this deep technology.

The platform aims to empower doctors, policy makers, insurance companies and their customers (patients) to make the right decision at the right moment.

Launched in 2018, the health-tech startup is the brainchild of Venkata Narasimham Peri and Dr Suresh Attili.

The former comes with more than 20 years of experience in deep management consulting, while Dr Suresh is a practicing medical oncologist and physician scientist with over 200 stem cell transplants and 170 publications to his credit.

Can early signs of diseases be found before it manifests in the human body?

This question led two master minds to utilise algorithms for reading the vast data available.

The end result -- understanding a patient's medical history, the risks associated with it, and timely intervention, in turn saving the governments, insurance companies and individuals wealth.

Peri says, "A disease occurs long before it manifests. Can we find those signals at the early stage?

And, even though different patients are suffering from the same disease, every patient is different because of their genetic profiles, lifestyle choices, drugs used etc.

Here comes in the need of customisation of treatment for every patient, even though they are suffering from the same set of diseases."

Dr Suresh adds, "Personalised medicine beyond pharmacogenomics is a big gap to plug. It has a deeper meaning, than simple gene signature, medical record and lifestyle.

Besides, to have a holistic look of 3,000 patients is humanly impossible. We must provide an abstract (summary) to make it handy for the doctors.

CognitiveCare's platform maps past medical reports, matched subject's data, lifestyle pattern etc, to come with a summary in the form of a score, which is understandable, reproducible and dynamic."

The two models -- Maternal and Infant health Insights and Cognitive intelligence score (MIHIC) and Onco survival score and adverse event score for cancer patients.

Not to forget, these models/scores undergo clinical validation from doctors for final approval before implementation.

At present, CognitiveCare is in partnership with five centres in the US and India, providing data.

For this (data), they work with research institutes, maternal and infant health not for profit organisations, clinical laboratories, etc.

The health-tech adheres to the standards of HIPAA (Health Insurance Portability and Accountability Act of 1996), a United States legislation that provides data privacy and security provisions for safeguarding medical information.

The pressing issue witnessed by CognitiveCare in the US and India is in regard to the maternal and infant health. "We have already identified 19 risks a mother can undergo in prenatal, perinatal, neonatal and postnatal stages and 16 risks an infant could face.

Officially the MIHIC model is launched, we are in talks with prospective clients for the same. Our Vizag facility is dedicated to work on adding more risks into this model," says Peri.

Coming to cancer, their initial focus areas are Multiple Myeloma Cancer and Glioblastoma cancer.

In this, the CognitiveCare Onco survival score and adverse event score depicts the chances of survival and adverse effect of cancer treatment along with other parameters. The founders expect this model to be ready for the market by next three months.

The third model the startup is working on – most sought by the US health departments - is related to mental health. This platform will provide the score reading propensity towards substance abuse, informs Dr Suresh.

"However, our current focus is the two cancer categories mentioned earlier and maternal infant health for which we are building the quantitative scores.

Our goal will be to bring in as much data as possible to build the models. Besides, we believe that as we have already built few models based on high end computing AI neural network, we would look to add on other diseases too, going ahead," adds Peri.

CognitiveCare has its presence in Hyderabad, Vishakapatnam and San Francisco in the US. The B2B model will have clients representing hospitals, insurance companies and the government.

Yet to count their first sale, the founders say they will be charging based on usage requirement of a client, with room to customise the model too.

Founded in 2017, the founders have self-funded the startup with a seed capital of Rs 3 crore. Till date, over Rs 1 crore has gone into working expense – looking forward to earn their revenues.

The founders claim to have a line-up of Rs 6 crore investment from family and friends, which will be ample enough till they bag their first client, they say, adding, "we are open for partnerships, only on the basis of value addition not capital addition. Our belief, the best money is the customer's money, hence intend to be self-funded until the first customer is acquired."

Currently, their team strength is of highly skilled eight scientists, including the founders. By end of FY19, they have plans to expand the team up to 30 to 50 members.

After having added on clients, the founders intend to expand their platform by adding other diseases. That time they look at aggressively scaling by introducing individual centre of excellence for different health segments like cardiology, nephrology etc.

While founded in 2017 itself, CognitiveCare received a valuation of $5 million, in 2018 it elevated to $7.5 million and at present, the talks are of up to $10 million, according to the founders.

This represents the command an exclusive idea holds, the growth an AI health-tech platform can clock headed and potential of Hyderabad to produce the next Unicorn.