Artificial intelligence (AI) will usher in an era of better diagnosis of companion animal illnesses
I believe that artificial intelligence will, in time, revolutionise the diagnosis of illnesses in both people and their pets and I am not alone. For the sake of clarity, artificial intelligence (AI) is a process whereby computers begin to think like humans which means they can learn autonomously to improve their abilities. You can see that I’m referring to computers as humans in that sentence which is what they almost become. In fact it is projected that they will be able to diagnose illnesses better than humans. They will be more reliable, more consistent and harder working. After all computers can work 24/7 and they don’t need feeding or to sleep.
If you add into the mix virtual reality and augmented reality you can visualise a situation whereby veterinary work will be completely different in perhaps about 10 to 20 years time. It may come sooner because there is rapid development of AI all over the planet in a multitude of startups (new businesses starting up in a new niche with the prospect of making it big).
Start-up companies are everywhere and they are receiving seed funding by hedge funds and other investors in the millions of US dollars. For example, with respect to human medicine, a business which calls itself IDx has received $33 million in seed funding. They make an autonomous AI diagnostic tool for diabetic retinopathy. Another company based in California, Pr3vent Inc. has been given $1.5 million to progress their start-up business based on artificial intelligence to screen newborn eye diseases.
The American Veterinary Medical Association (AVMA) have a page on AI in veterinary medicine. They mention that AI technology has the power to rapidly analyse huge amounts of data. In 2019 the National Academy of Medicine wrote that, “AI has the potential to revolutionise healthcare”.
One danger is that people have to manage their expectations because at present their expectations are very high and they may be a tad overoptimistic. However, AI will progress rapidly and one fan is Dr Krystle Reagan, a veterinary internist at the UC Davis Veterinary Medical Teaching Hospital. She helped develop an algorithm to detect Addison disease which has an accuracy greater than 99 percent.
In dogs, Addison’s disease (endocrine disorder causing insufficient hormones) presents very vague clinical signs and can look like kidney or liver disease. They used blood work results from 1,000 dogs treated at the teaching hospital to train the AI program to detect patterns which suggested the presence of the disease. Once it had been programmed it was able to determine whether a new patient had Addison disease.
They use machine learning algorithms to identify “subtle patterns in the blood work”. They are hopeful that they can find patterns in the data which will help them to diagnose with some certainty that a companion animal has a certain disease.
AI can also be used to recognise patterns in x-rays. I recall that AI programs can read some x-rays more reliably and more precisely than a well-trained radiographer can.
Artificial intelligence is here to stay and it is not a fad or a temporary trend. It will have a huge effect on diagnosing illnesses in animals. There will be a need to manage ethical considerations when using AI in veterinary medicine. I guess that they will have to programme their computers with morals and ethics. If they do they will have morals and ethics which are more reliable that is present in veterinarians.
This inevitably draws me to a neat conclusion on cat declawing which is strictly against the oath of American veterinarians but they still carry it out. In five years time if they asked and artificial intelligent computer to make a decision about declawing it would tell them in no uncertain terms, No.
SOME ARTICLES ON VETS: