Automated Machine Learning in the Sonographic Diagnosis of Non-alcoholic Fatty Liver Disease

doi:10.37015/AUDT.2020.200008

Abstract

Abstract:

Objective: This study evaluated the performance of automated machine-learning to diagnose non-alcoholic fatty liver disease (NAFLD) by ultrasound and compared these findings to radiologist performance.
Methods: 96 patients with histologic (33) or proton density fat fraction MRI (63) diagnosis of NAFLD and 100 patients without evidence of NAFLD were retrospectively identified. The “Fatty Liver” label included 96 patients with 405 images and the “Not Fatty Liver” label included 100 patients with 500 images. These 905 images made up a “Comprehensive Image” group. A “Radiology Selected Image” group was then created by selecting only images considered diagnostic by a blinded radiologist, resulting in 649 images. Cloud AutoML Visionbeta (Google LLC, Mountain View, CA) was used for machine learning. The models were evaluated against three blinded radiologists.
Results: The “Comprehensive Image” group model demonstrated a sensitivity of 88.6% (73.3-96.8%) and a specificity of 95.3% (84.2-99.4%). Radiologist performance on this image group included a sensitivity of 81.0% (74.3-87.6%) and specificity of 86.0% (72.6-99.5%). The model’s overall accuracy was 92.3% (84.0-97.1%), compared with mean individual performance (83.8%, 78.4-89.1%). The “Radiology Selected Image” group model demonstrated a sensitivity of 88.6% (73.3 - 96.8%) and specificity of 87.9% (71.8-96.6%). Mean radiologist sensitivity was 92.4% (86.9-97.9%) and specificity was 91.9% (83.4-100%). The model’s overall accuracy was 88.2% (78.1-94.8%) which was comparable to the individual radiologist performance (92.2%, 90.1-94.2%) and consensus performance (95.6%, 87.6-99.1%).
Conclusions: An automated machine-learning algorithm may accurately detect NAFLD on ultrasound.

Key words: Non-alcoholic fatty liver disease; Machine learning; Ultrasound

Sriharsha Gummadi, MD, Nirmal Patel, Haresh Naringrekar, MD, Laurence Needleman, MD, Andrej Lyshchik, MD PhD, Patrick O’Kane, MD, Jesse Civan, MD, John R Eisenbrey, PhD. Automated Machine Learning in the Sonographic Diagnosis of Non-alcoholic Fatty Liver Disease. Advanced Ultrasound in Diagnosis and Therapy, 2020, 4(3): 176-182.

References

[1]	Younossi ZM, Koenig AB, Abdelatif D, Fazel Y, Henry L, Wymer M. Global epidemiology of nonalcoholic fatty liver disease-meta-analytic assessment of prevalence, incidence, and outcomes. Hepatology 2016; 64:73-84.
[2]	Younossi ZM, Stepanova M, Afendy M, Fang Y, Younossi Y, Mir H, et al. Changes in the prevalence of the most common causes of chronic liver diseases in the united states from 1988 to 2008. Clin Gastroenterol Hepatol 2011; 9:524-30.
[3]	Sheth SG, Chopra S. Epidemiology, clinical features, and diagnosis of nonalcoholic fatty liver disease in adults. Waltham, MA: UpToDate Inc; 2018.
[4]	Cho CS, Curran S, Schwartz LH, Kooby DA, Klimstra DS, Shia J, et al. Preoperative radiographic assessment of hepatic steatosis with histologic correlation. J Am Coll Surg 2008; 206:480-8.
[5]	Springer F, Machann J, Claussen CD, Schick F, Schwenzer NF. Liver fat content determined by magnetic resonance imaging and spectroscopy. World J Gastroenterol 2010; 16:1560-6.
[6]	Borra RJ, Salo S, Dean K, Lautamaki R, Nuutila P, Komu M, et al. Nonalcoholic fatty liver disease: Rapid evaluation of liver fat content with in-phase and out-of-phase MR imaging. Radiology 2009; 250:130-6.
[7]	Hernaez R, Lazo M, Bonekamp S, Kamel I, Brancati FL, Guallar E, et al. Diagnostic accuracy and reliability of ultrasonography for the detection of fatty liver: A meta-analysis. Hepatology 2011; 54:1082-90.
[8]	Mottin CC, Moretto M, Padoin AV, Swarowsky AM, Toneto MG, Glock L, et al. The role of ultrasound in the diagnosis of hepatic steatosis in morbidly obese patients. Obes Surg 2004; 14:635-7.
[9]	Tang A, Tan J, Sun M, Hamilton G, Bydder M, Wolfson T, et al. Nonalcoholic fatty liver disease: MR imaging of liver proton density fat fraction to assess hepatic steatosis. Radiology 2013; 267:422-31.
[10]	Janssen I, Katzmarzyk PT, Ross R. Body mass index, waist circumference, and health risk: Evidence in support of current national institutes of health guidelines. Arch Intern Med 2002; 162:2074-9.
[11]	Chalasani N, Younossi Z, Lavine JE, Charlton M, Cusi K, Rinella M, et al. The diagnosis and management of nonalcoholic fatty liver disease: Practice guidance from the american association for the study of liver diseases. Hepatology 2018; 67:328-57.
[12]	Acharya UR, Fujita H, Bhat S, Raghavendra U, Gudigar A, Molinari F, et al. Decision support system for fatty liver disease using gist descriptors extracted from ultrasound images. Information Fusion 2016; 29:32-39.
[13]	Byra M, Styczynski G, Szmigielski C, Kalinowski P, Michalowski L, Paluszkiewicz R, et al. Transfer learning with deep convolutional neural network for liver steatosis assessment in ultrasound images. Int J Comput Assist Radiol Surg 2018; 13:1895-903.
[14]	Chartrand G, Cheng PM, Vorontsov E, Drozdzal M, Turcotte S, Pal CJ, et al. Deep learning: A primer for radiologists. Radiographics 2017; 37:2113-31.
[15]	Daniels K, Gummadi S, Zhu Z, Wang S, Patel J, Swendseid B, et al. Machine learning by ultrasonography for genetic risk stratification of thyroid nodules. JAMA Otolaryngol Head Neck Surg 2019, 10.1001/jamaoto.2019.30731-6.
[16]	S Wang, JB Liu, ZY Zhu, Eisenbrey J. Artificial intelligence in ultrasound imaging: Current research and applications.Advanced Ultrasound in Diagnosis and Therapy 2019; 3:53-61.