Automated quantification of steatosis: agreement with stereological point counting

Abstract

BackgroundSteatosis is routinely assessed histologically in clinical practice and research. Automated image analysis can reduce the effort of quantifying steatosis. Since reproducibility is essential for practical use, we have evaluated different analysis methods in terms of their agreement with stereological point counting (SPC) performed by a hepatologist.MethodsThe evaluation was based on a large and representative data set of 970 histological images from human patients with different liver diseases. Three of the evaluated methods were built on previously published approaches. One method incorporated a new approach to improve the robustness to image variability.ResultsThe new method showed the strongest agreement with the expert. At 20× resolution, it reproduced steatosis area fractions with a mean absolute error of 0.011 for absent or mild steatosis and 0.036 for moderate or severe steatosis. At 10× resolution, it was more accurate than and twice as fast as all other methods at 20× resolution. When compared with SPC performed by two additional human observers, its error was substantially lower than one and only slightly above the other observer.ConclusionsThe results suggest that the new method can be a suitable automated replacement for SPC. Before further improvements can be verified, it is necessary to thoroughly assess the variability of SPC between human observers.

Systematic Evaluation of the Safety Threshold for Allograft Macrovesicular Steatosis in Cadaveric Liver Transplantation

“…The H&E stain has lower accuracy in terms of prediction of steatotic severity for high inter-observer variation (Fiorini et al, 2004). Computerized analysis of Oil Red O (ORO)-stained samples were recommended to optimize the classification and minimize these random errors (Fiorini et al, 2004; Homeyer et al, 2017). Thirdly, potential bias might be caused by unavailable information on allograft biopsies in a portion of the donors.…”

Focused scores enable reliable discrimination of small differences in steatosis

“…Also, it requires no manual interaction and virtually no additional computational costs over the standard score. The focused scores approach is complementary to many previously published methods for computing steatosis area fractions [ 9 – 14 ]. Deriving focused scores from these methods can be a simple way to improve their reliability.…”

“…This process is independent of the particular method for computing steatosis area fractions and arbitrary methods can be used. In [ 9 ], different methods for computing steatosis area fractions were compared in terms of their agreement with human observers. One algorithm was significantly superior to the others and suggested as a suitable automated replacement for manual analysis.…”

“…Visual estimation by a hepatopathologist is the traditional method of assessing steatosis in histological slides [ 4 ]. However, in recent years, various automated analysis methods were developed to make the measurement of steatosis in histological images more efficient and reproducible [ 9 – 14 ]. These methods quantify steatosis in terms of the steatosis area fraction, that is, the area fraction of macrovesicular fat droplets with respect to the total tissue area.…”

Liver Steatosis Segmentation With Deep Learning Methods

“…Hailed as a new alternative solution, digital pathology is an emerging field that uses digital high-resolution images of tissue sections for machine-based image processing. Although multiple automated methods for liver steatosis measurement have demonstrated computational advantages over human reviewing process [4,7,8,9,10,11,12], they are not sufficiently accurate to support precise steatosis quantification, especially when overlapped steatosis regions with weak separating borders are in presence. As a result, it still remains challenging to develop a robust image analysis program to support precise liver steatosis analysis.…”