Training of computational algorithms to predict NAFLD activity score and fibrosis stage from liver histopathology slides.

BACKGROUND: The incidence of non-alcoholic fatty liver disease (NAFLD) and its progressive form, non-alcoholic steatohepatitis (NASH), has been increasing for decades. Since the mainstay is lifestyle modification in this mainly asymptomatic condition, there is a need for accurate diagnostic methods. OBJECTIVES: This study proposes a method with a computer-aided diagnosis (CAD) system to predict NAFLD Activity score (NAS scores-steatosis, lobular inflammation, and ballooning) and fibrosis stage from histopathology slides. METHODS: A total of 87 pathology slides pairs (H&E and Trichrome-stained) were used for the study. Ground-truth NAS scores and fibrosis stages were previously identified by a pathologist. Each slide was split into 224 × 224 patches and fed into a feature extraction network to generate local features. These local features were processed and aggregated to obtain a global feature to predict the slide's scores. The effects of different training strategies, as well as training data with different staining and magnifications were explored. Four-fold cross validation was performed due to the small data size. Area Under the Receiver Operating Curve (AUROC) was utilized to evaluate the prediction performance of the machine-learning algorithm. RESULTS: Predictive accuracy for all subscores was high in comparison with pathologist assessment. There was no difference among the 3 magnifications (5x, 10x, 20x) for NAS-steatosis and fibrosis stage tasks. A larger magnification (20x) achieved better performance for NAS-lobular scores. Middle-level magnification was best for NAS-ballooning task. Trichrome slides are better for fibrosis stage prediction and NAS-ballooning score prediction task. NAS-steatosis prediction had the best performance (AUC 90.48%) in the model. A good performance was observed with fibrosis stage prediction (AUC 83.85%) as well as NAS-ballooning prediction (AUC 81.06%). CONCLUSIONS: These results were robust. The method proposed proved to be effective in predicting NAFLD Activity score and fibrosis stage from histopathology slides. The algorithms are an aid in having an accurate and systematic diagnosis in a condition that affects hundreds of millions of patients globally.

Effect of a Low-Rank Denoising Algorithm on Quantitative Magnetic Resonance Imaging-Based Measures of Liver Fat and Iron.

PURPOSE: This study aimed to assess the effect of a low-rank denoising algorithm on quantitative magnetic resonance imaging-based measures of liver fat and iron. MATERIALS AND METHODS: This was an institutional review board-approved, Health Insurance Portability and Accountability Act-compliant, retrospective analysis of 42 consecutive subjects who were imaged at 3T using a multiecho gradient echo sequence that was reconstructed using the multistep adaptive fitting algorithm to obtain quantitative proton density fat fraction (PDFF) and R2* maps (original maps). A patch-wise low-rank denoising algorithm was then applied, and PDFF and R2* maps were created (denoised maps). Three readers independently rated the PDFF maps in terms of vessel and liver edge sharpness and image noise using a 5-point scale. Two other readers independently measured mean and standard deviation of PDFF and R2* values for the original and denoised maps; values were compared using intraclass correlation coefficients (ICCs) and mean difference analyses. RESULTS: Qualitatively, the denoised maps were preferred by all 3 readers based on image noise (P < 0.001) and by 2 of 3 readers based on vessel edge sharpness (P < 0.001-0.99). No reader had a significant preference regarding liver edge sharpness (P = 0.16-0.48). Quantitatively, agreement was near perfect between the original and denoised maps for PDFF (ICC = 0.995) and R2* (ICC = 0.995) values. Mean quantitative values obtained from the original and denoised maps were similar for liver PDFF (7.6 ± 7.7% vs 7.7 ± 7.8%; P = 0.63) and R2* (52.9 ± 40.3s vs 52.8 ± 41.1 s, P = 0.74). CONCLUSIONS: Applying the low-rank denoising algorithm to liver fat and iron quantification reduces image noise in PDFF and R2* maps without adversely affecting mean quantitative values or subjective image quality.