Swine Ultrasonography Numerical Modeling for Pregnancy Diagnosis and Prediction of Litter Size

Abstract

Early and reliable diagnosis of pregnancy in sows is considered one of the most crucial parameters for the proper management in pig farming. Reproductive activity of the sows is directly correlated with the overall productivity of the enterprise. The best and most precise methodology to diagnose pregnancy in farm animals, including pigs, is ultrasonography. For getting optimal outputs of real-time ultrasonography application in pigs, both gestation status and litter size of sows is of high importance. Hence, a detailed evaluation and careful reading of the retrieved ultrasonic image could serve to the prediction of the sow’s litter size, providing valuable information to the producer. In the present article we studied the ultrasonic typification of sows in an effort to develop a methodology for the precise pregnancy diagnosis and the prediction of the litter size. To achieve this goal, 1214 matings and 1010 pregnancies were recorded and evaluated over a period of 20 months, in a pig unit in Northern Greece. Two thorough scans were performed on each sow, classifying the ultrasonic image in a scale of 0 to 10, 0 being ‘non pregnant’ and 10 given to the most reassuring image of pregnancy. Evaluations took place in a wide range of time since mating, hence between days 16 (the earliest first scan) and 68 (the latest that a second scan was performed) and the mean values obtained were 8,36 and 8,83 for the first and second scan respectively. Furthermore, the litter size of each subsequent parturition was recorded (mean: 12.22) and correlated to the score of each of the two scans performed during pregnancy. The results showed that more reassuring images and higher scores for both examinations of pregnancy were achieved with increasing time interval from mating (P<0.05). It is therefore suggested that a detailed real-time ultrasonic scanning, can provide a very useful tool, not only for pregnancy diagnosis, but also for the prediction of litter size and eventually the precise management of pregnant sows.