Data Analysis of COVID-19 Hospital Records Using Contextual Patient Classification System

Annals of Data Science https://doi.org/10.1007/s40745-022-00378-9 Data Analysis of COVID-19 Hospital Records Using Contextual Patient Classification System Vrushabh Gada1 · Madhura Shegaonkar1 · Madhura Inamdar1 · Sharath Dinesh1 · Darshan Sapariya1 · Vedant Konde1 · Mahesh Warang1 · Ninad Mehendale1 Received: 14 April 2021 / Revised: 1 November 2021 / Accepted: 19 February 2022 © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022 Abstract Humanity today is suffering from one of the most dangerous pandemics in history, the Coronavirus Disease of 2019 (COVID-19). Although today there is immense advancement in the medical field with the latest technology, the COVID-19 pandemic has affected us severely. The virus is spreading rapidly, resulting in an escalation in the number of patients admitted. We propose a contextual patient classification system for better analysis of the data from the discharge summary available from the research hospital. The classification was done using the Knuth–Morris–Pratt algorithm. We have also analyzed the data of COVID-19 and non-COVID-19 patients. During the analysis, studies on the medicines, medical services and tests, pulse count, body temperature, and the overall effect of age and gender was done. The death versus survival ratio for the COVID-19 positive patients has also been studied. The classification accuracy of the contextual patient classification system achieved was 97.4%. The combination of data analysis and contextual patient classification will be helpful to all the sectors to be better prepared for any future waves of the COVID-19 pandemic. Keywords Data analysis · Patient classification system · Contextual search 1 Introduction A catastrophic virus originated in early December 2019, in the Wuhan province of China. Later on, it became a worldwide crisis termed Coronavirus Disease of 2019 (COVID-19) by the World Health Organization (WHO) which is still affecting the world [1]. COVID-19 is still a serious challenge for doctors and hospitals. Even though B Ninad Mehendale 1 K. J. Somaiya College of Engineering, Mumbai, Maharashtra 400077, India 123 Annals of Data Science hospitals are trying their best to overcome this difficult situation, this pandemic is becoming more severe day by day as the number of variants is increasing. The COVID-19 pandemic has resulted in uncontrollable havoc in India. Since this was an unexpected circumstance, many local hospitals were not prepared to handle this crisis. The number of patients getting admitted because of COVID-19 is still increasing rapidly and this has caused a strain on hospital resources like ventilators, beds, medication (drugs), ICU beds, oxygen supply, etc. [2]. It makes the situation even more difficult for doctors and related staff such as nurses, ward boys, etc. This chaotic situation has majorly affected the patients as well. The proper allocation of resources has become a tough challenge for hospitals. Because of this, there is a possibility that many patients may not get proper treatment. If the trends in the current situation of the COVID-19 pandemic in terms of patient condition and availability of hospital resources are studied and analyzed correctly, it can help in the organized planning of any future waves of the COVID-19 pandemic [3]. This will eventually help in quick decision-making and proper allocation of the hospital resources.Data science is one of the tools to get the trends from a large dataset. Data science uses scientific methods and algorithms on unstructured data to extract useful insights, which help different businesses, health care, and other organization to improve their goods and services [4]. In India, different hospitals have different ways and software for maintaining their patient records [5]. A centralized system of maintaining the records is required. For proper resource management, we need the history of a patient to be presented in a wellorganized manner. There are a good deal of software already available that can be used for hospital resource management if organized data is present. The patient summary written by doctors varies from doctor to doctor [6]. Hence, we need a context-based patient classification system that can give segregated data which can be useful for hospital resource allocation. In our proposed method, data analysis is done on the anonymous data provided by a local hospital. This data was present in an unorganized form. The received raw data from the hospital contained eight different databases as excel sheets. Out of the eight databases provided by the hospital, seven were used. They named the seven sheets as patient list, registration list, ward list, medicine list, service list, test list, discharge summary list. We then organized this data and passed it as an input to the contextual patient classification system. The organized data was given to the contextual patient classification system to classify COVID-19 and non-COVID-19 patients. The classification was done using the KMP algorithm [7]. The classification that was done helped us in performing a comparative analysis between the COVID-19 and non-COVID-19 patient characteristics. We could compare the COVID-19 and non-COVID-19 patients based on the effect of gender, age, and services provided to them by the hospitals in terms of treatment. The death versus survival ratio of COVID-19 patients was obtained based on differences in gender and differences in age. This classification and comparison will help in the early prediction for the resource allocation and treatment process of COVID-19 patients using the data present in the discharge summary section of the organized data. Figure 1a shows the conceptual diagram of the process of data analysis and contextual patient classification system. The data is filtered and arranged in an organized 123 Annals of Data Science Fig. 1 a Concept diagram of the proposed method for data analysis. The unorganized data obtained from the hospital was filtered with the help of python programming. The filtration resulted in an organized representation of the raw data given by the hospital. In the organized data, the data for attributes related to hospital services, medicines, and discharge details were mapped against the unique MR numbers of each patient. The organized dataset was given to a contextual patient classification system. The discharge summary list from the organized dataset was then used to classify COVID-19 and non-COVID-19 patients. This classification further helped in data analyses and visualizing the differences in various aspects between COVID-19 and non-COVID-19 patients. b Based on the discharge summary of the patients obtained from the records provided by the hospital, the patients were segregated into non-COVID-19 and COVID-19. This was done using a contextual patient classification system that used the KMP algorithm for pattern mapping manner using (...truncated)