Automatic counting of retinal ganglion cells in the entire mouse retina based on improved YOLOv5

Optimizing retinal ganglion cell nuclear staining for automated cell counting

The retinal ganglion cells (RGCs) serve as the critical pathway for transmitting visual information from the retina to the brain, yet they can be dramatically impacted by diseases such as glaucoma. When investigating disease processes affecting RGCs in mouse models, accurately quantifying affected cells becomes essential. However, the use of pan RGC markers like RBPMS or THY1 presents challenges in accurate total cell counting. While Brn3a serves as a reliable RGC nuclear marker for automated counting, it fails to encompass all RGC subtypes in mice. To address this limitation and enable precise automated counting, our research endeavors to develop a method for labeling nuclei in all RGC subtypes. Investigating RGC subtypes labeled with the nuclear marker POU6F2 revealed that numerous RGCs unlabeled by Brn3a were, in fact, labeled with POU6F2. We hypothesize that using antibodies against both Brn3a and POU6F2 would label virtually all RGC nuclei in the mouse retina. Our experiments confirmed that staining retinas with both markers resulted in the labeling of all RGCs. Additionally, when using the cell body marker RBPMS known to label all mouse RGCs, all RBPMS-labeled cells also exhibited Brn3a or POU6F2 labeling. This combination of Brn3a and POU6F2 antibodies provides a pan-RGC nuclear stain, facilitating accurate automated counting by labeling cell nuclei in the retina.

From beasts to bytes: Revolutionizing zoological research with artificial intelligence

Since the late 2010s, Artificial Intelligence (AI) including machine learning, boosted through deep learning, has boomed as a vital tool to leverage computer vision, natural language processing and speech recognition in revolutionizing zoological research. This review provides an overview of the primary tasks, core models, datasets, and applications of AI in zoological research, including animal classification, resource conservation, behavior, development, genetics and evolution, breeding and health, disease models, and paleontology. Additionally, we explore the challenges and future directions of integrating AI into this field. Based on numerous case studies, this review outlines various avenues for incorporating AI into zoological research and underscores its potential to enhance our understanding of the intricate relationships that exist within the animal kingdom. As we build a bridge between beast and byte realms, this review serves as a resource for envisioning novel AI applications in zoological research that have not yet been explored.

Improved UAV-to-Ground Multi-Target Tracking Algorithm Based on StrongSORT

Unmanned aerial vehicles (UAV) are essential for aerial reconnaissance and monitoring. One of the greatest challenges facing UAVs is vision-based multi-target tracking. Multi-target tracking algorithms that depend on visual data are utilized in a variety of fields. In this study, we present a comprehensive framework for real-time tracking of ground robots in forest and grassland environments. This framework utilizes the YOLOv5n detection algorithm and a multi-target tracking algorithm for monitoring ground robot activities in real-time video streams. We optimized both detection and re-identification networks to enhance real-time target detection. The StrongSORT tracking algorithm was selected carefully to alleviate the loss of tracked objects due to factors like camera jitter, intersecting and overlapping targets, and smaller target sizes. The YOLOv5n algorithm was used to train the dataset, and the StrongSORT tracking algorithm incorporated the best-trained model weights. The algorithm's performance has greatly improved, as demonstrated by experimental results. The number of ID switches (IDSW) has decreased by sixfold, IDF1 has increased by 7.93%, and false positives (FP) have decreased by 30.28%. Additionally, the tracking speed has reached 38 frames per second. These findings validate our algorithm's ability to fulfill real-time tracking requisites on UAV platforms, delivering dependable resolutions for dynamic multi-target tracking on land.

Pan-retinal ganglion cell markers in mice, rats, and rhesus macaques

Univocal identification of retinal ganglion cells (RGCs) is an essential prerequisite for studying their degeneration and neuroprotection. Before the advent of phenotypic markers, RGCs were normally identified using retrograde tracing of retinorecipient areas. This is an invasive technique, and its use is precluded in higher mammals such as monkeys. In the past decade, several RGC markers have been described. Here, we reviewed and analyzed the specificity of nine markers used to identify all or most RGCs, i.e., pan-RGC markers, in rats, mice, and macaques. The best markers in the three species in terms of specificity, proportion of RGCs labeled, and indicators of viability were BRN3A, expressed by vision-forming RGCs, and RBPMS, expressed by vision- and non-vision-forming RGCs. NEUN, often used to identify RGCs, was expressed by non-RGCs in the ganglion cell layer, and therefore was not RGC-specific. γ-SYN, TUJ1, and NF-L labeled the RGC axons, which impaired the detection of their somas in the central retina but would be good for studying RGC morphology. In rats, TUJ1 and NF-L were also expressed by non-RGCs. BM88, ERRβ, and PGP9.5 are rarely used as markers, but they identified most RGCs in the rats and macaques and ERRβ in mice. However, PGP9.5 was also expressed by non-RGCs in rats and macaques and BM88 and ERRβ were not suitable markers of viability.