Application of a fuzzy pattern classifier to decision making in portal verification of radiotherapy

With the large volume of electronic portal images acquired and stringent ti me constraints, it is no longer feasible to follow the convention whereby t he radiation oncologist reviews and approves or rejects all portals. For th at purpose we have developed a portal image classifier based on the fuzzy k -nearest neighbour (L-NN) algorithm Each portal image is represented by a f eature vector that consists of translational and rotational errors in the p lacement of radiation held borders that were measured in the portal image. Memberships in the acceptable portal class for the reference portal images within a training dataset were defined by a radiation oncologist expert. Th e fuzzy k-NN portal image classifier was trained and tested on a dataset of 328 portal images acquired during tangential irradiations of the breast. T he memberships in the acceptable portal class produced by the fuzzy k-NN al gorithm agreed very well with those defined by the expert. The linear corre lation coefficient was equal to 0.89. Performance of the fuzzy k-NN classif ier was also evaluated from the portal decision-making point of view using the measures of accuracy, sensitivity and specificity. The fuzzy L-NN porta l classifier was capable of identifying almost all the truly unacceptable p ortals with an acceptably low false alarm rate.