pubmed.ncbi.nlm.nih.gov

Transfer learning with deep convolutional neural network for liver steatosis assessment in ultrasound images - PubMed

Transfer learning with deep convolutional neural network for liver steatosis assessment in ultrasound images

Michał Byra et al. Int J Comput Assist Radiol Surg. 2018 Dec.

Abstract

Purpose: The nonalcoholic fatty liver disease is the most common liver abnormality. Up to date, liver biopsy is the reference standard for direct liver steatosis quantification in hepatic tissue samples. In this paper we propose a neural network-based approach for nonalcoholic fatty liver disease assessment in ultrasound.

Methods: We used the Inception-ResNet-v2 deep convolutional neural network pre-trained on the ImageNet dataset to extract high-level features in liver B-mode ultrasound image sequences. The steatosis level of each liver was graded by wedge biopsy. The proposed approach was compared with the hepatorenal index technique and the gray-level co-occurrence matrix algorithm. After the feature extraction, we applied the support vector machine algorithm to classify images containing fatty liver. Based on liver biopsy, the fatty liver was defined to have more than 5% of hepatocytes with steatosis. Next, we used the features and the Lasso regression method to assess the steatosis level.

Results: The area under the receiver operating characteristics curve obtained using the proposed approach was equal to 0.977, being higher than the one obtained with the hepatorenal index method, 0.959, and much higher than in the case of the gray-level co-occurrence matrix algorithm, 0.893. For regression the Spearman correlation coefficients between the steatosis level and the proposed approach, the hepatorenal index and the gray-level co-occurrence matrix algorithm were equal to 0.78, 0.80 and 0.39, respectively.

Conclusions: The proposed approach may help the sonographers automatically diagnose the amount of fat in the liver. The presented approach is efficient and in comparison with other methods does not require the sonographers to select the region of interest.

Keywords: Convolutional neural networks; Deep learning; Hepatorenal index; Nonalcoholic fatty liver disease; Transfer learning; Ultrasound imaging.

PubMed Disclaimer

Conflict of interest statement

Conflict of interest

The authors do not have any conflict of interest.

Ethical statements

All procedures performed in studies involving human participants were in accordance with the Ethical Standards of the Medical University of Warsaw and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Figures

Fig. 1
Fig. 1

Histogram of steatosis level across the population of patients in the study group

Fig. 2
Fig. 2

Liver B-mode images and the ROIs selected for HI calculation, a steatosis level of 3% and b 25%, respectively

Fig. 3
Fig. 3

Illustration of feature extraction using the Inception-ResNet-v2 model [28]

Fig. 4
Fig. 4

The validation pipeline

Fig. 5
Fig. 5

The ROC curves for the HI method (AUC = 0.959), the GLCM algorithm (AUC = 0.893) and the classifier developed using CNN features (AUC = 0.977)

Fig. 6
Fig. 6

The usefulness of a the HI parameter (SCC = 0.80), b GLCM-based features (SCC = 0.39) and c the CNN-based features (SCC = 0.78) in steatosis level assessment

Fig. 7
Fig. 7

a The relation between the HI parameter and the Lasso regression (SCC = 0.78) and b the corresponding Bland–Altman plot

Similar articles

Cited by

References

    1. Beeman SC, Garbow JR. Imaging and metabolism. New York: Springer; 2018.
    1. Adams LA, Scott Harmsen JL, Charatcharoenwitthaya P, Enders FB, Therneau T, Angulo P. Nonalcoholic fatty liver disease increases risk of death among patients with diabetes: a community-based cohort study. Am J Gastroenterol. 2010;105:1567–1573. doi: 10.1038/ajg.2010.18. - DOI - PMC - PubMed
    1. Adams LA, Anstee QM, Tilg H, Targher G. Non-alcoholic fatty liver disease and its relationship with cardiovascular disease and other extrahepatic diseases. Gut. 2017;66:1138–1153. doi: 10.1136/gutjnl-2017-313884. - DOI - PubMed
    1. Tapper EB, Lok ASF. Use of liver imaging and biopsy in clinical practice. N Engl J Med. 2017;377:756–768. doi: 10.1056/NEJMra1610570. - DOI - PubMed
    1. Luapuadat AM, Jianu IR, Ungureanu BS, Florescu LM, Gheonea DI, Sovaila S, Gheonea IA. Non-invasive imaging techniques in assessing non-alcoholic fatty liver disease: a current status of available methods. J Med Life. 2017;10:19–26. - PMC - PubMed

MeSH terms