Welcome to LEAP for dogs, the leptospirosis AI prediction tool.

Leptospirosis is life threatening disease caused by a bacterial infection. In dogs, this disease can result in severe kidney and liver damage. With early treatment, prognosis for even the most severely affected dogs can be good\(_{(1)}\). This tool utilizes machine learning to look for patterns in health data to predict a diagnosis of leptospirosis in dogs.

When this tool was tested in dogs in Northern California, it correctly identified all dogs with leptospirosis and 90% of dogs without leptospirosis\(_{(2)}\). We invite you to help us test this tool in a wider population of dogs suspected of having leptospirosis.

How to use:

1. Enter health data for any dog that is sick and suspected of having leptospirosis. This will include breed, weight, complete blood count, serum chemistry panel, and urinalysis information. If leptospirosis antibody titers are available (microscopic agglutination test), you can enter that information as well. (Warning: If some of the input data are missing the result MAY vary).

2. That information will be utilized to make AI prediction on the final diagnosis of leptospirosis. If an antibody titer result is supplied, two predictions will be made, one with that information and one without that information.

3. A follow up email will be sent to you in 1 week to gather more health information including results of leptospirosis testing (PCR or antibody testing).

References:

1. Sykes, Jane E., et al. "2010 ACVIM small animal consensus statement on leptospirosis: diagnosis, epidemiology, treatment, and prevention." Journal of Veterinary Internal Medicine 25.1 (2011): 1-13.

2. Reagan, Krystle L., et al. "Use of machine-learning algorithms to aid in the early detection of leptospirosis in dogs." Journal of Veterinary Diagnostic Investigation (2022): 10406387221096781.