Available online 29 April 2023, 107541

Author links open overlay panel, , , , , , , , ABSTRACTAIMS

Myocardial fibrosis (MF) is a common pathological process in a wide range of cardiovascular diseases. Its quantity has diagnostic and prognostic relevance. We aimed to assess if the complementary use of an automated artificial intelligence software might improve the precision of the pathologist´s quantification of MF on endomyocardial biopsies (EMB).

METHODS AND RESULTS

Intraoperative EMB samples from 30 patients with severe aortic stenosis submitted to surgical aortic valve replacement were analysed. Tissue sections were stained with Masson´s trichrome for collagen/fibrosis and whole slide images (WSI) from the experimental glass slides were obtained at a resolution of 0.5μm using a digital microscopic scanner. Three experienced pathologists made a first quantification of MF excluding the subendocardium. After two weeks, an algorithm for Masson´s trichrome brightfield WSI (at QuPath software) was applied and the automatic quantification was revealed to the pathologists, who were asked to reassess MF, blinded to their first evaluation. The impact of the automatic algorithm on the inter-observer agreement was evaluated using Bland-Altman type methodology.

Median values of MF on EMB were 8.33% [IQR 5.00-12.08%] and 13.60% [IQR 7.32-21.2%], respectively for the first pathologist´s and automatic algorithm quantification, being highly correlated (R2: 0.79; p<0.001). Inter-observer discordance was relevant, particularly for higher percentages of MF. The knowledge of the automatic quantification significantly improved the overall pathologist´s agreement, which became unaffected by the degree of MF severity.

CONCLUSIONS

The use of an automated artificial intelligence software for MF quantification on EMB samples improves the reproducibility of measurements by experienced pathologists. By improving the reliability of the quantification of myocardial tissue components, this adjunctive tool may facilitate the implementation of imaging-pathology correlation studies.

Section snippetsINTRODUCTION

Myocardial fibrosis (MF), meaning the excessive deposition of extracellular matrix (ECM) components in the myocardium, is a common pathological process in a wide spectrum of chronic heart diseases [1,2]. Being associated with the disruption of normal myocardial structure, it is behind the mechanistic base for adverse cardiac remodeling [3]. Indeed, it is a key contributor to heart failure and its progression, having recognized prognostic implications in both ischemic and non-ischemic cardiac

Study design

Present analysis was conducted on EMB samples from 30 patients undergoing elective surgical aortic valve replacement at our tertiary center between April 2019 and January 2022 because of isolated severe symptomatic AS, defined according to European guidelines on valvular heart disease [12]. This is part of a correlation research protocol involving both pre- and post-operative LV structural and functional assessment by multimodality imaging and myocardial histopathology study from EMB, specially

Population characteristics

This study was conducted on EMB samples from 30 patients undergoing elective surgical AVR, with a median age of 73 (68-77) years. Demographic and clinical characteristics of the population are presented in supplementary material – supplemental Table 1.

Myocardial fibrosis quantification by the Pathologists versus the Automatic Algoritm (QuPath)

MF was quantified in the tissue samples by the three experienced pathologists (supplemental Table 2). ‘Gold-standard’ MF value, defined as the average of the experienced pathologists’ initial quantification, had a median value of 8.33% [IQR

DISCUSSION

The main findings of our study were that: (1) the use of an automatic algorithm as an adjunctive tool for MF quantification at individual EMB samples improves the overall agreement between experienced pathologist´s quantifications, indistinctive of the MF severity; and (2) in the absence of such tool, the inter-observer variability is relevant and, importantly, the inter-observer discordance increases for higher percentages of fibrosis. Finally, to the best of our knowledge this is the first

LIMITATIONS

Contrary to some correlation studies we did not specifically assess the impact of the automatic algorithm quantification across distinct stains for collagen, such as Picrosirius red or Azan. This may be important when different institutional protocols are compared. Variations in staining intensity, which is related to specificity stain affinity for collagen fibers and further ECM components, staining hue, contrast, and tendency to fade over time may all impact the level of colour threshold

CONCLUSIONS

The use of an automatic digital pathology algorithm for the quantification of MF on EMB samples significantly improves the reproducibility of measurements by experienced pathologists. This was achieved through a machine learning imaging hierarchical protocol, developed on QuPath, an open-source software. As precision increases and automation is inferred, this quantification tool should be pursued for wide availability in the assessment of MF, a decisive biomarker on a huge range of

DISCLOSURES

None.

Uncited References

[15]

Declaration of Competing Interest

None to declare (on behalf of all authors)

SOURCES OF FUNDING

Nothing to declare.

ACKNOWLEDGEMENTS

The authors would like to thank to the technician Fernanda Silva, from Pathology Department, IPOFG, Lisboa, for microscopy slides digitization, and to Fundação Champalimaud, Lisboa, for the opportunity to use computer vision for the first applications of QuPath.

REFERENCES (21)

There are more references available in the full text version of this article.

View full text

© 2023 Elsevier Inc. All rights reserved.