Related Experiment Video
Updated: Aug 5, 2026
11:00
Visualization, Quantification, and Mapping of Immune Cell Populations in the Tumor Microenvironment
Published on: March 25, 2020
A study of hepatocellular carcinoma using morphometric and densitometric image analysis
B S Erler1, H M Truong, S S Kim
1Department of Pathology and Laboratory Medicine, Cedars-Sinai Medical Center, Los Angeles, California 90048.
American Journal of Clinical Pathology
|August 1, 1993
Summary
Image analysis accurately distinguishes benign from malignant hepatocytes, aiding hepatocellular carcinoma diagnosis. This method uses nuclear features to classify cells, improving diagnostic accuracy in challenging specimens.
Area of Science:
- Hepatobiliary pathology
- Computational pathology
- Medical imaging analysis
Background:
- Hepatocellular carcinoma (HCC) diagnosis is challenging in cytologic and biopsy samples.
- Distinguishing benign from malignant hepatocytes requires careful morphologic assessment.
- Existing diagnostic methods may have limitations in accuracy and efficiency.
Purpose of the Study:
- To evaluate the utility of image analysis in differentiating benign and malignant hepatocytes.
- To identify nuclear morphometric and densitometric parameters effective for HCC classification.
- To assess the accuracy of image analysis-based classification models.
Main Methods:
- Comparison of 42 malignant and 26 benign hepatocyte groups from HCC patient biopsies.
- Nuclear measurements using a microcomputer-based image analysis system and flexible imaging software.
- Evaluation of 22 nuclear morphometric and densitometric parameters.
- Classification using optimized linear discriminant functions.
Main Results:
- The nuclear major axis was the best single discriminator.
- High positive predictive values (PV+) and negative predictive values (PV-) were achieved: 95.0% PV+ and 85.7% PV- for the major axis.
- Combined morphometric/densitometric parameters yielded 95.5% PV+ and 100% PV-.
- Multivariate analysis demonstrated accurate classification capabilities.
Conclusions:
- Image analysis of nuclear features provides accurate classification of benign and malignant hepatocytes.
- This computational pathology approach can enhance diagnostic accuracy for hepatocellular carcinoma.
- Image analysis offers a valuable tool for improving the diagnosis of challenging liver lesions.
