Optimizing HCV Diagnosis and Hospital Resource Management Using Machine Learning Techniques

Shilpi Bisht, Neeraj Bisht, Bishwajeet Kumar Pandey, Vedant Patil, Sonalika Srivastava, Anshul Srivastava

2024

Closed access Metadata (CSL-JSON)
DOI 10.1109/delcon64804.2024.10866596
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

The aim of this research effort is to improve hospital profitability and management by using cutting-edge machine learning algorithms to accurately diagnose the Hepatitis C virus (HCV). Given that over 80% of individuals have no symptoms and that HCV can induce potentially fatal liver malignancies, this virus poses a serious threat to public health. Efficient hospital administration and resource optimization depend on timely and reliable HCV prediction. The study we conducted makes use of the ‘HCV’ dataset from the UCI Machine Learning Repository, which contains laboratory results from 615 cases involving HCV patients and blood donors, some of which have missing values. The dataset includes 13 attributes altogether and includes 533 instances of blood donors, 7 suspected cases, and 75 instances of HCV patients. We used Linear Regression to fill in the missing data to keep our predictive models accurate and reliable. We selected to use Analysis of Variance (ANOV A) for feature selection since it is a powerful tool for evaluating the correlations between features, which allowed us to increase model efficiency and decrease computation time. The decision tree algorithm, which is resilient in HCV identification, attained the greatest accuracy of 98.78% among the many machine learning techniques used, both before and after applying ANOV A. This paper presents novel strategies that can greatly increase diagnostic efficiency and accuracy in hospital settings. Examples of these strategies include the use of ANOV A for feature selection and Linear Regression for managing missing data. We tried to improve patient outcomes and maximize hospital resource use by utilizing machine learning for improved HCV prediction, which ultimately drives profitability. To determine which machine learning algorithm is the most successful, future research and real-world applications in comparable medical situations can benefit greatly from a comparative examination of several algorithms.