Prediction of Liver Disease using Deep Learning Methods

Abstract

The liver is one of the largest organs in the human body and it performs more than 500 functions in the human body. It also supports most of the organs, which are vital for our survival. Liver disease is a major health challenge nationally and globally. Generally, liver diseases are detected too late and raise the complexity of treatment. Early diagnosis is essential for preventing serious damage to the liver and for decreasing healthcare costs. However, the accuracy of the conventional diagnostic techniques for early prediction of liver disease is currently not at a satisfactory level. Due to the unavailability of enough specialist doctors and the high healthcare cost, people in third-world countries like Bangladesh remain away from regular health checkups, and therefore, like other serious diseases, liver disease is not detected early. Predictive models can serve a promising role in detecting liver disease, primarily before going to a doctor. In recent years, machine learning (ML)/deep learning (DL) models have become a promising technique for improving the diagnosis of human diseases, including liver disease. Therefore, we developed a deep learning-based method for early prediction of liver disease using demographic and clinical data downloaded from Kaggle.com. Several DL models including convolutional neural networks-long short-term memory (CNN-LSTM), long short-term memory (LSTM), gated recurrent units (GRU), categorical boosting (CatBoost), and deep neural network (DNN) were trained and among these models LSTM was selected as the best model based on the evaluation metrics like accuracy, precision, recall (sensitivity), specificity, F1-Score, receiver operating characteristic (ROC) curve, and area under curve (AUC). The prediction accuracy of LSTM was 99% with excellent discrimination (ROC-AUC 0.9991) and feature interpretability. The highest contribution in liver disease was confirmed by the key predictors Alkaline Phosphotase, SGPT Alamine Aminotransferase, and SGOT Aspartate Aminotransferase.

IJSS, Vol. 26(1), March, 2026, pp 53-64