NUCLEAR TEXTURE IN POORLY DIFFERENTIATED SMALL ROUND-CELL TUMORS - IMAGE-ANALYSIS STUDY OF FINE-NEEDLE ASPIRATION MATERIAL

Gradient analysis and pattern spectr um decomposition based on mathema tical morphology concepts were used to explore nuclear texture pattern s in a pool of 108 cells obtained by fine needle aspiration of five un differentiated small round cell tumors of childhood, including one cas e each of Wilms' tumor, neuroblastoma, lymphoblastic lymphoma, Ewing's sarcoma and rhabdomyosarcoma. The aim of the study was to determine t he presumptive value of nuclear pattern to correctly allocate each iso lated cell to each of the five patients. The cells were examples of fi ve histogenetically different tumors, all undifferentiated and with a close microscopic resemblance to one another. High gradient structures (heterochromatin-euchromatin and nuclear membrane edges) were estimat ed by a difference-of-boxes filter, and pattern spectrum decomposition was obtained by successive openings and closings performed on the inp ut gray tone image. One important feature of these procedures was that no prior selection by thresholding of the structures to be studied wa s required, thus obviating subjective bins. Percentages of correctly a llocated cells by canonical analysis ranged from 70.0% (rhabdomyosarco ma) to 92.9% (Ewing's sarcoma). Although the five cases could be disti nguished using seven texture variables, this does not imply generaliza tion of the results for the differential diagnosis of these tumors. No netheless, the possibility that undifferentiated small round cells pre sent distinctive nuclear patterns when studied by sensitive image anal ysis techniques is suggested by our results.