Quantification of myocardial fibrosis by digital image analysis and interactive stereology

Abstract

Background: Cardiac fibrosis disrupts the normal myocardial structure and has a direct impact on heart function and survival. Despite already available digital methods, the pathologist's visual score is still widely considered as ground truth and used as a primary method in histomorphometric evaluations. The aim of this study was to compare the accuracy of digital image analysis tools and the pathologist's visual scoring for evaluating fibrosis in human myocardial biopsies, based on reference data obtained by point counting performed on the same images.

Methods: Endomyocardial biopsy material from 38 patients diagnosed with inflammatory dilated cardiomyopathy was used. The extent of total cardiac fibrosis was assessed by image analysis on Masson's trichrome-stained tissue specimens using automated Colocalization and Genie software, by Stereology grid count and manually by Pathologist's visual score.

Results: A total of 116 slides were analyzed. The mean results obtained by the Colocalization software (13.72 ± 12.24%) were closest to the reference value of stereology (RVS), while the Genie software and Pathologist score gave a slight underestimation. RVS values correlated strongly with values obtained using the Colocalization and Genie (r>0.9, p<0.001) software as well as the pathologist visual score. Differences in fibrosis quantification by Colocalization and RVS were statistically insignificant. However, significant bias was found in the results obtained by using Genie versus RVS and pathologist score versus RVS with mean difference values of: -1.61% and 2.24%. Bland-Altman plots showed a bidirectional bias dependent on the magnitude of the measurement: Colocalization software overestimated the area fraction of fibrosis in the lower end, and underestimated in the higher end of the RVS values. Meanwhile, Genie software as well as the pathologist score showed more uniform results throughout the values, with a slight underestimation in the mid-range for both.

Conclusion: Both applied digital image analysis methods revealed almost perfect correlation with the criterion standard obtained by stereology grid count and, in terms of accuracy, outperformed the pathologist's visual score. Genie algorithm proved to be the method of choice with the only drawback of a slight underestimation bias, which is considered acceptable for both clinical and research evaluations.

Virtual slides: The virtual slide(s) for this article can be found here: http://www.diagnosticpathology.diagnomx.eu/vs/9857909611227193.

Figure 1

Fibrosis mark-up on digitized slide: (A) Masson trichrome original staining, (B) Colocalization algorithm, (C) Genie algorithm.

Figure 2

Training and using of the genie: (A) Defining and marking the classes of interest,(B) Testing and using the new specific classifier.

Figure 3

Fibrosis mark-up on digitized slide using stereology protocol: (A) ImageScope V11 view incorporating grid (sampling step 200 pixels and size of the pattern 20 pixels), (B, C) structures of interest (glass, fibrosis, myocardium, other) manually highlighted by observer.

Figure 4

Single linear regression models with reference values: (A) colocalization and stereology; (B) genie and stereology. Linear regression line is presented within 95% confidence interval.

Figure 5

Single linear regression models with reference values: (A) Pathologist mean score and stereology; (B) pathologist score at week 0 and week 2; (C) colocalization and genie. Linear regression line is presented within 95% confidence interval.

Figure 6

Bland-Altman plots and histograms of the method score differences: (A) colocalization and stereology; (B) genie and stereology; (C) pathologist mean score and stereology. Horizontal line represents mean difference within limits of agreement, which are defined as the mean difference ± 2 standard deviations.