Youtuber designs software to monitor her cats poop

Two years of hard work paid off – the machine learning program can now tell whether any of her two cats are constipated.

A software engineer, who goes as Estefannie on Youtube, had a problem. She caught Teddy, her "64-cat-years-old" feline, eating plastic – too much of which can be lethal.

Estefannie took Teddy to the veterinarian, who could not tell how much plastic the cat ate. It would cost her over $3,000 to find out. So, she didn't do it, instead opting for an alternative – to just watch and track if Teddy poops.

If he does, then there's likely no plastic in his intestines. Which means "he won't die, and he might actually love me back," Estefannie said in a video she posted on Youtube.

As she went on to write a Python code and set up a sensor to take pictures of Teddy using his litter box, she realized the futility of her efforts. This was a shared litter box, after all. Her other cat, Luna, also used it.

"So when I see poop, I don't know who pooped; there're too many poops. Also, Teddy was not impressed with my Python script," Estefannie explained.

Naturally, she then researched machine learning and found that what's needed for it to work is a bunch of pictures of her cats. She ended up taking over 50,000 photos of them pooping using a picture-taking script she already had – and the algorithm could now tell the difference between Teddy and Luna.

At this point, nothing can hold Estefannie back, so she builds an entire dedicated bathroom for her cats out of plywood. She uses it to disguise a computer, sensors, and infrared cameras to record her cats' bowel movements, day or night.

A couple of additional stumbling blocks later, she had a prediction model that could accurately tell when and which of her cats pooped or peed. She also set up a pretty website. It shows when the litter box is vacant and when it's in use by either Teddy or Luna.

"I basically made a poop monitor for my cats. Like a baby monitor but a poop monitor," Estefannie said in a blog post, where she explained her project in detail. "Yes, I watched them poo and pee for almost two years. Don't ask me about my mental health, I did it for science".