J. Malone, K. Law, S. Prabhu, and P. Goddard (UK)
textural analysis, machine learning, com puted tomography.
There has been much work recently in the classification of interstitial lung disease from CT scans using texture anal ysis. The process generally involves one or more radiolo gists labelling regions of the lung parenchyma as represen tative of a particular condition. The character of these re gions falls broadly into two groups: those which the radiol ogists present and, it is supposed, all others would agree are highly representative of a particular pattern; and the regions which, although deemed to belong to the class in question, are not necessarily perfect examples and need not possess an homogeneous texture throughout. In short, the data can be clean or contain noise. There are circumstances in which information from only one of these categories may be avail able and it may be necessary to train a machine learning al gorithm and classify regions which belong to the other type or are of unknown origin. Here we evaluate the decrease in accuracy associated with such incomplete information us ing a number of common classifiers.
Important Links:
Go Back
