Related Experiment Videos
Markov optical texture parameters as prognostic indicators in ovarian carcinoma
J. P. Geisler1, H.E. Geisler, G.A. Miller
1Division of Gynecologic Oncology, Department of Obstetrics and Gynecology; Division of Oncology Research, Department of Medicine, St. Vincent Hospitals and Health Services and Department of Pathology, Laboratory for Diagnostic and Analytical Cytometry, Indianapolis, Indiana.
Summary
Markov nuclear texture analysis, using computer-aided image analysis, identified sum entropy as an independent prognostic indicator for survival in epithelial ovarian cancer patients. This method, alongside FIGO stage and cytoreduction, aids in predicting patient outcomes.
Area of Science:
- Oncology
- Medical Imaging
- Biostatistics
Background:
- Texture analysis quantifies surface heterogeneity in medical images.
- Computer-aided image analysis enables objective texture measurement.
- Prognostic indicators are crucial for managing epithelial ovarian carcinoma.
Purpose of the Study:
- To prospectively evaluate Markov nuclear texture features as prognostic indicators of survival in epithelial ovarian cancer.
- To determine the independent predictive value of texture features compared to clinical factors.
Main Methods:
- Prospective study of 99 epithelial ovarian carcinoma patients treated with initial surgery.
- Quantification of 20 Markov nuclear texture features using image analysis.
- Multivariate analysis correlating texture features, FIGO stage, cytoreduction, and survival.
Main Results:
- Five optical texture features correlated significantly with survival.
- Sum entropy, a Markov texture feature, was an independent predictor of survival (P = 0.035).
- FIGO stage (P = 0.0031) and level of cytoreduction (P < 0.0001) were also independent predictors.
Conclusions:
- Optical texture analysis using image analysis is feasible for quantifying nuclear features.
- The Markov texture feature, sum entropy, is an independent prognostic indicator for survival in epithelial ovarian cancer.
- FIGO stage and optimal cytoreduction remain critical independent prognostic factors.