Computer recognition of regional lung disease patterns

We have developed an objective, reproducible, and automated means for the r egional evaluation of the pulmonary parenchyma from computed tomography (CT ) scans. This method, known as the Adaptive Multiple Feature Method (AMFM) assesses as many as 22 independent texture features in order to classify a tissue pattern. In this study, the six tissue patterns characterized were: honeycombing, ground glass, bronchovascular, nodular, emphysemalike, and no rmal. The lung slices were evaluated regionally using 31 x 31 pixel regions of interest. In each region of interest, an optimal subset of texture feat ures was evaluated to determine which of the six patterns the region could be characterized as. The computer output was validated against experienced observers in three settings. In the first two readings, when the observers were blinded to the primary diagnosis of the subject, the average computer versus observer agreement was 44.4 +/- 8.7% and 47.3 +/- 9.0%, respectively . The average interobserver agreement for the same two readings were 48.8 /- 9.1% and 52.2 +/- 10.0%, respectively. In the third reading, when the ob servers were provided the primary diagnosis, the average computer versus ob server agreement was 51.7 +/- 2.9% where as the average interobserver agree ment was 53.9 +/- 6.2%. The kappa statistic of agreement between the region s, for which the majority of the observers agreed on a pattern type, versus the computer was found to be 0.62. For regional tissue characterization, t he AMFM is 100% reproducible and performs as well as experienced human obse rvers who have been told the patient diagnosis.