Development and validation of a neural network for NAFLD diagnosis

Abstract

Non-Alcoholic Fatty Liver Disease (NAFLD) affects about 20-30% of the adult population in developed countries and is an increasingly important cause of hepatocellular carcinoma. Liver ultrasound (US) is widely used as a noninvasive method to diagnose NAFLD. However, the intensive use of US is not cost-effective and increases the burden on the healthcare system. Electronic medical records facilitate large-scale epidemiological studies and, existing NAFLD scores often require clinical and anthropometric parameters that may not be captured in those databases. Our goal was to develop and validate a simple Neural Network (NN)-based web app that could be used to predict NAFLD particularly its absence. The study included 2970 subjects; training and testing of the neural network using a train-test-split approach was done on 2869 of them. From another population consisting of 2301 subjects, a further 100 subjects were randomly extracted to test the web app. A search was made to find the best parameters for the NN and then this NN was exported for incorporation into a local web app. The percentage of accuracy, area under the ROC curve, confusion matrix, Positive (PPV) and Negative Predicted Value (NPV) values, precision, recall and f1-score were verified. After that, Explainability (XAI) was analyzed to understand the diagnostic reasoning of the NN. Finally, in the local web app, the specificity and sensitivity values were checked. The NN achieved a percentage of accuracy during testing of 77.0%, with an area under the ROC curve value of 0.82. Thus, in the web app the NN evidenced to achieve good results, with a specificity of 1.00 and sensitivity of 0.73. The described approach can be used to support NAFLD diagnosis, reducing healthcare costs. The NN-based web app is easy to apply and the required parameters are easily found in healthcare databases.

Figure 1

Visualizing data with t-SNE.

Figure 2

ROC curve and AUC neural network in training.

Figure 3

ROC curve and AUC neural network in test phase.

Figure 4

Confusion matrix values during the test. 0 indicates no presence of NAFLD, 1 indicates presence of NAFLD.

Figure 5

Histogram of feature relevance of anthropometric and biochemical parameters considered during training.

Figure 6

Histogram of feature relevance of anthropometric and biochemical parameters considered during test.

Figure 7

Global interpretation using Shapley values of anthropometric and biochemical parameters considered during training.

Figure 8

Global interpretation using Shapley values of anthropometric and biochemical parameters considered during test.

Figure 9

Application of LIME for diagnosis of sick subject during the test.

Figure 10

Application of LIME for diagnosis of healthy subject during the test.