Using Real-Time-Ultrasound to estimate body composition of tilapia (Oreochromis niloticus Linnaeus)

• Main Author: Gonçalves, Catarina Alexandra Peixoto
• Publication Date: 2018
• Format: Master thesis
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/10348/9079
• topic: tilapia; fillet yield; real-time ultrasonography

The knowledge of fish composition using non-destructive methods is an important issue for nutrition, genetics and physiology fields. In this way this study was undertaken with the objective to predict tilapia fillet (Oreochromis niloticus) using real-time ultrasonography and image analysis. A heterogeneous group of tilapia (Oreochromis niloticus) was used for this experiment (n= 51, mean weight= 167.35 ± 56.09 g, range from 47.35 to 305.07 g). All animals were under mild anaesthesia, during the scanning procedure. They were scanned with a realtime ultrasound (RTU), equipped with a linear array transducer of 7.5 MHz. The RTU images were taken in six cross-sectional slices. The dorsal fillet area was determined using Fiji software. The RTU fillet weight was determined by taking into account the value of 1.0798 g/cm3 for tilapia fish muscle specific density. The correlation between determined fillet weight and volume and the RTU equivalent measurements is significant (r ranged between 0.94 and 0.96; P<0.001). The linear model can explain 89% of the variation of the fillet volume.In conclusion, the obtained results suggest that RTU and image analysis is a viable nondestructive and non-invasive method for fillet volume evaluation in Oreochromis niloticus.