What is it about?
LFT data comprises 6 - 8 separate markers, with new analyses suggesting that results are sometimes redundant and difficult to interpret. As an example of addressing redundacy, the combination of single decision trees and support vector machines predicted serum GGT with ALP and ALT.
Featured Image
Why is it important?
Pattern recognition via machine learning can predict other LFT results, including with reference to other blood markers; e.g. cholesterol and uric acid for GGT and alcohol abuse.
Perspectives
Time to reconsider standard reference intervals for pathology data, and look to data patterns to discern disease processes, and so on ...
Associate Professor Brett A Lidbury
Australian National University
Read the Original
This page is a summary of: Assessment of machine-learning techniques on large pathology data sets to address assay redundancy in routine liver function test profiles, Diagnosis, January 2015, De Gruyter,
DOI: 10.1515/dx-2014-0063.
You can read the full text:
Contributors
The following have contributed to this page
