Project Page

Back to search results

ANDI-II-A Taking research tools into clinical Practice: Model-based image "type" detection

Dr Andrew Janke, Dr Nicolas Cherbuin, Dr Moyra Mortby
Research Centre DCRC Early Diagnosis and Prevention
Partner Institution Australian National University
Project Description

As part of any processing one of the fundamental steps is to first ensure that the data you are receiving and performing analysis is indeed of the correct anatomical area, that cropping is correct and that for example, the correct sequences have been used for an MRI image.

For smaller projects this process can be checked manually after extraction of the data from the DICOM headers. In larger projects such as ANDI, this checking time becomes prohibitively expensive.

Without correct identification and QC (Quality Control) of data the resulting errors can have small but often undetectable effects to the eye in the results. The results of course are compromised generally via an increase in variance.

Project Portfolio

Back to search results