Building a Better Brain: Saffron Cognitive Computing Platform Replicates How We Associate Facts

Saffron_logoSaffron Technology has been on a quest since 1999 to replicate the way the human brain learns using associative memory. Saffron is now commercially available as a cognitive computing platform following beta testing for real-time operational risk intelligence and decision support in defense, energy, healthcare and manufacturing applications.

When you think brain-like, it’s likely that IBM Watson comes to mind. Watson may be brainy, but it does not think like a brain does! Instead, Saffron has invented the a true artificial intelligence that mimics aspects of the human brain by inventing an Associative Memory approach to organizing data, which is similar to how the human brain processes the ‘information organization task’ in order to draw conclusions and ‘take action’.

As CEO Gayle Sheppard and chief product officer Ian Hersey explain, the heart of the company’s intellectual property processes data on three types of entities–people, places, and things. This database is geared toward helping users find meaningful patterns in the data, including who’s connected to whom, why, when, and where, across time and space. In the video below, CEO and Chairman Gayle Sheppard discusses Saffron Technology at GigaOM Structure Data 2014:

This is all in contrast to IBM Watson which requires an intensive knowledge-engineering approach to tune it to support different domains like healthcare. Moreover, relational databases lose much of this valuable information a priori, as the restrictions of columns and fields are rigid, predefined, often due to the constraints on size, structure and management of computer maintenance.

Saffron instead deals with unstructured and structured data and can store attributes and associations between data naturally whether that data is being ingested from smart devices, databases, or unstructured notes that will be parsed into structured data through its natural language processing component. Saffron not only reads free text to identify relationships, it analyzes the strength of relationships based on correlation, count and context.

Key for organizations is that, like a brain, Saffron finds connections among data across diverse sources, without the need for rules or modeling, while also learning incrementally and anticipating outcomes based on patterns it finds in the data. A good example is in the areas of cardiology, where distinguishing between two types of heart disease—restrictive cardiomyopathy and constrictive pericarditis—continues to pose a challenge even for top cardiologists, who provide correct diagnoses of these diseases only 76 percent of the time.

Dr. Partho Sengupta, Director of Cardiac Ultrasound Research and Associate Professor of Medicine in Cardiology at The Mount Sinai Hospital, needed a way to accurately identify disease patterns resulting from echocardiograms in order to improve diagnostics and save more lives. He wanted to use all available data attributes, not just the seven variables commonly used by his colleagues. To answer his request, Saffron ingested data from 10,000 attributes per heartbeat and per patient, using 90 metrics in six locations of the heart, collected 20 times data in a single heartbeat—finding real-time patterns from this huge amount of data that no human can perform. Saffron, without the use of restrictive models or extensive data training, improved the diagnosis accuracy by 90 percent, outperforming top physicians and state-of-the-art decision trees (54 percent).

Cognitive Computing in the style of IBM Watson, based on analyzing a question, generating a hypothesis of what the answer needs to look like, and then trying to find the best match from its facts base, doesn’t provide the tools to distinguish between the two conditions, when you don’t know in advance which attributes or combination of attributes will be important for determining one condition from another.

 

Sign up for the free insideBIGDATA newsletter.

Leave a Comment

*

Resource Links: