Sum variance--a most valuable parameter in computer-assisted image analysis of voided urines.

1996 
The chromatin pattern of retrospectively Feulgen–Schiff-stained exfoliative urothelial cells was determined by computer-assisted analysis for 12 detailed texture image parameters. These were compared with density parameters and geometric parameters for their ability to discriminate between benign and malignant samples. The discriminatory power of the mean variable of the texture parameter sum variance was found to equal that of densitometric parameters which reflect the nuclear DNA content of the cell. The results suggest that determination of the texture parameter sum variance is valuable in distinguishing nuclei of specimens negative from those positive for malignancy.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    3
    Citations
    NaN
    KQI
    []