Liver Disease Detection

Abstract

Intelligent predictive systems are showing a greater level of accuracy and effectiveness in early detection of critical diseases like cancer and liver and lung disease.Predictive models assist medical practitioners in identifying the diseases based on symptoms and health indicators like hormone,enzymes,age,bloodcounts,etc.This study proposes a framework to use classification models to accurately detect chronic liver disease by enhancing the prediction accuracy through cutting-edge analytics techniques.The article proposes an enhanced framework on the original study by Ramana et al. (2011).It uses evaluation measures like Precision and Balanced Accuracy to choose the most efficient classification algorithm in INDIA and USA patient datasets using various factors like enzymes,age,etc.Using Youden’s Index, individual thresholds for each model were identified to increase the power of sensitivity and specificity.A framework is proposed for highly accurate automated disease detection in the medical industry,and it helps in strategizing preventive measures for patients with liver diseases.

Applications of edge analytics: a systematic review

“…The most often domain is healthcare. More concretely in our set of SLR's primary studies we have papers using edge analytics to help cancer treatment (P11 [5]), patient monitoring (P19 [29], P24 [33], P35 [30]), EEG analytics (P20 [41]), e-healthcare (P31 [37]), voice disorder (P35 [30]), liver disease detection (P38 [36]). The labels in parentheses are the codes of the primary studies listed in Table 1 and Table 2 of this work (selected primary studies list of SLR on edge analytics part 1 and part 2).…”

Investigating the Prediction of Breast Cancer Diagnosis by Use of Support Vector Machines

“…It successfully classified the lesion with an sensitivity of 100%, specificity of 60%, accuracy of 85.7%, and training time of 0.8507 seconds (Arora et al, 2020). Pan et al (2022) reviewed various machine learning algorithms and the scope for optimization of the same on liver cancer. Separate thresholds for each model were reported using Youden's index to enhance their sensitivity and uniqueness.…”