Automated Spine Segmentation and Human-Computer Interface System

The code was written by Chein Chang, Darren Lee, Jamie Wang and Yoyo Zheng.

Outline

1. Introduction
2. Set up
3. Qt Note
4. Reference

Introduction

Motivation

Design an artificial intelligence algorithm that can predict the probability of a patient who has undergone spinal fusion surgery experiencing a vertebral fracture within one year, and provide a simple and clear interface system (GUI) for medical staff to create a visualized prediction model for the convenience of researchers to use and optimize.

Demonstrations

A DICOM contain an advanced function - Spine Segmentation.

More Detailed

Set up

1. Create a new enviroment

conda create -n qt_env python=3.8

2. Activate the new enviroment

activate qt_env

3. Install PyQt5

pip install pyqt5
pip install pyqt5-tools

4. Check

import PyQt5

Qt Note

Qt Designer to Python Code

link

• Shell

pyuic5 -o ____.py ____.ui

qrc file to py file

• Shell

pyrcc5 ____.qrc -o ____.py

Reference

1. zhixuhao, Unet using Tensorflow
2. DICOM影像格式影像格式標籤
3. PyQt API
4. Find vertices in image of convex polygon
5. Kim YJ, Ganbold B, Kim KG. Web-based spine segmentation using deep learning in computed tomography images. Healthc Inform Res 2020;26:61-7.
6. Asian Spine Journal 2020 - Review of the use of AI in spinal Diseases
7. Computer Methods and Programs in Biomedicine 2020 - Automatic detection and segmentation of lumbar vertebra
8. Global Spine Journal 2020 - Automated Measurement of Lumbar Lordosis
9. Osteoporosis International 2019 - Prediction Of Lumbar Vertebral Strength
10. Better Height Restoration, Greater Kyphosis Correction, and Fewer Refractures of