CHROMATIN TEXTURE SIGNATURES IN NUCLEI FROM PROSTATE LESIONS

OBJECTIVE: To characterize nuclei from prostatic lesions in a highly s pecific manner by developing a nuclear chromatin texture signature and to characterize lesions by means of their composition of nuclei with diverse degrees of deviation from normal. STUDY DESIGN: High-resolutio n digitized imagery of nuclei from normal prostates, from prostatic ne oplastic lesions of low and high grade and from histologically normal appearing regions of prostates with low and high grade prostatic intra epithelial neoplasia (PIN) lesions were recorded. A set of 65 features descriptive of the spatial and statistical distribution of nuclear ch romatin was computed for each nucleus. These features were arranged an d processed to form a distinctive signature. A distance metric from '' normal'' was defined and computed for each nucleus. RESULTS: Profiles of feature values can, after suitable scaling, be presented as distinc tive feature value signatures. For many practical applications, profil es based on It standardized distance from normal nuclei may be more us eful. Such profiles allow the derivation of a progression curve, showi ng increasing distances for diagnostic groups with increasing lesion p rogression tip to high grade PIN lesions. Within each diagnostic group different cases show distinctive distributions of nuclei with differi ng degrees of deviation from normal, allowing the derivation of a lesi on signature. CONCLUSION: Nuclear chromatin texture signatures may be of value for the characterization of both nuclei and lesions. They are based on a more comprehensive use of information offered by the nucle ar chromatin pattern than that included in classification methods. Whi le these signatures offer a more specific characterization of a clinic al sample, they also are subject to move variability within a diagnost ic category. This may not be due to randomness but may reflect some ac tual differences between lesions.