Adaptive simulated annealing for CT image classification

This paper presents a pattern classification method that combines the classical Perceptron algorithm with simulated annealing. The approach is applied to the recognition of focal liver tumors presented in the DICOM format. On test sets of 100+100 examples (disjoint from the learning set) we obtain a correct classification of more than 98%.
This work was carried out as part of a collaboration with medical practitioners based at the Institute of Radiology, Humboldt University of Berlin. This paper builds upon work first presented at ESANN 2001.