Abstract
Hospital readmission systems increase the efficiency of initial treatment at hospitals. This chapter proposes a novel prediction model for identifying risk factors using machine learning techniques, and the proposed model is tested using 10-fold cross-validation for generalization and finds hidden patterns in the diagnosis, medications, lab test results, and basic characteristics of patients related to readmissions. This model predicts a statistically problem solving using searching patterns. Based on the findings of this study, for the given dataset, pruning dataset manifested the most accurate prediction of readmissions to the hospital with 94.8% accuracy for patients admitted in a year.
TopIntroduction
Prevention is better than cure, Benjamin Franklin’s famous aphorism, is worth recalling at this troubling time for healthcare and the economy especially in Covid-19 pandemic (Abdel et al., 2021). There are various schemes sponsored by Indian government under flagship Ayushman Bharat National Health Protection Mission (AB-NHPM) hybrid of two major health initiatives, namely Health and wellness Centres (HWC) and National Health Protection Scheme (NHPM). This research explores the effectiveness of preventive care models using Intelligent Healthcare System to predict early hospital readmission in Indian healthcare system. In this research study, a novel Intelligent Software System for preventive care is proposed for reducing hospital readmission by machine learning analytics techniques on patients of high and low-risk categories using just-in-time deduction. The just-in-time (JIT) analytics not only ensure best readmission quality, but also handles patient’s readmission by identifying and prioritizing them according to medical aid needed so patients with higher comorbidities can be given better medical facility based on their past medical history like lab tests, medicines thereby enhancing reliability of the intelligent software system. Intelligent Software System (ISS) comprises programs, methodologies, rules and related documentation and research that empower the client to collaborate with a computer, its equipment i.e. hardware, or perform errands. It comprises 4 V’s and suffices the first V (Volume) of Big Data. The speed of healthcare data created from patient encounters and patient monitors are increasing, in and out of the clinic - second V (Velocity). Over 80 percentage of medical data resides in unstructured formats, such as doctors’ notes, images, and charts from monitoring instruments – third V (Variety) and Fourth V (Veracity) deals with unsure or vague data. Most healthcare data from clinic and hospital records is afflicted with errors, as while entering data, technicians frequently attach research to the wrong person’s record or copy research incorrectly. This section provides the context of the study research and its aim and objectives. It then demonstrates the significance of this research. Thus, the role played by the intelligent system is appreciated in the healthcare industry as Intelligent Software System can handle huge volumes of data, the amount of research with respect to the capacity for its storage and management in patient readmission cases.
TopBackground
Indian Healthcare system needs major reinvention based on income levels difference, ageing population, rising health awareness and positive outlook towards preventive healthcare are expected to increase demand of healthcare services in near future (Abdelrahman et al., 2020) thus making healthcare industry a critical and fastest growing industry in India with expectation to touch $280 billion in 2025. With the unscheduled dawn of the digital era and massive growth in healthcare, the vast amount of data can be anticipated from different health science data sources including data from patient electronic records, claims system, lab test results, pharmacy, social media, drug research, gene sequencing, home monitoring mobile apps etc. This data is called Big Data and Big data analytics can possibly change the way healthcare providers utilize modern innovations to extract knowledge from their medical data repositories and settle on educated choices. The new trend of medical data digitization is leading to an optimum model change in the healthcare industry As a result the healthcare industry is experiencing an increment in sheer volume of data regarding unpredictability, timeliness and diversity. Successfully acquiring and effectively analyzing a variety of healthcare data for a long period of time can shed light on a significant number of approaching healthcare challenges as and connected to Aarogya Setu and this requires identifying high-risk patients at the time of discharge from hospital.
Figure 1.
Four dimensions of big data in designing intelligent software system for Indian healthcare
Key Terms in this Chapter
Model Validation: It is a phase of machine learning that quantifies the ability of an ML or statistical model to produce predictions or outputs with enough fidelity to be used reliably to achieve business objectives.
Deep Learning: Deep learning is a subset of machine learning, which is essentially a neural network with three or more layers. These neural networks attempt to simulate the behavior of the human brain—albeit far from matching its ability—allowing it to “learn” from large amounts of data.
Model Testing: In machine learning, model testing is referred to as the process where the performance of a fully trained model is evaluated on a testing set. The testing set consisting of a set of testing samples should be separated from the both training and validation sets, but it should follow the same probability distribution as the training set. Each testing sample has a known value of the target. Based on the comparison of the model’s predicted value, and the known target, for each testing sample, the performance of the trained model can be measured. There are a number of statistical metrics that can be used to assess testing results including mean squared errors and receiver operating characteristics curves. The question of which one should be used is largely dependent on the type of models and the type of application. For a regression (Regression Analysis) model, the standard error of estimate is widely used.
Machine Learning: Machine learning is a tool used in health care to help medical professionals care for patients and manage clinical data. It is an application of artificial intelligence, which involves programming computers to mimic how people think and learn.
Ayushman Bharat-National Health Protection Mission (AB-NHPM): Ayushman Bharat Yojana (ABY) is a central government-funded free healthcare coverage scheme. The scheme is focused on nearly 11 crore poor and vulnerable families in rural and urban India. It is the largest scheme of its kind in the world. ABY envisions a two-pronged, unified approach by both government and private hospitals, to provide a comprehensive healthcare on primary, secondary and tertiary levels. This is planned to be accomplished through Health and Wellness Centres (HWCs) and Pradhan Mantri Jan Arogya Yojana (PM-JAY).
Particle Swarm Optimization (PSO): It is an artificial intelligence (AI) technique that can be used to find approximate solutions to extremely difficult or impossible numeric maximization and minimization problems.
Support Vector Machine: A support vector machine (SVM) is a type of deep learning algorithm that performs supervised learning for classification or regression of data groups. In AI and machine learning, supervised learning systems provide both input and desired output data, which are labelled for classification.