LEARN algorithm: a novel option for predicting non-alcoholic steatohepatitis

Abstract

Background: There is an unmet need for accurate non-invasive methods to diagnose non-alcoholic steatohepatitis (NASH). Since impedance-based measurements of body composition are simple, repeatable and have a strong association with non-alcoholic fatty liver disease (NAFLD) severity, we aimed to develop a novel and fully automatic machine learning algorithm, consisting of a deep neural network based on impedance-based measurements of body composition to identify NASH [the bioeLectrical impEdance Analysis foR Nash (LEARN) algorithm].

Methods: A total of 1,259 consecutive subjects with suspected NAFLD were screened from six medical centers across China, of which 766 patients with biopsy-proven NAFLD were included in final analysis. These patients were randomly subdivided into the training and validation groups, in a ratio of 4:1. The LEARN algorithm was developed in the training group to identify NASH, and subsequently, tested in the validation group.

Results: The LEARN algorithm utilizing impedance-based measurements of body composition along with age, sex, pre-existing hypertension and diabetes, was able to predict the likelihood of having NASH. This algorithm showed good discriminatory ability for identifying NASH in both the training and validation groups [area under the receiver operating characteristics (AUROC): 0.81, 95% CI: 0.77-0.84 and AUROC: 0.80, 95% CI: 0.73-0.87, respectively]. This algorithm also performed better than serum cytokeratin-18 neoepitope M30 (CK-18 M30) level or other non-invasive NASH scores (including HAIR, ION, NICE) for identifying NASH (P value <0.001). Additionally, the LEARN algorithm performed well in identifying NASH in different patient subgroups, as well as in subjects with partial missing body composition data.

Conclusions: The LEARN algorithm, utilizing simple easily obtained measures, provides a fully automated, simple, non-invasive method for identifying NASH.

Keywords: Non-alcoholic fatty liver disease (NAFLD); bioeLectrical impEdance Analysis foR Nash (LEARN) algorithm; body composition; non-alcoholic steatohepatitis (NASH).

Figure 1

Flowchart of the study. NAFLD, non-alcoholic fatty liver disease.

Figure 2

Flowchart of the deep neural network algorithm for prediction of NASH (namely the LEARN algorithm). Input layer: input the normalized data into this layer consisting of four modules. First, a FC layer, to synthesize the features extracted from the previous section. Second, a BN layer, to simplify the calculation and make the data retain its original expression ability as far as possible after the normalization processing. Third, Tanh function, a nonlinear function to help machines learn complex mappings. Last, the Dropout layer, reducing overfitting, extract a matrix containing 1,536 features. Hidden layer: the matrix containing 1,536 features is input into this layer, and the data needs to be looped four times through the residual module. Output layer: the output layer is consisted of a FC layer and a Softmax function. Through the Softmax function, we can map the output values to the interval (0, 1) for the final classification. FC, full connected; BN, batch normalization; NASH, non-alcoholic steatohepatitis; LEARN, bioeLectrical impEdance Analysis foR Nash.

Figure 3

Diagnostic performance of LEARN algorithm and sensitivity, specificity of the dual cut-off values in the training and validation groups. (A,B) Training group. (C,D) Validation group. AUC, area under the curve; LEARN algorithm: deep neural network model for identifying non-alcoholic steatohepatitis; LEARN, bioeLectrical impEdance Analysis foR Nash.

Figure 4

Boxplot of the LEARN algorithm versus histopathological severity of NAFLD in the training group: (A) steatosis grade, (B) ballooning grade, (C) lobular inflammation grade, and (D) presence of definite NASH. NASH, non-alcoholic steatohepatitis; LEARN algorithm: deep neural network model for identifying non-alcoholic steatohepatitis; LEARN, bioeLectrical impEdance Analysis foR Nash; NAFLD, non-alcoholic fatty liver disease.