Skip to content

BiAPoL/microscope-gym

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Microscope Gym

Installation

create a conda environment and install this repository

conda create -n microscope-gym-api python=3.9
conda activate microscope-gym-api
pip install git+https://gitlab.mn.tu-dresden.de/bia-pol/microscope-gym.git

if you want to run the notebooks from the documentation, it is recommended to install devbio-napari into the same environment:

conda install mamba
mamba install devbio-napari

Code structure

flowchart TD
    A[Smart feature] --> |uses| B(Microscope Gym API)
    AA[Smart feature] --> |uses| B
    AAA[Smart feature] --> |uses| B
    AAAA[Smart feature] --> |uses| B
    B -.-> |implemented by| C[mock_scope]
    B -.-> |implemented by| D[Vendor 1 API adapter]
    B -.-> |implemented by| E[Vendor 2 API adapter]
    D --> |uses| G{{Microscope Vendor 1 API}}
    E --> |uses| H{{Microscope Vendor 2 API}}
Loading

microscope GYM API

defines interfaces by inheriting from abc.abc for example:

from abc import ABC

class Stage(ABC):
    @property
    @abstractmethod
    def x_position(self):

    @abstractmethod
    move_x_to(target_x_position: float):

    @abstractmethod
    move_x_by(relative_x_position: float):

Versioned interface

as few setter methods as possible

People that are involved

  • Jamie White
  • Hugo Sebastiao
  • Marcus Jahnel
  • Robert Haase
  • Bert Nitzsche

People that might be interested to get involved

  • Honki Moon (Java smart microscope software)
  • Benedict Diederich (from UC2 microscope)

Next steps

  • Schedule Meeting with Marcus and Hugo
  • Create GitHub repository
  • order open uc2 microscope
  • write microscope emulator
  • write notebook explaining basic imaging api
  • write a notebook that uses napari to train an apoc model to identify microtubule crossings in the sample data
  • write smart feature that uses a trained apoc model generated in napari to find and imag certain features with the mock scope and then images a z-stack at the identified positions
  • create napari plugin that implements the above feature in napari

Example microscope APIs

pycromanager (python API for micromanager) possible template for which kind methods to implement openUC2 REST API Natari

python microscope (bad example)

About

An open python smart-microscopy API

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •  

Languages