Nuclear morphometry of soft tissue tumors of fibrous origin.

In this study we investigated the diagnostic significance of a set of different morphometric nuclear parameters in the differential diagnosis of soft tissue tumors. Nuclear area, the standard deviation of the nuclear area, the shape factor and other parameters such as Feret's Diameter and Martin's Radii were assessed using a computer assisted image analyzer system. A statistically significant difference (p < 0.01) between benign and malignant tumors and tumor-like lesions could be confirmed for the nuclear area and the standard deviation of the nuclear area, with the significance level being lower (p = 0.5) for the latter parameter. The shape factor also discriminated between the examined groups. Reclassification of the assessed histological diagnosis was performed by linear discriminant analysis using all possible combinations of the different nuclear parameters. This procedure disclosed an increasing rate of correctly reclassified cases with rising number of parameters applied. We conclude that the assessment of nuclear parameters may be helpful in the correct diagnosis and differential diagnosis of soft tissue tumors and tumor-like lesions of fibrous origin.