TEXTURAL ANALYSIS OF LYMPHOID-CELLS IN SEROUS EFFUSIONS - A MATHEMATICAL MORPHOLOGIC APPROACH

Nuclear texture of reactive and well-differentiated neoplastic lymphoc ytes from serous effusions was studied by an approach based on princip les of mathematical morphology. Density features were obtained before and after gray level nuclear image transformation by morphologic closi ng (dilation followed by erosion) and application of a top-hat functio n, which detects light or dense spots with a determined width and cont rast. Five cases of benign reactive lymphocytic serous effusions and 1 1 cases of effusions in well-differentiated lymphocytic lymphomas were analyzed retrospectively. Each lymphoid cell was characterized by 24 densitometric features. Rank-order transformation was used for linear discriminant analysis given nonnormal distributions of variables. In a first model formed by pooling all cells, discriminant function distin guished between 110 reactive and 216 malignant lymphocytes in the lear ning set and between III reactive and 226 malignant lymphocytes in the test set with better than 81% accuracy in both. In a second model, co rrect classification of cases as reactive or malignant was achieved in 515 reactive and 11/11 malignant lymphoid effusions. The results indi cate that mathematical morphologic transformations of the gray level i mage may be an effective adjunct to other textural descriptors of cell ular atypia, especially in the differential diagnosis of lymphoid sero us effusions.