SEGMENTATION of COVID-19 POSITIVE PATIENTS REGARDING SYMPTOMS AND COMPLAINTS

Abstract

The COVID-19 has spread rapidly among people living in all around the world and become a global threat. COVID-19 is approaching approximately 46 million cases worldwide according to the World Health Organization (WHO). There are limited number of COVID-19 test kits because of the rapid increasing cases daily. The fatality rate of ill patients with COVID-19 is very high in all around the world. Therefore, it is critical to cluster COVID-19 cases by applying clustering methods and provide the features of each. In this paper, we present symptom statistics of COVID-19 diagnosed patients to be used to foresee whether a patient will suffer through the illness severely or not. A clustering model by applying Fuzzy C-Means and PCA data reduction and visualization of data in a scatter diagram is also presented in the study. Clustering results shows patients may be segmented as risky or not in terms of the symptoms observed. We used the complaints and symptoms of 1.313 PCR-confirmed COVID-19 positive patients admitted to a university hospital in Istanbul. The findings from clustering method suggest that weakness, cough and sore throat were the most common COVID-19 symptoms and all of symptoms are separated into 3 clusters. Herein we report which symptoms are serious that may lead patients to critical situation.

Keywords

• COVID-19
• Coronavirus
• Fuzzy C-Means
• Segmentation

References

1. [1] World Health Organization (WHO) Homepage, https://www.who.int/emergencies/diseases/novel- coronavirus-2019, last accesses 2020/11/03.
2. [2] Huang C, Wang Y, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China, Lancet 2020; 395 (10223), 497–506.
3. [3] Singhal T. A review of coronavirus disease-2019 (COVID-19), Indian J. Pediatr 2020; 87, 281–286.
4. [4] Yifan T, Ying L, Chunhong G et all. Symptom Cluster of ICU Nurses Treating COVID-19 Pneumonia Patients in Wuhan, China, Journal of Pain Pain and Symptom Management, 2020; 60.
5. [5] Geva A, Patel MM, Newhams MM, et all. Data-driven clustering identifies features distinguishing multisystem inflammatory syndrome from acute COVID-19 in children and adolescents, EclinicalMedicine, 2021; 40.
6. [6] Afzal A, Ansari Z, Alshanrani S, et all. Clustering of COVID-19 data for knowledge discovery using c-means and fuzzy c-means, Results in Physics, 2021; 29.
7. [7] Nejatian S, Parvin H, Faraji E. Using sub-sampling and ensemble clustering techniques to improve performance of imbalanced classification, Neurocomputing, 2018; 276, pp.55–66.
8. [8] Rashidi F, Nejatian S, Parvin H, Rezaie V. Diversity based cluster weighting in cluster ensemble: an information theory approach, Artif Intell, 2019; 52(2), 1341–68.

9. [9] Niu H, Khozouie N, Parvin H, Alinejad-Rokny H, Beheshti A, Mahmoudi MR. An ensemble of locally reliable cluster solutions, Appl Sci, 2020; 10(5), 1891.
10. [10] Bagherinia A, Minaei-Bidgoli B, Hossinzadeh M, Parvin H. Reliability-based fuzzy clustering ensemble, Fuzzy Sets Syst, 2020.
11. [11] Mojarad M, Nejatian S, Parvin H, Mohammadpoor M. A fuzzy clustering ensemble based on cluster clustering and iterative fusion of base clusters, Appl Intell, 2019; 7, 2567–81.
12. [12] Dunn JC. A fuzzy relative of the ISODATA process and its use in detecting compact well-separated clusters, J Cybern, 1973; 3(3), 32-57.
13. [13] Silahtaroğlu G. Veri Madenciliği Kavram ve Algoritmaları (4. Basım). Papatya Bilim Yayıncılık. 2019.
14. [14] Lam C, Calvert J, Siefkas A, et all. Personalized stratification of hospitalization risk amidst COVID-19: A machine learning approach, Health Policy and Technology, 2021; 10.