In the recent years Deep learning algorithms have emerged as a potential means for the diagnosis of medical diseases owing to their capability to extract composite features and patterns from huge datasets. The results of the proposed work demonstrate the efficiency of the Single Layer Perceptron, Multi-Layer Perceptron and auto encoder algorithms through extensive validation in precisely detecting the early signs of liver failure. The efficacy of the algorithms are compared based on performance metrics such as accuracy, F1 score, and recall. The comparative analysis shows that the performance of Multi-Layer Perceptron is superior. The highest accuracy is obtained by MLP as it has True Positive of 1.0, True Negative of 0.975, False Positive of 0.024 and False Negative of 0.0 leading to the accuracy 99.41 and f1 score is obtained as 99.61 and has Precision of 99.23 and Recall of 99.41.