You do the science, we do the recording.

Zebrafish research today: Tremendous promise to be unlocked with behavioral studies

In medical research, and drug discovery in particular, the zebrafish (Danio Rerio) has demonstrated its extraordinary potential to accelerate research timelines while reducing costs. Medical research using the model is increasingly relying on the impressive behavior repertoire exhibited by the transparent larvae, but this requires specialized, expensive equipment that comes with many constraints that limit experiments. These limitations include:

Our technological solution: MagnaTank™ a revolutionary behavior monitoring platform

Our revolutionary MagnaTank™ behavior monitoring platform records in parallel groups of free swimming larvae in custom Petri dish-sized containers. Illuminated horizontally (darkfield illumination), larvae are recorded using our novel Remanent Imaging paradigm: they literally draw their own trajectories on the black background of the images. This reduces processing needs and data footprint, making possible real time processing of life long recordings, a prowess out of reach of any other existing technology.

Remote operation and access to the platform is made possible by its underlying software architecture and PiscisTrack™ suite of custom software. PiscisTrack™ performs all image capture, data storage and data processing tasks via a unified, consistent, cloud based architecture controlled via web browser.

Experiment-As-A-Service: WatchYourFish™ brings MagnaTank™ benefits to any lab

WatchYourFish™ service makes our revolutionary MagnaTank™ accessible and available to any zebrafish researcher. Simply send us:

We conduct experiments and provide:

Costs and delays from equipment acquisition, user training, and experiment turnaround time are removed by the WatchYourFishTM service. Initially targeting zebrafish larvae, our service allows scientists to conduct naturalistic behavior experiments not possible before and interact with their experiments and recordings in an unprecedented manner and without any training.

Contact us to learn more


The MagnaTank™ platform is designed to provide the best possible darkfield images so that motion data can be extracted with little or no image processing and that the region of interest includes the entirety of the tanks, making all organisms always visible regardless of their size.

At the same time, the platform has been conceived to free users from making any adjustments, such as centering or focusing.

The superior image capture capability is provided by a combination of unique designs that provides fine control of illumination wavelength and light field topology. A major emphasis has been put on enforcing rotational symmetry whenever possible so that experimental conditions are identical and finely controlled.

Consistency and synchronicity are enforced thanks to an unqualified commitment to rotational symmetry for all station components.

The Tank

MagnaTank™ custom tanks are designed to hold 30mL of water, provide a 5.2 mm deep water column, and enforce a meniscus free surface orthogonal to the camera recording axis.

Meniscus formation is prevented by sharp ledges on the inner and outer walls of the tank 5.2 mm from the tank bottom.

In addition, both walls are tapered to perfectly align with the camera and provides a region of interest that includes the entire water column with no dead angles for organisms to escape the field of view.

The Tank Container

The MagnaTank™ tank container surround tanks with water to provide:

  • optical sink to reduce light reflections on the tank’s outside wall
  • thermal buffer and locus to measure temperature constantly without interfering with image capture.
  • noise isolation buffer to prevent propagation to other tanks in experiments involving tapping
  • the option to float the tanks by increasing the water volume and for enhanced buffering
  • provide a locus to measure extrinsic vibrations (not created by the movements of organisms), leveraging Remanent Imaging aptitude at detecting and measuring vibrations of small water particles.

Illumination Layer

Dual illumination for centering (red) and imaging (IR) which are split (beam splitters) and reflected (mirrors) into the 4 tanks from below.

As a result, consistent, synchronized, collimated illumination is provided to each of the 4 tanks.

Capture Layer

Capture Layer

Board level cameras and narrow bandpass filters held in custom holders support capture in a narrow IR spectrum. Capture is controlled by custom software in our PiscisTrack™ suite. A centrally placed LED will illuminate experiments with the appropriate light schedule.

Remanent Imaging

Remanent Imaging

Remanent Imaging (“RI”) is a video imaging paradigm that records motion blur for improving motion detection and analysis at massively reduced computational load and data storage footprint; it is implemented by setting the camera exposure to match the time interval between frames. Darkfield illuminated RI subjects have highly visible trajectories that are continuous from image to image; stacking image sequences recapitulate what a single long exposure photo would have recorded over the sequence duration. This recapitulation power is illustrated on the right with a stack of 75 images of 12 five days post fertilization (“dpf”) zebrafish larvae spanning 20 seconds.

Because motion continuity is preserved from frame to frame, the precise fish motion is captured by the crest lines of the intensity profile of the image stack, and the rest of the image can be discarded, yielding massive data compression—10,000X in the above example. Most crucially, this compression is achieved without sacrificing time resolution: all C-starts, the sharp snapping fish escape initiations that require millisecond sampling rates, are clearly visible on long exposures. Due to this data footprint, RI can realistically be used for lifetime recording.

Integrative by design, Remanent Imaging enables another major innovation: direct measure extraction from stacked image sequences at user-specified time resolution. In traditional image processing, videos are processed frame by frame to identify and track moving objects, build their trajectories, then compile various measures; because systematic error accumulates as a result of frame by frame compilation, further treatment is required for mitigation. RI sequences can be processed in this manner to track individual fish.

PiscisTrack™ Architecture

The PiscisTrack™ software architecture, illustrated below, is designed on the following foundational principles:

  • Process data according to a dataflow model. As much as possible, recorded pixel values are processed as they are streamed from the camera rather than stored as images to be processed post-mortem after a recording is complete. This model provides support for uninterrupted, continuous recording.
  • Store images as sequences of individual frames rather than as video files.
  • Handle, store, and query analysis data along a relational database model.
  • All user interface components run in web browsers for universal accessibility.
  • PiscisTrack™ software agencies (details below) are launched as programs on physical or cloud based servers, or within a user browser by visiting a dedicated URL.

Meet Our Team

Read Full Bios