Ultrasound image dataset for ovarian follicular development detection in pigs

Description

Pigs are one of the largest scale livestock species globally. In large-scale pig farms, the reproductive management of sows is a critical control point to ensure production capacity and reduce costs. The formulation of the sow breeding management program mainly depends on the accurate assessment of the development of ovarian follicles and ovulation status in sows. A comprehensive dataset of high-resolution images, capturing various size follicles in different stages and conditions, has been assembled. The dataset includes 868 ultrasound images, categorizing follicles as four sizes: Small, Medium, Large, Preovulatory follicles in gilts and sows . This initiative aims to enhance the reproductive efficiency of sows in pig farms and facilitate research into sow reproductive physiology.

Steps to reproduce

The dataset is composed of ultrasound images of sow ovary. All these high-quality ultrasound images are collected by Honda HS-1600 V device (Honda Electronics, Tokyo, Japan). The bulk of the training data was collected over the course of four years from four different pig farms located in Beijing, China, which are the Great Wall Danyu livestock, Beijing Pig Breeding Co., Ltd. Pinggu branch, Nankou original breeding pig farm and Xishao original breeding pig farm.

Institutions

• China Agricultural University

Categories

Funders

• the National Natural Science Foundation of China

  Grant ID: 31930101
• National Center of Technology Innovation for Pigs

  Grant ID: NCTIP-XD/B03
• Ningbo Major Science and Technology Project

  Grant ID: 2021Z112
• National Key R&D Program

  Grant ID: 2022YFD1300301
• Beijing Innovation Consortium of Livestock Research System

  Grant ID: BAIC05-2024

Related Links

Licence