Automated Silicon-substrate Ultra-microtome (ASUM)

ASUM is a novel mechanism based on circular silicon substrates for automatic collection of brain sections

This repository contains the comprehensive release files of the ASUM:

• Mechanical design drawings of the ASUM
• Detection and control code of the ASUM
• GUI of the ASUM

---

Mechanical design drawings of the ASUM

• Open the design drawings by UG or Solidworks

./Mechanical_drawing/Assembly_source_files-overall/装配体2(1).stp

---

Detection and control code of the ASUM (Include GUI)

Note: In this section, we assume that you are always in the directory $PROJECT_ROOT/

Compatability

Currently this repo is compatible with ubuntu16.04/ubuntu18.04, Python 3, PyQt5.

Running tutorial

1. Environment configuration (for ubuntu OS)

pip install -r requirements.txt

2. Download Pre-trained Model

• Download address: Pre-trained SSD detection model for brain sections
• Put the pre-trained model as ./Detection_and_control/ASUM-GUI-with-detection-and-control/models/SSD_sections_det.pth

3. Make sure the ASUM is installed successfully on the Ultra-microtome

• make sure the serial port communication authority of the host is provided
• make sure the electric motor has been successfully connected to computer
• make sure that the CCD camera has been successfully connected to computer

4. Run GUI

(Normal mode: run with normal mode when there is an installed ASUM device)

python3 ./Detection_and_control/ASUM-GUI-with-detection-and-control/main-asum-gui-final.py

(Debug mode: run with normal mode when there is no existing ASUM device, only the GUI will be displayed by default under the debug mode)

python3 ./Detection_and_control/ASUM-GUI-with-detection-and-control/main-asum-gui-final.py --debug

---

Declare

The Mechanical design drawings, GUI and the automatic control system of the ASUM can be used only for the purpose of academic study

Citation

@article{cheng2021automated,
  title={Automated silicon-substrate ultra-microtome for automating the collection of brain sections in array tomography},
  author={Cheng, Long and Liu, Weizhou and Zhou, Chao and Zou, Yongxiang and Hou, Zeng-Guang},
  journal={IEEE/CAA Journal of Automatica Sinica},
  volume={8},
  number={2},
  pages={389--401},
  year={2021},
  publisher={IEEE}
}