Transfer learning with pre-trained deep convolutional neural networks for the automatic assessment of liver steatosis in ultrasound images.

AIM: In this paper we proposed different architectures of convolutional neural network (CNN) to classify fatty liver disease in images using only pixels and diagnosis labels as input. We trained and validated our models using a dataset of 629 images consisting of 2 types of liver images, normal and liver steatosis. MATERIAL AND METHODS: We assessed two pre-trained models of convolutional neural networks, Inception-v3 and VGG-16 using fine-tuning. Both models were pre-trained on ImageNet dataset to extract features from B-mode ultrasound liver images. The results obtained through these methods were compared for selecting the predictive model with the best performance metrics. We trained the two models using a dataset of 262 images of liver steatosis and 234 images of normal liver. We assessed the models using a dataset of 70 liver steatosis im-ages and 63 normal liver images. RESULTS: The proposed model that used Inception v3 obtained a 93.23% test accuracy with a sensitivity of 89.9%% and a precision of 96.6%, and areas under each receiver operating characteristic curves (ROC AUC) of 0.93. The other proposed model that used VGG-16, obtained a 90.77% test accuracy with a sensitivity of 88.9% and a precision of 92.85%, and areas under each receiver operating characteristic curves (ROC AUC) of 0.91. CONCLUSION: The deep learning algorithms that we proposed to detect steatosis and classify the images in normal and fatty liver images, yields an excellent test performance of over 90%. However, future larger studies are required in order to establish how these algorithms can be implemented in a clinical setting.

Correlation of Angiogenesis and Inflammation with Post-Operative Complications in Patients with Fatty Liver Disease Undergoing Liver Resection.

BACKGROUND: Hepatic steatosis has been identified as an independent risk factor for post-operative complications. The aim of our research was to assess how inflammation and neoangiogenesis associated with different stages of hepatic steatosis are related to post-operative complications in patients who undergo hepatic resection. METHODS: Our study included 19 patients with hepatic steatosis undergoing liver resection for primary or secondary tumors. For every patient we performed immunostaining using a panel of 5 primary antibodies (CD3, CD20, CD68, CD31, CD34) to highlight inflammation and neoangiogensis in the non-tumoral hepatic parenchyma. RESULTS: Taking into consideration the number of vessels as well as the signal area and integrated optical density (IOD) forCD3, CD20, CD68, and also the degree of steatosis, the univariate analysis with a log-rank (Mantel-Cox) test revealed that patients with higher values of CD31 and CD34 had a higher rate of post-operative complications on a 30-day follow-up period. Also, we used a Mann-Whitney U and Kruskal-Wallis H tests for group distributions. We noticed thatCD34 was significantly increased in patients diagnosed with steatosis compared to the control group and there was a statistically significant difference between CD31 median values of S0 (27.6) and S1 (55.8) grades. CONCLUSION: Patients with steatosis that presented higher values of CD31 and CD34 had a higher rate of post-operative complications. Further studies should assess the value of pre-operative evaluation of angiogenesis in patients with liver steatosis submitted to liver surgery.