Shrimp classification for white spot syndrome detection through enhanced gated recurrent unit-based wild geese...

Area of Science:

• Aquaculture
• Veterinary Virology
• Machine Learning in Biology

Background:

• White Spot Syndrome Virus (WSSV) is a major viral pathogen in cultivated shrimp, causing up to 100% mortality within days.
• WSSV affects various crustacean hosts, predominantly commercially farmed marine shrimp across all age groups.
• Understanding WSSV pathogenicity is complex, with ongoing research into genomic variations and protein factors.

Purpose of the Study:

• To develop and evaluate a novel deep learning methodology for classifying WSSV-infected shrimp.
• To accurately distinguish between healthy and WSSV-affected shrimp using advanced image analysis techniques.
• To improve early detection and management of WSSV outbreaks in aquaculture.

Main Methods:

• Data collection from shrimp farms and online sources, followed by image pre-processing using the Local Binary Pattern (LBP) technique.
• Image segmentation via the TGVFCMS approach and feature extraction using the Probabilistic Linear Discriminant Analysis (PLDA) technique.
• Classification of shrimp using an Enhanced Gated Recurrent Unit (EGRU) model, with parameter tuning by a wild Grey Wolf Optimizer (GWO) algorithm for accuracy maximization.

Main Results:

• The proposed deep learning methodology demonstrated high accuracy in classifying shrimp as either healthy or WSSV-affected.
• Performance indicators, particularly accuracy, were compared favorably against various conventional methods.
• The study successfully identified WSSV illness in shrimp using the developed computational approach.

Conclusions:

• The novel deep learning approach offers a robust and accurate solution for WSSV detection in shrimp.
• This methodology can significantly aid in the early identification of WSSV, mitigating economic losses in shrimp farming.
• Further research can explore broader applications of this technique for other aquaculture diseases.