AI/Watson

Cognitive canines: Boosting graduation rates for guide dogs

Share this post:

Guiding Eyes for the Blind is on a mission to provide exceptionally trained guide dogs to people with vision loss. Each dog is bred for ideal attributes, both physical and behavioral, then trained for 20 months before finally paired with its blind human counterpart. Guiding Eyes dogs help individuals gain greater independence and confidence.

We rely on data to ensure the highest-possible graduation rates for our dogs. We’re now also benefiting from cognitive insights. Improved insights are especially helpful since the more of our dogs that graduate, the more we can meet the growing need for guide dogs.

Breeding healthier pups

We are increasing the success rates of our guide dogs by using data that calculates and identifies the genetically best dogs to keep for breeding. In 2003 we completed our database to track genetics data and have made huge progress in decreasing the incidence of diseases in our dogs. For example, before introducing data analytics, we had about a nine percent rate of hip dysplasia, a very common health problem with dogs. Now, nearly zero percent of our dogs develop this health issue.

Training more successful guide dogs

Identifying the medical issues is relatively straightforward, but the behavioral element is much harder. We need to produce dogs that can handle the responsibility of guiding a person with impaired vision. The guide dog must be able to stay emotionally calm despite stressors such as oncoming traffic in busy cities and determining when is the best time to cross the street. Identifying and training the ideal temperament in future guide dogs is no small task.

Much of our data over the years has come from trainer feedback and videos. We worked with Dr. Chris Singh of San Jose State University and his students to pull from that unstructured data using cognitive insights from IBM Watson. Early findings are quite impressive and point to opportunities to better match our volunteer puppy raisers with the right dogs for better results.

Each dog costs $46,000 to get through our program, and that’s whether they graduate or not. All of our funds come from donations, and the dogs are free to the people who get them. So each additional dog that we can successfully raise and train as a guide dog benefits us, not only on the human side, but also helps to properly manage our resources. Our current goal is to place 10 percent more dogs a year, taking us from the 150 we successfully place now to 175 matched dogs. That’s a drop in the bucket for helping the many people that are in need of and would like to use a guide dog. And, remember, each dog that successfully completes our program helps one more person.

Future vision: Combining research to help more people

In the future, I’m hoping to have Watson help us tie together the other work we’re doing with universities. For example, researchers at the Broad Institute of MIT have been working with us to identify a gene that causes cancer in dogs. As Watson learns more and more, it’s one of my dreams that we also expand our behavioral data research. By combining the hundreds of thousands of data points that we have with Watson, I believe we can not only bring guide dogs to more people in need, but also enhance work that’s being done in different fields.

Listen to the IBM Wild Ducks podcast below to hear more about Guiding Eyes cognitive solution journey. And, read the full case study to learn more about how Guiding Eyes is using cognitive computing to advance the art and science of raising guide dogs.

Learn-More   Find-A-Rep

Save

Save

Save

Save

Save

Save

Guiding Eyes for the Blind, Director of Genetics and Breeding

More stories

How AI helps Japan Airlines personalize the travel experience

For airlines, the sheer volume of flights and travelers can sometimes make it difficult to provide a personalized customer experience. When airports are busy and flights are full, passengers sometimes feel that the airline simply sees them as objects to be transported from point A to point B. In response, Japan Airlines decided to set […]

Continue reading

The problem of rural internet access: How AI is helping

A disconnected rural population Nineteen million people in the United States lack high-speed internet access. Many of these people live in rural areas, where fiber, coaxial cable, DSL or any other kind of broadband infrastructure is scarce or nonexistent. This is a huge issue for rural community administrators and school officials who have been struggling […]

Continue reading

Converting weather forecasts into smart business decisions with AI

Few would doubt the idea that weather has a big impact on business. Think of how something as basic as holiday retail spending can be sharply influenced by a blizzard or frigid temperatures. While those kinds of big, macro effects are easy to understand, the real nature of the relationship between weather and business is […]

Continue reading