June 6, 2023
The Role of ChatGPT in Veterinary Health
By: Lizzie and Nicky, Medical Strategy Team, Evoke Mind+Matter
We asked ChatGPT to evaluate its role in veterinary health
ChatGPT burst onto the veterinary scene in March after a pet owner used the AI platform to diagnose a tick-borne disease in their dog, Sassy the border collie.
The pet owner took to Twitter, declaring, ‘GPT4 saved my dog’s life’, and the story exploded. With 9.8m views and Google news reports stretching from India to New York – a pandora’s box of questions, concerns, and possibilities for AI in veterinary medicine has now opened.
So, we decided to do some digging and put ChatGPT to the test.
The open AI language model helped Sassy get a diagnosis
Sassy’s owner pasted lab results into the chatbot from bloodwork done by the owner’s veterinary practice, which had revealed anaemia. ChatGPT provides a disclaimer for any request for veterinary advice, stating, “I am not a veterinarian, but I can try to help…” However, in this case ChatGPT offered differential diagnoses, including babesiosis (a parasitic tick-borne disease) and associated immune mediated haemolytic anaemia (IMHA).
Sassy’s owner went to another vet for a second opinion, supported by ChatGPT’s suggested diagnoses. Following additional blood tests, the AI’s differentials of babesiosis and IMHA were confirmed, and Sassy is now on-route to a full recovery.
Here’s what ChatGPT said…
The potential impacts ChatGPT could have on the global veterinary industry are broad and far-reaching, and the risk of reduced reliance on veterinary professionals will require close observation.
So, we decided to ask ChatGPT to evaluate its role by asking: “What are the implications of AI use on the global veterinary industry?”
ChatGPT responded: “use of AI in the veterinary industry has the potential to transform the ways animals are diagnosed and treated,” highlighting these four areas for consideration:
Increased accuracy – AI can analyse large amounts of data quickly to identify disease patterns that may not be apparent to humans, leading to more accurate diagnoses and treatment plans, improving the overall health and wellbeing of the animal.
Cost-effective care – identifying potential health problems earlier in the disease timeline, allowing for earlier interventions and potentially reducing costs. AI can also assist veterinarians in identifying appropriate treatment options, reducing unnecessary and ineffective treatments.
Global access to veterinary care – access to veterinary care via AI-integrated telemedicine platforms in areas with a shortage of vets or limited accessibility, allowing owners to connect remotely with veterinary professionals, reducing barriers to accessing care and improving welfare.
Ethical considerations – at the risk of dehumanising the veterinary care process, nothing can beat a physical examination from your vet.
We also tasked ChatGPT with helping us find the cause for some specific symptoms in a dog: “Why has my dog been urinating more frequently than normal and drinking a lot of water?”
ChatGPT responded with a list of five potential causes —all plausible but not comprising a comprehensive list. It reassuringly concluded by saying: “These are just a few potential causes for increased urination and thirst in dogs. A veterinarian can perform a physical examination, run diagnostic tests and provide a proper diagnosis based on their findings. It’s important to seek veterinary care promptly if you notice any changes in your dog’s urination or drinking habits, as early intervention can improve outcomes.”
What does this all mean for the future?
Potential risks and ethical implications are unpredictable. ChatGPT itself has recommended that veterinary professionals be actively involved in the development of AI models to ensure that they are designed to meet the needs of animals and their owners.
Sassy’s owner on Twitter said, “I can’t imagine what medical diagnostics will look like 20 years from now.”
…Can you?