NAFLDPNet: Non-Alcoholic Fatty Liver Disease Prediction using Deep Learning Model with AdaBoost Algorithm for Ultrasound Images

Abstract

Non-alcoholic disease detection has become one of the major areas of research in recent years. Modern lifestyle changes, including unhealthy eating habits, environmental factors, and stress, have led to increased cases of obesity and diabetes, which in turn contribute to various health complications. While alcohol consumption is a known cause of Fatty Liver Disease (FLD), detecting this condition in non-alcoholic individuals remains a significant challenge. Early diagnosis is particularly difficult because the initial symptoms of FLD closely resemble those of other illnesses, often resulting in misdiagnosis or delayed treatment. Studies indicate that about 30% of FLD patients experience a sudden worsening of the condition, leading to severe outcomes such as heart attacks, strokes, or even death. Therefore, timely and accurate diagnosis of Non-Alcoholic Fatty Liver Disease (NAFLD) based on symptoms like abdominal pain, fatigue, and unexplained weight loss is crucial. This paper focuses on identifying NAFLD using pathological and genomic data analyzed through advanced Deep learning techniques. Several algorithms were implemented to evaluate their effectiveness in diagnosing the disease. The proposed approach based on the usage of the convolutional neural network NAFLDPNet for feature extraction of a complex nature and decision-making by the AdaBoost classifier is hence introduced. Extremely high capability of feature extraction by NAFLDPNet made its implementation reach a detection accuracy of 92%, in general 5% better than earlier practices. Results have proven that this approach can extract unique features from ultrasound images, which would be useful in diagnosing and managing fatty liver diseases. This could be instrumental in helping the physicians to be right on the spot in the accuracy and quality of diagnosis.