image:

With the help of AI-driven infrared imaging, the area of interest around the eye was clearly identified (highlighted in green hatching).

Credit: Osaka Metropolitan University

Conventional veterinary methods commonly rely on rectal thermometers to check an animal's body temperature. While well-established, this technique can be distressing for animals and does not allow for continuous temperature monitoring. Advances in infrared thermography now offer a non-invasive way to gauge body temperature, but challenges remain—mainly the difficulty in tracking ongoing temperature shifts and inconsistencies caused by manually setting target areas in the data.

To address these issues, a research group led by Assistant Professor Sueun Kim from Osaka Metropolitan University’s Graduate School of Veterinary Science conducted a study involving 11 calves. The team used infrared cameras to collect temperature data while simultaneously capturing video for a complete record of the animals’ facial temperature changes, particularly focusing on the eyes and nose.

Artificial intelligence helped automatically define the regions of interest (eyes and nose) within the video frames through segmentation techniques. The analysis revealed nearly 200 unique temperature variation patterns. Notably, there was a strong alignment between the top 10% and top 30% temperature values recorded in both the eyes and nose, indicating a reliable trend in thermal changes at these locations.

“This study allowed for the detection of steady, non-contact temperature patterns in cattle, paving the way for detailed statistical assessments,” explained Dr. Kim. “We anticipate this method could greatly advance how we evaluate animal health and psychological stress moving forward.”

The research findings are available in BMC Veterinary Research.

###

About OMU

Osaka Metropolitan University, one of Japan’s most prominent public institutions, is situated in Osaka. The university dedicates itself to advancing society through interdisciplinary collaboration and high-level research initiatives.

Journal

BMC Veterinary Research

DOI

10.1186/s12917-025-04919-1

Method of Research

Data and statistical analysis

Subject of Research

Animals

Article Title

AI-enhanced infrared thermography for reliable detection and spatial mapping of temperature patterns in calf eyes and muzzles

Article Publication Date

15-Jul-2025

COI Statement

The author confirms there are no commercial or financial ties that could be viewed as a potential conflict of interest related to this study.