Artificial Intelligence

Basic Imaging Processing and Artificial Intelligence in Radiology

Duration: 120 min    Date & Time: Oct9,2019 , 15:00 – 18:00    Location:  Kempinski AlOthman Hotel

Workshop Faculty:

 Metab Alkubeyyer, MD ( Consultant Body MRI and Imaging Informatics)

King Khalid University Hospital, Riyadh , Saudi Arabia

King Saud University Medical City, Riyadh , Saudi Arabia

Target Audience: 

Radiologist, Radiographer,  Image cancer researcher , Medical Physicist, quality officer, scientist researcher

Prerequisites:

Bring your own laptop and download ImageJ or FIJI (Fiji Is Just ImageJ with extra plugins)

https://imagej.nih.gov/ij/download.html ( for image processing only )

http://fiji.sc (for image processing and machine learning )

https://www.cs.waikato.ac.nz/ml/weka/downloading.html ( for machine learning and artificial intelligence only)

 

Workshop Description:

Open source software has had an expanding role in the development of image analysis routines for radiology research applications. Compared to commercial proprietary software, open source software is often more adaptable and economical. The purpose of this workshop is to familiarize participants with the features of ImageJ/FIJI , an open source image processing program The utility of ImageJ for performing automated image filtration, segmentation and registration tasks that are repeatable and
can be tailored using information on the DICOM headers of images will be discussed. Case studies using clinical images will be used to demonstrate the capabilities of ImageJ. These demonstrations will also show how the capabilities of this software can be expanded using macros scripts and plug-ins (FIJI only). A quick demonstration of WEKA (https://www.cs.waikato.ac.nz/ml/weka/ ) a machine learning package embedded in FIJI.

Learning Objectives:

Upon completion of this workshop, attendees should be able to:

  1. Open DICOM images and check DICOM tags
  2. Apply filters (noise removal, enhance edges)
  3. Enhance Image contrast.
  4. Extract some imaging features & measurement.
  5. Apply artificial intelligence & machine learning algorithm for segmentation or
    predicting cancer.