Deep learning for robust and flexible tracking in behavioral studies forC. elegans

Abstract

Robust and accurate behavioral tracking is essential for ethological studies. Common methods for tracking and extracting behavior rely on user adjusted heuristics that can significantly vary across different individuals, environments, and experimental conditions. As a result, they are difficult to implement in large-scale behavioral studies with complex, heterogenous environmental conditions. Recently developed deep-learning methods for object recognition such as Faster R-CNN have advantages in their speed, accuracy, and robustness. Here, we show that Faster R-CNN can be employed for identification and detection of Caenorhabditis elegans in a variety of life stages in complex environments. We applied the algorithm to track animal speeds during development, fecundity rates and spatial distribution in reproductive adults, and behavioral decline in aging populations. By doing so, we demonstrate the flexibility, speed, and scalability of Faster R-CNN across a variety of experimental conditions, illustrating its generalized use for future large-scale behavioral studies.

A Review on High Throughput Phenotyping for Vegetable Crops

“…As a result of this difficulty, the genetic and functional foundations of root phenes are less established than those of aboveground phenes [68,69]. To close this "phenotyping gap," a shift away from traditional phenotyping toward image-based phenotyping has occurred, which enables relatively high throughput while maintaining root measurement accuracy [70]. Numerous platforms make use of two-dimensional imaging via cameras and propagate plants via seedlings on agar plates, germination paper, or fabric cloth in bins [71,72].…”

High-throughput imaging of Caenorhabditis elegans aging using collective activity monitoring

“…Recent studies have shown that machine learning and convolutional neural network (CNN) based approaches can reliably identify worms in variable quality images (Hakim et al, 2018;Bornhorst et al, 2019;Bates et al, 2021). We trained a CNN to differentiate worms from background in 20x20 pixel kernels extracted from our images (see above).…”

Towards Lifespan Automation for Caenorhabditis elegans Based on Deep Learning: Analysing Convolutional and Recurrent Neural Networks for Dead or Live Classification

“…Recently, [22,23] used different convolutional neural network models to estimate the physiological age of C. elegans. A method for the identification and detection of C. elegans based on Faster R-CNN was proposed in [24]. Lastly, [25] developed a CNN that classifies young adult worms into short-lived and long-lived.…”