A low-cost, submersible, digital holographic microscope for in situ microbial imaging
Description
Microscopes used for studying environmental microbes are typically designed for use in a laboratory setting, this usually requires extracting a sample from its place of origin before examination. Here we present a low-cost submersible digital holographic microscope (DHM) designed to image small marine organisms (such as bacteria and plankton) in their natural underwater environment. The system achieves sub-micron spatial resolution and uses artificial intelligence for detection and tracking. The instrument also aims to reduce the cost of manufacturing. “Off the shelf” components were selected that provide accurate results without sacrificing data quality. The DHM itself costs under one thousand dollars, and features a low-cost high-resolution camera, the Arducam MT9J001, interfaced with our artificial intelligence via a Raspberry Pi and Google Coral Tensorflow accelerator. Preliminary testing indicates data acquisition in our submersible for up to 3 hours at depths upwards of 30 meters. Additionally, our artificial intelligence is currently capable of tracking up to 10 areas of interest in a fraction of a second via the neural net used in combination with the Google Coral.
A low-cost, submersible, digital holographic microscope for in situ microbial imaging
Microscopes used for studying environmental microbes are typically designed for use in a laboratory setting, this usually requires extracting a sample from its place of origin before examination. Here we present a low-cost submersible digital holographic microscope (DHM) designed to image small marine organisms (such as bacteria and plankton) in their natural underwater environment. The system achieves sub-micron spatial resolution and uses artificial intelligence for detection and tracking. The instrument also aims to reduce the cost of manufacturing. “Off the shelf” components were selected that provide accurate results without sacrificing data quality. The DHM itself costs under one thousand dollars, and features a low-cost high-resolution camera, the Arducam MT9J001, interfaced with our artificial intelligence via a Raspberry Pi and Google Coral Tensorflow accelerator. Preliminary testing indicates data acquisition in our submersible for up to 3 hours at depths upwards of 30 meters. Additionally, our artificial intelligence is currently capable of tracking up to 10 areas of interest in a fraction of a second via the neural net used in combination with the Google Coral.