Srivastav

BROOKLINE, Mass. — It's not hard to tell what's on the mind of company officials at Brookline-based Cognika Inc.

The firm is busy trying to help the health care industry make better decisions about therapeutic options with the company's signature ESP analytics platform, according to company co-founder and Chief Executive Officer Ram Srivastav.

One way to do that is to streamline what Srivastav defines as a time-consuming and costly process by which Cognika's clients typically make decisions related to clinical drug trial results.

The cognitive computing engine, also known as Cognika's ESP, learns through experience in a fashion somewhat similar to the human brain. The engine is exposed to vast quantities of factual information, from which it "learns," making automated analyses of data that is entered as search criteria. ESP can then make inferences based on search criteria, predicting the reaction a certain patient might have to a particular medication, said Srivastav.

This "prediction" is made possible through the processing of vast amounts of data, which allows ESP to identify patterns and discover relationships between terms entered into ESP's database.

The user — often a medical professional — will enter a criterion for which ESP may find a hidden relationship. ESP allows for more precise, better decision-making in the world of health care, according to the company.

Often, language evolves over time, causing medical terms to have multiple associations, Srivastav said. In order for the hidden relationships, or what's also known as recognition of new content, to emerge, ESP will parse through millions of pieces of data provided by various hospitals and health-care professionals and produces models as its output. A new model is created for each new prediction, said Cognika Chief Software Architect Shashi Kant.

Srivastav and Kant co-founded Cognika in 2006, but the seeds for the company were planted more than a decade ago, when the duo worked together in India.

The company has four full-time employees at its Brookline headquarters, as well as a number of consultants and contractors in the United States. In India, Cognika uses an information technology subcontractor called MIND, with 450 IT professionals to draw from.

Kant first began thinking about cognitive computing concepts 12 years ago, when he and Srivastav worked for a company in India that provides machine-learning services, Srivastav said.

Kant more fully developed the idea for the ESP platform while doing graduate work at the Massachusetts Institute of Technology in Cambridge, Srivastav said.

The company continues to have ties with MIT through the school's "Ventureship" program, an informal partnership that seeks to pair entrepreneurially minded business students with area businesses, giving hands-on experience to tomorrow's business leaders, Srivastav said.

"From our perspective, [the program offers] access to young people, very enthusiastic, [who are] willing to work on cutting-edge stuff," Srivastav said.

The students are engaged in evaluating new business ideas. In the case of Cognika, MIT students worked on market research and drew up a business plan.

Cognika also is taking advantage of MIT's Venture Mentoring Service, which advises and mentors smaller businesses.

Cognika is currently working with VMS to have a group of student mentors review Cognika's progress and help guide the company through it is growth phase, according to Srivastav.

Srivastav said Cognika's product line will continue to focus exclusively on the life sciences and health-care fields.

The company's products evolve constantly as feedback is received from clients, said Srivastav. He said Cognika has "to monitor how customers are evolving [because those are the people who are] driving the strategy on what the product needs to be."

"You have to keep your ear to the ground • and do things as the market evolves," he added.

Cognika is developing a series of products aimed at increasing the ease of carrying out clinical trials.

The Trial Accelerator, an offshoot of Cognika's ESP technology, which Kant likens to the Turbo Tax of clinical trial protocol, integrates data from 700,000 clinical trials and 6 million journal articles.

Reports indicate that Cognika collaborated with Massachusetts General Hospital in Boston to test an early ESP insulin-dosing model, although Kant and Srivastav did not specifically name the hospital due to a confidentiality agreement.

The Trial Accelerator is the first in a proposed series of six products in the Cognika pipeline aimed at meeting the information analysis needs of the pharmaceutical industry, thereby increasing the ease of clinical operations, said Srivastav.

The Trial Accelerator attempts to alleviate the challenge in the pharmaceutical industry of designing a clinical trial, said Srivastav. If a medical compound is commercially known by one name, but referred to clinically by another, smart analytics will find the two terms and correlate them, reducing confusion for researchers conducting the trial.

The software is in beta and being tested by various local pharmaceutical companies, although Srivastav would not say which firms Cognika is working with.

A customer pays a subscription fee, logs on to a Web application and gets instant access to all of the database information.

For example, in the case of diabetes, companies researching the disease can use the Trial Accelerator to sift through mountains of data to inform them which interventions have already been tested and the outcome of those tests.

The software walks the user through the information, aggregated from Pubmed, a search engine for accessing biomedical and life science citations and articles, and various clinical trials.

A company like Pfizer, that has performed thousands of trials, will add its own data to that of the public domain when using Trial Accelerator, further increasing the information in the database.

The user can narrow the search scope to the exclusion of any irrelevant information, said Srivastav.

As the name suggests, the software is specifically designed to be fast. According to Srivastav, each day spent testing a drug can add tens of thousands of dollars per patient to the cost of clinical trials.

According to Srivastav, Cognika's revenue has increased every year since inception, although he would not divulge the privately held firm's revenue figures. Srivastav said he has "stepped back from revenue, as hard as it is" in order to train his sights more wholly on building Cognika's projects out as a means of measuring the company's success.

Srivastav says the nature of his role as Cognika's CEO stands apart from his previous management experience at other companies, including oil and gas giant Sclumberger, where Srivastav once headed the Paris-based company's India Ltd. division, and Davul Power, where Srivastav acted as chief executive.

By comparison, Cognika has "a very different flavor." Most firms, he explained, are much more about keeping the cash flow in control, whereas with Cognika the focus comes down to finding and keeping customers.

"We had to work our ways towards [finding a clientele base] in the life sciences," said Srivastav.

Born in Kampur, India, Srivastav spent much of his boyhood in Delhi and considers it his "home on the Indian front."

He graduated from Columbia University's Graduate School of Business and earned a bachelor's degree from the Indian Institute of Technology.

Kant received a master's degree in engineering and management from the Massachusetts Institute of Technology and MIT's Sloan School of Management.