Mass Surveilance of C. elegans—Smartphone-Based DIY Microscope and Machine-Learning-Based Approach for Worm Detection

Abstract

The nematode Caenorhabditis elegans (C. elegans) is often used as an alternative animal model due to several advantages such as morphological changes that can be seen directly under a microscope. Limitations of the model include the usage of expensive and cumbersome microscopes, and restrictions of the comprehensive use of C. elegans for toxicological trials. With the general applicability of the detection of C. elegans from microscope images via machine learning, as well as of smartphone-based microscopes, this article investigates the suitability of smartphone-based microscopy to detect C. elegans in a complete Petri dish. Thereby, the article introduces a smartphone-based microscope (including optics, lighting, and housing) for monitoring C. elegans and the corresponding classification via a trained Histogram of Oriented Gradients (HOG) feature-based Support Vector Machine for the automatic detection of C. elegans. Evaluation showed classification sensitivity of 0.90 and specificity of 0.85, and thereby confirms the general practicability of the chosen approach.

Mask R-CNN Based C. Elegans Detection with a DIY Microscope

“…Based on the general aim to develop a do-it-yourself microscope setup to capture complete Petri dishes and identify multiple C. elegans, we faced challenges with our previous smartphone-based system [13], such as variations of focal lengths and firmware (applying various pre-processing steps and autofocus methods) among mobile phonecameras and computational-complexity of the applied Support Vector Machine. Thus, in this article, we propose a low-cost DIY microscope with a fixed camera and a controlling unit (based on a Raspberry Pi) to guarantee standardized image acquisition conditions and overcome the necessity of readjustment per camera model.…”

“…The previously proposed smartphone based C. elegans tracker [13] encountered significant issues, including memory and compute-power constraints of the smartphone. With the chosen sliding window approach and a 50% overlap between subsequent windows, a reduced image resolution of 3024 × 4032 px was shown to be maximally practical for smartphones' processing capabilities, while a higher image resolution was expected to result in higher accuracy.…”

“…With these components, an even higher cost-efficiency is achieved since the necessity of a smartphone is omitted, and the overall material costs of this system range around EUR 200. 1) the camera module with lenses, (2) the agar plate with C. elegans, (3) the lighting system, reused from [13] and (4) a movable stage to determine the correct distance to the lenses and thus guarantee sharp images (only in this prototype system, distance will be fixed later). The Pi saves taken images on itself.…”

“…To alleviate these problems, we proposed an approach [13] that utilizes low-cost smartphones within a do-it-yourself microscope setup to capture complete Petri dishes and identify multiple C. elegans via a SVM [14]. This approach enabled the capturing of complete Petri dishes with nematodes with a recall of 0.90 and specificity of 0.85 via the included illumination, optics, and the smartphone camera.…”

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

“…WorMachine [ 15 ] is a tool that uses machine learning techniques for the identification, sex classification, and extraction of different phenotypes of C. elegans . A support vector machine (SVM) was used for the automatic detection of C. elegans via a smartphone app in [ 16 ]. A method that classifies different strains of C. elegans using convolutional neural networks (CNN) was presented in [ 17 ].…”