**AWARDS ISSUED PRIOR TO JANUARY 20, 2025, WERE FUNDED UNDER PREVIOUS ADMINISTRATIONS AND MAY NOT REFLECT THE PRIORITIES AND POLICIES OF THE CURRENT ADMINISTRATION.** PRE-WEANING MORTALITY HAS DETRIMENTAL EFFECTS ON BOTH PIGLET WELL-BEING AND ECONOMIC RETURNS TO U.S. PORK PRODUCERS. SOW CRUSHING ACCOUNTS FOR ROUGHLY 50% OF THE TOTAL PRE-WEANING MORTALITY. DESPITE THE WIDELY APPRECIATED MAGNITUDE OF THIS PROBLEM, LITTLE PROGRESS HAS BEEN MADE IN THE PAST FEW DECADES IN UNDERSTANDING MATERNAL BEHAVIOR SUCH AS POSTURAL CHANGE SEQUENCES BETWEEN SOWS WITH HIGH AND LOW PIGLET CRUSHING RATES. THE VARIATION IN THESE BEHAVIORAL SEQUENCES CAN BE THE CRITICAL PHENOTYPES RELATED TO A SOW'S MOTHERING ABILITY, ONE OF THE CRITICAL FACTORS DETERMINING PIGLET PRE-WEANING MORTALITY.THE OVERALL GOAL OF THIS PROJECT IS TO DEVELOP AND DEPLOY A CYBERINFRASTRUCTURE FRAMEWORK TO SUPPORT THE AUTOMATIC IDENTIFICATION AND ANALYSIS OF SOW POSTURE AND PIGLET ACTIVITY AND SPACE UTILIZATION. WE PROPOSE TO 1) ESTABLISH A CLOUD-BASED SERVER TO SUPPORT THE CURATION OF SOW'S AND PIGLET'S IMAGE AND PRODUCTION DATA; 2) DEVELOP DEEP LEARNING MODELS FOR AUTOMATIC BEHAVIORAL PHENOTYPES IDENTIFICATION FROM THE SEQUENTIAL IMAGERY, AND 3) CREATE A MOTHERING ABILITY INDEX (MAI) TO CLASSIFY A SOW'S ABILITY TO SUCCESSFULLY RAISE A LITTER BASED ON STATISTICAL MODELING WITH THE IDENTIFIED BEHAVIORAL PHENOTYPES AND PRODUCTION RECORDS.THIS FRAMEWORK WILL PROVIDE A WAY TO MORE EFFICIENTLY AND EFFECTIVELY STUDY ANIMAL BEHAVIORS AND AID THE ENGINEERING DESIGN AND GENETICS BREEDING. UPON COMPLETING THIS WORK, A SET OF DATA SCIENCE APPLICATIONS WILL BE AVAILABLE TO THE SCIENTIFIC COMMUNITY AND CAN BE ADAPTED FOR GENERAL ANIMAL POSTURE, ACTIVITY, AND SPACE UTILIZATION STUDIES. ADDITIONALLY, A LABELED DATASET OF SOW AND PIGLET BEHAVIORS WILL BE MADE AVAILABLE FOR FURTHER MODEL DEVELOPMENT. THESE WILL BE VALUABLE BEYOND THE LIFETIME OF THIS PROJECT.