作者
Filipe Soares,Andreia Raposo,Rodrigo Mendes,Marina Luise Viola de Azevedo,Jorge Dias,Ângelo Nobre,Luís E.C. Conceição,Tomé S. Silva
摘要
ficoEst – Fish Composition Estimator is a public web tool to estimate the whole-body proximate composition of farmed fish (https://webtools.sparos.pt/ficoest/). The tool was designed for researchers in fish nutrition and fish farmers, and is available for six commercially relevant species: gilthead seabream (Sparus aurata), European seabass (Dicentrarchus labrax), meagre (Argyrosomus regius), rainbow trout (Onchorhynchus mykiss), Atlantic salmon (Salmo salar), and Nile tilapia (Oreochromis niloticus). ficoEst uses three different types of mathematical models (BC1, BC2, and BC3) to estimate the body composition of fish in terms of crude protein, crude lipids, water, ash, phosphorus, and energy. The models differ in the input data used to perform the estimates. BC1 models consider only body weight, BC2 models consider both body weight and water, and BC3 models consider body weight, water, and ash as inputs. The model evaluation results demonstrate that considering water and ash as additional inputs to body weight (BC3 models) significantly improves the accuracy in predicting some body composition components, such as crude lipids (e.g., up to 67.9 % and 28.1 % more accurate, compared to BC1 and BC2 models, respectively, depending on the species considered). ficoEst can be used as a complementary tool to analytical methods to obtain additional information about fish body composition. As a public web tool, ficoEst has the potential to be a valuable resource for researchers and fish farmers interested in estimating the body composition of farmed fish.