Dogs are amazing. But as many pet parents will tell you, they can be remarkably good at hiding when something is wrong. My Labrador-Great Dane mix, for example, contracted tick fever several times over his 14 long years, and it was always at least a couple of weeks before I suspected something was off.

It nearly gave me anxiety ulcers a few times. Hoomanely, a new startup out of Palo Alto, thinks it can help people like me spot health problems in their dogs much sooner with an AI platform that gathers data using a sensor-laden feeding station.

Dubbed the EverBowl, the station measures dogs’ food and water consumption and eating speeds, and records chewing and swallowing sounds, facial thermals, oral motions, and a few other signals. For related coverage, explore our detailed analysis on Technology and Digital Systems.

Company’S Platform Analyzes Data Updates

The company’s AI platform analyzes that data to establish a baseline and then builds a health record. Any subsequent and prolonged departures from that baseline are shown to the pet owner via an app and in long-term reports, which the company says can provide more useful data to veterinarians if and when they get involved.

The app also accepts information from the user, so the baseline can be kept updated with information outside the sensors’ scope. “Practically, the bowl is the best capture environment in a pet’s life: same place, same posture, same routine, twice a day, for years,” Sai Supriya Sharath, co-founder and CEO of Hoomanely (pictured above, in the middle), told TechCrunch.

She added that appetite, hydration, and oral comfort are among the first things to be disrupted “Dogs are evolved to mask lethargy and limping. They are far less able to mask how they eat and drink … reduced or altered intake and changed drinking are presenting signs across a wide range of conditions.” Hoomanely says its platform was built after an 18-month beta-testing phase during which the company gathered about 5 million data points across more than 80 dogs.

Testing Platform Detected Changes Updates

During testing, the platform detected changes in one dog’s eating patterns, which were eventually tied to a chipped tooth that was starting to become infected. In another case, the company’s app demonstrated a sustained change from the dog’s regular eating and temperature baselines, which prompted the owner to take their dog to the vet and get a diagnosis for tick fever.

Read More: Meta agrees: to sweeping changes to restrict kids’ access to its apps as part

While this sounds useful, there are some caveats. The cohort for this beta was quite small, and the company seems to have shipped only about 50 devices so far. It currently offers the EverBowl and the app reports through a $29-per-month subscription service.

Sharath also shared the company does not yet have independent sensitivity, specificity, or false-positive rates for clinical events, as that would require a study comparing the system’s alerts against actual diagnoses. Hoomanely admits that its platform is not a replacement for a veterinary diagnosis but is instead meant to help owners and vets to identify potential issues faster.

The company is using its data for formal veterinary studies to find out whether its system works across a larger group of dogs. “The veterinary studies being designed will measure sensitivity, specificity, positive predictive value, and false positives for each alert category, and we intend to report results Along with those studies, Hoomanely plans to build more devices to gather data.

EverSense is planned to be a wearable that will measure movement and rest information, and EverHub will be able to accept inputs from third-party devices such as smart collars, feeders, or home devices to record environmental data. Sharath shared the long-term plan for Hoomanely is to eventually become an animal health data company Because the data capture system isn’t specifically designed for dogs, she shared it may eventually expand to other animals, such as cats, livestock, or horses.

The startup has raised $1.8 million in pre-seed funding so far and says it’s starting conversations for a seed round. When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

You can contact or verify outreach from Ram iyer@techcrunch.com. Don’t miss out. The startup community will gather to answer a pivotal question: How do you build sustainably in the AI era?

Fitbit founders launch Luffu Link, an LTE health and safety band Two years after launch, Walmart’s Flipkart is closing in on India’s quick-commerce leaders Inherent, founded Michael Polansky is training an AI model on skin that’s still alive How AI accounting startup Rillet raised $100M and became a unicorn in 48 hours Tesla’s solar roof is dead — here’s what went wrong Oura faces legal case accusing it of misleading consumers about sleep-tracking accuracy