Page 106 - AI for Good Innovate for Impact

AI for Good Innovate for Impact

2 Use Case Description

2�1 Description

The AI-powered equine health management platform (TT-care Equine) [1] is designed to solve
two critical problems in equine welfare: horse identification and early detection of lameness.
Horses, which have long had a close relationship with humans, are vulnerable to risks such as
lameness [2], and misidentification and mismanagement are critical to the preservation of the
species. Traditional methods to address these issues rely heavily on manual observation and
subjective assessment, resulting in inefficient, inconsistent, and delayed interventions.

The proposed platform integrates a smartphone-based platform with advanced AI algorithms
to provide two key features. First, the horse identification module uses a common smartphone
camera to capture unique physical patterns in images of the horse's face, body, and legs. AI
models automatically extract the unique features and compare them to a digital passport
system to verify the horse's identity. Digitizing the horse registration process increases data
accuracy, reduces administrative burden, and improves accessibility for horse owners and
veterinarians.

Second, the lameness detection module analyzes gait patterns by applying AI-based pose
estimation techniques from video footage taken with a smartphone. The system detects subtle
movement abnormalities that may indicate the early stages of lameness, providing objective
data to support a veterinary diagnosis. Unlike traditional methods that require specialized
equipment such as force meters, treadmills, and inertial measurement units (IMUs), this
approach uses widely available mobile devices, significantly lowering the barrier to regular
health monitoring. The platform also facilitates preventive health care strategies by enabling
longitudinal tracking of gait data.

The proposed solution addresses key limitations of existing practices, including reliance on
subjective judgment [3], high cost of measurement equipment, and lack of continuous data
collection and analysis systems. It also contributes to animal welfare and sustainable livestock
production by facilitating ongoing and proactive equine health management.

Use Case Status: Pilot

Partners: Korea Racing Authority (KRA)

2�2 Benefits of use case

This attempt to combine horse management and AI has the potential to improve the welfare of
horses through ongoing population protection and health management, while also contributing
to sustainable livestock farming.

2�3 Future Work

Building on the current platform, AIforPet will focus on three key areas to further advance animal
welfare through AI-powered solutions.

First, the scale and diversity of datasets will be expanded by collecting images and gait
information from a broader range of horse breeds, age groups, and activity types. This

70