AI listens to pig coughs to catch disease early on Korean farms

Kim Seong-guk : Aug. 13, 2026
https://mbiz.heraldcorp.com/article/10839576

Technology that analyzes pig farm locations, weather data and disease records to predict livestock epidemic spread risk in real time — and an AI system that listens to pigs coughing to issue early disease warnings — has arrived on Korean farms. Efforts to apply AI to agri-food public data to tackle pressing challenges in agriculture and rural communities, from livestock disease and crop pests to drought and produce retail, are gaining momentum.

The Ministry of Agriculture, Food and Rural Affairs held an awards ceremony Thursday at the Korea Racing Authority auditorium for the 11th Agriculture and Rural Public Data + AI Startup Competition, selecting 18 winning entries — nine in the product and service development category and nine in the idea planning category. The total purse is 56 million won ($39,600).

This year’s competition drew 290 entries, up 105, or 56.8 percent, from 185 last year. Eight affiliated organizations took part in running the event, including the Ministry of Agriculture, Food and Rural Affairs, the Rural Development Administration, the Korea Rural Community Corporation and the Korea Agro-Fisheries and Food Trade Corporation (aT), as well as the Korea Racing Authority.

A defining feature of this year’s winning entries was their focus on directly solving problems in the field by combining AI, farm mapping data and agri-food public datasets. The range of applications also broadened, spanning livestock epidemic prevention, pest control, agricultural drought forecasting, produce retail, agri-solar power and smart livestock farming.

The grand prize in the product and service development category went to team “zyra” for its “AI-based livestock epidemic pre-emptive biosecurity navigation” service. The prize carries 7 million won.

The service establishes domestic diagnostic standards for porcine reproductive and respiratory syndrome based on locally identified genetic variant samples, then combines disease occurrence records, farm locations and weather data to diagnose and predict in real time the epidemic spread risk at pig farms nationwide. Its core value lies in shifting from reactive response to proactive biosecurity support.

An AI tool for responding to livestock disease also took the grand prize in the idea planning category. The winning entry, submitted by team “HanVet AI,” is a pig barn cough AI monitoring and early-warning system.

The AI learns to recognize pig coughs, analyzes the direction in which coughing spreads inside a barn alongside ventilation patterns, and checks for disease outbreaks within a 30-kilometer radius to assess risk. When readings exceed a set threshold, the system automatically sends an alert to the responsible veterinarian along with an actual audio clip of the coughing. The entry was highly rated for enabling livestock farmers to detect and respond to animal disease early.

AI-driven ideas for addressing climate change also featured among the winners. Team “Groundwater Forecast” won an excellence award for proposing a system that uses monitoring network water levels, public well data and reservoir storage rates to predict agricultural drought 15 days in advance.

A service for assessing the economic viability and feasibility of agri-solar power projects before launch also made the winners’ list. Team “Sunshine Farmer” developed an AI diagnostic service that combines address-based farm mapping data with agricultural statistics, land-use regulations and solar irradiance figures to determine whether an agri-solar project is viable for a given plot of farmland.

The grand prize in the farm-mapping data category went to team “Cheolbuji” for “All-Farm,” a digital twin farming simulator that integrates crop cultivation data and weather information from farm maps with wholesale and retail produce prices. The platform lets new and experienced farmers virtually experience and analyze farm management outcomes under a range of conditions before they begin actual cultivation.

Other winning entries included a platform that uses AI to identify the optimal wholesale market outlet for agricultural produce, a system that detects irregular distribution of goods by origin by analyzing import volumes and auction prices, and a platform that simultaneously optimizes carbon output and revenue while forecasting rice yields.

The ministry plans to support winners with tailored public data resources, AI education, and marketing and investment consulting to help turn their ideas into startups and commercial ventures. The top four entries will also advance to the integrated national finals of the Ministry of Interior and Safety’s 14th Pan-Government Public Data and AI Startup Competition, to be held in September and October.

Lee Si-hye, the ministry’s director general for agri-industry innovation policy, said high-quality data is the core resource for the digital transformation of agriculture and rural communities. “We will continue to expand the opening and use of agri-food public data so that it can resolve diverse problems in the field and give rise to new services and industries,” she said.