AGAR a microbial colony dataset for deep learning detection

Abstract

The Annotated Germs for Automated Recognition (AGAR) dataset is an image database of microbial colonies cultured on agar plates. It contains 18 000 photos of five different microorganisms as single or mixed cultures, taken under diverse lighting conditions with two different cameras. All the images are classified into countable, uncountable, and empty, with the countable class labeled by microbiologists with colony location and species identification (336 442 colonies in total). This study describes the dataset itself and the process of its development. In the second part, the performance of selected deep neural network architectures for object detection, namely Faster R-CNN and Cascade R-CNN, was evaluated on the AGAR dataset. The results confirmed the great potential of deep learning methods to automate the process of microbe localization and classification based on Petri dish photos. Moreover, AGAR is the first publicly available dataset of this kind and size and will facilitate the future development of machine learning models. The data used in these studies can be found at https://agar.neurosys.com/.

Generation of microbial colonies dataset with deep learning style transfer

“…Finally, a deep learning detector was trained entirely on this synthetic dataset and evaluated on the test part of the higher-resolution AGAR subset. This way, the obtained results are comparable with the ones from [12], obtained for the same DL detector architecture, but trained entirely on the real dataset (i.e. training part of the higher-resolution subset) containing over 7k images.…”

“…The use of neural networks also makes it possible to automate industrial processes such as counting microbial colonies on Petri dishes, which is an important step in the microbiological laboratory to evaluate the cleanliness of the samples. Traditionally, the counting task is done manually or semi-automatically 10,11 , but recent studies suggest that DL-based methodology will accelerate the process [12][13][14][15][16] .…”

“…To evaluate the number of microbial colonies grown on a Petri dish, neural network models called detectors may be applied 12 . The most representative group of detectors are two-stage models.…”

“…However, despite the broad possibilities and the recent interest in the topic [12][13][14] , the application of DL-based methods in medical, pharmaceutical, or food industries still faces some challenges. DL algorithms are quite efficient when learning from large amounts of data, whose collection and annotations require a lot of time and financial support.…”

“…We used the higher-resolution subset of the recently introduced AGAR microbial colony dataset 12 to conduct a synthetic dataset generation experiment. We randomly selected 100 annotated images from this subset and used them to feed the colony extraction part.…”