Recent Advances in Bioimage Analysis Methods for Detecting Skeletal Deformities in Biomedical and Aquaculture Fish Species

Abstract

Detecting skeletal or bone-related deformities in model and aquaculture fish is vital for numerous biomedical studies. In biomedical research, model fish with bone-related disorders are potential indicators of various chemically induced toxins in their environment or poor dietary conditions. In aquaculture, skeletal deformities are affecting fish health, and economic losses are incurred by fish farmers. This survey paper focuses on showcasing the cutting-edge image analysis tools and techniques based on artificial intelligence that are currently applied in the analysis of bone-related deformities in aquaculture and model fish. These methods and tools play a significant role in improving research by automating various aspects of the analysis. This paper also sheds light on some of the hurdles faced when dealing with high-content bioimages and explores potential solutions to overcome these challenges.

Keywords: aquaculture; bioimage analysis; biomedical; computer vision; fish; image processing; medaka; seabream; skeletal deformities; zebrafish.

Figure 1

A typical CNN architecture for image classification tasks.

Figure 2

Different phenotypes in zebrafish tail. Larvae were imaged live under a dissecting microscope under transmitted light illumination: (A) Downward curved tail; (B) Upward curved tail; (C) Short tail; (D) Normal phenotype [29].

Figure 3

Segmentation results obtained with deep learning on a zebrafish image [57]. Zebrafish larvae were stained for mineralized tissues using Alizarin red and imaged using a fluorescence microscope: (Left) Input image; (Right) Analyzed image.

Figure 4

Six landmark locations on a medaka juvenile stained with Alizarin red for calcified tissues from microscopy dataset [59]. These landmarks are as follows: 1: rostral tip of the premaxilla; 2: base of the neural arch of the 1st (anteriormost) abdominal vertebra bearing a rib; 3: base of the neural post-zygapophyses of the first hemal vertebra (viz., vertebra with hemal arch closed by a hemaspine); 4: base of the neural post-zygapophyses of the first preural vertebra; 5: base of the neural post-zygapophyses of the preural-2 vertebra; 6: posteriormost (caudad) ventral extremity of the hypural 1.

Figure 5

Sample image of zebrafish with 25 landmark locations. Alizarin red staining was performed on fixed larvae and imaged under a dissecting microscope. The landmark locations are annotated as follows: 1 and 24: Maxilla; 2 and 23: Branchiostegal ray 2; 3 and 11: Opercle; 4, 12, 13, and 14: Cleithrum; 5 and 19: Anguloarticular; 6 and 25: Ceratobranchial; 7 and 8: Hyomandibular; 9 and 20: Entopterygoid; 10: Notochord; 21, 15, and 18: Parasphenoid; 17 and 22: Dentary; 16: showing anterior end marking.

Figure 6

Sample X-ray image of pre-ongrowing phase gilthead seabream with 19 landmarks [59]. The image was obtained in air, on euthanized fish. The landmark locations are as follows: A: frontal tip of premaxillary; B: rostral head point in line with the eye center; C: dorsal head point in line with the eye center; D: dorsal extremity of the 1st predorsal bone; E: edge between the dorsal 1st hard ray pterygophore and hard ray; F: edge between the dorsal 1st soft ray pterygophore and soft ray; G: edge between the dorsal last soft ray pterygophore and soft ray; H: dorsal concave inflexion point of caudal peduncle; I: middle point between the bases of hypurals 2 and 3 (fork); L: ventral concave inflexion point of caudal peduncle; M: edge between the anal last pterygophore and ray; N: edge between the anal 1st ray pterygophore and ray; O: insertion of the pelvic fin on the body profile; P: preopercle ventral insertion on body profile; Q: frontal tip of dentary; R: neural arch insertion on the 1st abdominal vertebral body; S: neural arch insertion on the 1st hemal vertebral body; T: neural arch insertion on the 6th hemal vertebral body; U: between the pre- and post-zygapophyses of the 1st and 2nd caudal vertebral bodies.