Next Article in Journal
Two-Dimensional Convolutional Neural Network for Depression Episodes Detection in Real Time Using Motor Activity Time Series of Depresjon Dataset
Previous Article in Journal
Biomechanics of Pulmonary Autograft as Living Tissue: A Systematic Review
Previous Article in Special Issue
Identification of Secondary Biomechanical Abnormalities in the Lower Limb Joints after Chronic Transtibial Amputation: A Proof-of-Concept Study Using SPM1D Analysis
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:

MEF: Multidimensional Examination Framework for Prioritization of COVID-19 Severe Patients and Promote Precision Medicine Based on Hybrid Multi-Criteria Decision-Making Approaches

Karrar Hameed Abdulkareem
Mohammed Nasser Al-Mhiqani
Ahmed M. Dinar
Mazin Abed Mohammed
Mustafa Jawad Al-Imari
Alaa S. Al-Waisy
Abed Saif Alghawli
8,* and
Mohammed A. A. Al-Qaness
College of Agriculture, Al-Muthanna University, Samawah 66001, Iraq
College of Engineering, University of Warith Al-Anbiyaa, Karbala 56001, Iraq
Center for Advanced Computing Technology, Faculty of Information and Communication Technology, Universiti Teknikal Malaysia Melaka, Malacca 75300, Malaysia
Computer Engineering Department, University of Technology- Iraq, Baghdad 19006, Iraq
College of Computer Science and Information Technology, University of Anbar, Anbar 31001, Iraq
Department of Medical Laboratory Techniques, Al-Mustaqbal University College, Hillah 51001, Iraq
Computer Technologies Engineering Department, Information Technology Collage, Imam Ja’afar Al-Sadiq University, Baghdad 10064, Iraq
Computer Science Department, College of Sciences and Humanities, Prince Sattam Bin Abdulaziz University, Aflaj 16278, Saudi Arabia
College of Physics and Electronic Information Engineering, Zhejiang Normal University, Jinhua 321004, China
Faculty of Engineering, Sana’a University, Sana’a 12544, Yemen
Author to whom correspondence should be addressed.
Bioengineering 2022, 9(9), 457;
Submission received: 9 August 2022 / Revised: 2 September 2022 / Accepted: 4 September 2022 / Published: 8 September 2022
(This article belongs to the Special Issue Smart Sensing for Health Informatics)


Effective prioritization plays critical roles in precision medicine. Healthcare decisions are complex, involving trade-offs among numerous frequently contradictory priorities. Considering the numerous difficulties associated with COVID-19, approaches that could triage COVID-19 patients may help in prioritizing treatment and provide precise medicine for those who are at risk of serious disease. Prioritizing a patient with COVID-19 depends on a variety of examination criteria, but due to the large number of these biomarkers, it may be hard for medical practitioners and emergency systems to decide which cases should be given priority for treatment. The aim of this paper is to propose a Multidimensional Examination Framework (MEF) for the prioritization of COVID-19 severe patients on the basis of combined multi-criteria decision-making (MCDM) methods. In contrast to the existing literature, the MEF has not considered only a single dimension of the examination factors; instead, the proposed framework included different multidimensional examination criteria such as demographic, laboratory findings, vital signs, symptoms, and chronic conditions. A real dataset that consists of data from 78 patients with different examination criteria was used as a base in the construction of Multidimensional Evaluation Matrix (MEM). The proposed framework employs the CRITIC (CRiteria Importance Through Intercriteria Correlation) method to identify objective weights and importance for multidimensional examination criteria. Furthermore, the VIKOR (VIekriterijumsko KOmpromisno Rangiranje) method is utilized to prioritize COVID-19 severe patients. The results based on the CRITIC method showed that the most important examination criterion for prioritization is COVID-19 patients with heart disease, followed by cough and nasal congestion symptoms. Moreover, the VIKOR method showed that Patients 8, 3, 9, 59, and 1 are the most urgent cases that required the highest priority among the other 78 patients. Finally, the proposed framework can be used by medical organizations to prioritize the most critical COVID-19 patient that has multidimensional examination criteria and to promptly give appropriate care for more precise medicine.

Graphical Abstract

1. Introduction

The global COVID-19 epidemic outbreak has a serious impact on people’s daily life and healthcare systems. To stop the spread of this epidemic, the entire globe has struggled and fought [1]. Therefore, the growing cases of COVID-19 present a new issue to the healthcare system, together with the expected daily service delivery, where it is expected that the rate of death would be significant and the diagnosis procedure will take a long time [2]. The early diagnosis and isolation of infected people are crucial in combating the COVID-19 pandemic. Furthermore, it is important to make sure that the right response is provided for critical cases in order to decrease the mortality rate [3].
Medically, examination criteria such as C-reactive protein (CRP) levels can be employed in the early diagnosis of pneumonia, and patients presented with severe pneumonia had high CRP levels [4]. Additionally, shortness of breath is caused by hypoxia, which is defined as SpO2 90%. However, with COVID-19, the normal level of SpO2 is decreased and may even drop to 70%, 60%, or 50%, even if the patient did not have a feeling of being out of breath. An important component in understanding and managing patient care is SpO2. It compares how much hemoglobin is currently bound to oxygen to how much hemoglobin remains unbound. A pulse oximeter, a non-invasive medical gadget, is placed over a person’s finger to measure blood oxygen saturation. It is commonly utilized in a hospital’s intensive care unit (ICU), operation unit, and postoperative ward. Other examinations, such as normal WBC or low total WBC, high C-reactive protein (CRP), neutrophil, lymphocyte ratio, lymphopenia, bilateral pneumonia in CXR, low procalcitonin, ground-glass opacity (GGO), significant elevation of D-dimer, serum ferritin level, and crazy paving appearance in chest CT scan, indicate the existence of COVID-19 [5]. Dry cough, fever, headache, tiredness, sore throat, vomiting, sneezing, dyspnea, myalgia, nasal congestion, and rhinorrhea are the most common critical symptoms. Critical consequences such as cardiac damage, pulmonary edema, septic shock, and acute renal injury occur in patients with severe COVID-19 infection [6,7].
All of these criteria are crucial in monitoring a COVID-19 patient’s status, but the large number of these indicators poses a challenge for doctors and emergency services to decide or predict the patients’ severity [8]. Furthermore, the old and those suffering from chronic diseases such as diabetes, cancer identification, chronic respiratory disease, cardiovascular disease, and chronic respiratory disease are at a higher risk of contracting a serious infection [9]. A new approach of prioritizing services could be a solution to such challenges. However, such difficulties are increased when patients have several examination results, each with a separate set of references. Due to the large number of COVID-19 patients, specifically in the United States, and various European countries, the treatments and healthcare provided to infected individuals should be prioritized [10]. The issues described in the literature have emphasized the challenge of distinguishing between mild and severe health conditions for infected patients in terms of examination criteria specific to each patient [11,12].
Prior studies in a specific disease perspective have demonstrated that precision medicine and health outcome priorities of patients and healthcare professionals can differ [13]. Moreover, some studies have pointed to a disparity between what practitioners understand to be their patients’ priorities and what their patients actually prioritize [14]. Considering the various problems associated with COVID-19 [15], approaches that could triage COVID-19 patients may be able to help prioritize treatments for patients who are at high risk of serious disease. Furthermore, effective prioritization plays critical roles in precision medicine. COVID-19 severity levels are as follows: ordinary, mild, critical, and severe [16]. Severe cases require extra medical energy and attention than minor and regular ones. A high percentage of false-positive severe or critical cases may overload healthcare facilities (i.e., beds in the intensive care unit). Delays in reporting severe or urgent cases might also lead to patients at a higher risk of death receiving delayed care. As a result, finding acute conditions as soon as possible is crucial in order to deploy services and intensify care [17].
This goal could be fulfilled by employing a prioritizing approach that ranks patients and displays the priority they are given based on their situations in order to facilitate the process of finding patients with the most urgent cases. Prioritization is usually used to ensure that care is provided in an appropriate and timely way. As a result, patient prioritizing is mainly concerned with determining which patients can safely wait and which cannot. In accordance with medical principles, the main tool for determining priority should be the patient’s condition. This strategy can improve medical staff’s workflow by allowing them to focus their attention and efforts on providing the best treatment for patients who require immediate care. Other patients can wait at indoor healthcare providers (such as clinics and hospitals) in the meanwhile [18].
Multi-criteria decision-making (MCDM) approaches have played vital roles in patients’ prioritization. Two main approaches are used for this: one for defining the importance of criteria (weighting approach) and another one to state the final rank (ranking approach) for patients based on the identified criteria. The weighting methods are also divided into two directions namely subjective and objective weighting methods. The analytic hierarchy process (AHP) method [18] is used by respiratory experts to set the subjective weights for the biological laboratory examination criteria for COVID-19 patients. However, despite the success of AHP, the weighting procedure has a significant drawback, namely its inconsistency issue, as well as being affected by expert opinion, as there is no guarantee that the expert cannot produce any bias toward any examined criteria. Objective methods such as CRITIC and Entropy consider the variability in the information. The effects of decision-makers in the calculation of weights are eliminated, as objective methods’ weights are assigned by a mathematical method to the criteria [19].
For the purpose of ranking, each of the VIekriterijumsko KOmpromisno Rangiranje (VIKOR) and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) methods have been used widely for patients ranking [18,20]. TOPSIS’s guiding premise is the distance between positive and negative answers. In contrast, it ignores the significance of comparing these distances [21]. The VIKOR procedure is compatible with the discrete alternative because it takes into account the most realistic way of dealing with such situations in the real world. The main advantage of VIKOR is that the best alternative can be quickly identified. VIKOR is thus appropriate in circumstances with several alternatives and attributes [22]. VIKOR utilizes an approach that compromises the priority for various response optimizations. This technique simply ranks the various alternatives based on their closeness to the optimal solution and quickly and accurately identifies the best model [23]. Furthermore, the combination of the weighting method and ranking method in one platform is recommended in the medical domain [20,24]. Thus, the proposed work adopted a hybrid approach based on CRITIC and VIKOR methods. The basis of integration is formed when the weights for the criteria are allocated in accordance with CRITIC. VIKOR is suggested for prioritizing patients with COVID-19.
The innovations and contributions of this work can be summarized as follow:
  • Creation of a Multidimensional Evaluation Matrix (MEM) based on real data for different COVID-19 patients’ examination characteristics, such as demographic, laboratory findings, vital signs, symptoms, and chronic conditions.
  • Based on multidimensional examination criteria and CRITIC (CRiteria Importance Through Intercriteria Correlation) approach, we provide an evaluation for the most examination criteria for sever COVID-19 patients.
  • Develop a framework for prioritization of COVID-19 severe patients on the basis of the real data of constructed MEM and hybrid decision-making approaches (CRITIC and VIekriterijumsko KOmpromisno Rangiranje (VIKOR)). Therefore, we can help health authorities prioritize sever patients by considering the results of different examination criteria for patients infected with COVID-19 to provide the desired intensive care facilities and manage the health conditions of patients.
The rest of this paper is organized as follows: Section 2 summarizes the most related studies on AI with COVID-19 diagnosis. A detailed description and an overview of the steps for the proposed multidimensional framework for COVID-19 patients’ prioritization based on the most urgent cases are presented in Section 3. Section 4 presents the results and a discussion of the prioritization framework of the COVID-19 patients based on the CRITIC weighting and VIKOR ranking methods. Section 5 confirms the validity of proposed work results based on a sensitivity analysis. Finally, Section 6 concludes the study findings, and the future direction presented in this section.

2. Related Works

Infectious disease modeling is a method that has been used to explore the mechanisms by which diseases spread, forecast the future course of an outbreak, and assess epidemic control strategies [25]. Mathematics has permeated biology in a variety of ways, including statistics in experimental design; pattern-seeking in bioinformatics; models in evolution, ecology, and epidemiology; and much more [26,27].
In addition to applying mathematical models and AI models for assessing and predicting the significance of factors for evaluating healthcare systems, there are several publications in which the evaluation of healthcare systems is performed by using multi-criteria decision-making (MCDM) tools. Utilizing MCDM in prioritization is a popular topic and a complex issue for patients with COVID-19, especially for multidimensional criteria [28]. The term “multidimensional” is used in the sense that no single value is sufficient alone to determine which patients should receive hospital resources. However, this requires building a robust framework that can be adapted [29]. For more clarity, the traditional way of treating people “first come, first served” should not apply during the pandemic in general and COVID-19 specifically [30]. Therefore, prioritizing some factors (multidimensional factors) should be a better approach to consider the large number of patients.
Prioritization methods have solved many challenges, especially when patients have multi-biological laboratory examination results and vital functions called multi-perspective criteria [31,32]. In clinical prioritization, MCDM methods have proven effective in many healthcare data-focused applications. M. Abdel-Basst et al. applied a model to differentiate between COVID-19 and four other viral chest diseases in an uncertain environment, using the viruses’ primary symptoms and CT scans. The proposed model employs the best worst method (BWM) and the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) [33].
We next discuss several studies that deal with the aspects that we explained earlier: MCDM, prioritization, and the multidimensional aspect. Moreover, these related works are discussed in terms of the aim of the research, case study utilized, and criteria used, as shown in Table 1.
Recently, Albahri et al. published two detection-based prioritization frameworks. The first study presented a new multi-biological laboratory examination framework for prioritizing patients with COVID-19 based on integrated MCDA methods. The experiment was conducted based on three phases: patient datasets containing eight biological laboratory examination criteria for six patients with COVID-19 were derived and discussed. The analytic hierarchy process (AHP) method was then used by respiratory experts to set the subjective weights for the biological laboratory examination criteria, and in the final phase, the VIekriterijumsko KOmpromisno Rangiranje (VIKOR) method was employed to prioritize patients in the context of individual and group decision-making (GDM). The second study developed a new integrated decision-making framework for prioritizing COVID-19 patients and detecting the health conditions of asymptomatic carriers. The findings suggest and explain the multiple advantages of the proposed framework for detecting/recognizing patients’ health issues prior to discharge. In the evaluation and benchmarking of COVID-19 ML models, M. A. Mohammed et al. developed a methodology for selecting the best COVID-19 diagnostic model based on Entropy and TOPSIS methods, and they evaluated the proposed methodology by using 50 samples from the COVID-19 Chest X-ray Dataset.
On the other hand, MCDM methods have been utilized in non-clinical prisonization. A new TOPSIS MCDM approach and GMDH can identify and select the significant risk factors and continuously monitor death due to COVID-19 [34]. Another work that was recently published talks about the prioritization of activities. In this work, the authors attempted to identify several activities that should not be performed during the pandemic time. Some selected activities were explored in this paper, and their relevance in preventing COVID-19 was also evaluated. These activities were considered criteria [35].
However, as can be concluded from Table 1, we can raise several issues that we must take into account in order to achieve a robust and reliable prioritization system that is beneficial for health systems:
Real data: With the spread of SARS-CoV-2 worldwide, understanding the basic epidemiological parameter values of COVID-19 from real-world data in mega-cities is essential for disease prevention and control [36]. Comparing them with other models depends on simulated data, which need more validation before being adopted [37].
Dependent criteria: Many studies have shown that symptoms and biomarkers are independently associated with in-hospital death and decisions about the severity of cases [38,39].
Multi-perspective (multidimensional): The identification of hospitalized COVID-19 patients at risk for severe deterioration can be performed by using risk scores that combine several factors, including age; sex; and comorbidities, i.e., diseases such as hypertension, diabetes, and cancer or tumor [40,41,42]. Some risk scores also include additional markers of severity, such as low level of oxygen in the blood, clinical symptoms, and biological factors reflecting multi-organ failures [42,43,44,45,46]. However, there are many unresolved interferences between these factors, including both clinical and biological markers [47]; therefore, there is a need for a comprehensive decision matrix that is inclusive of all of the markers mentioned earlier.
Therefore, the main focus of this study was the prioritization of severe COVID-19 patients based on multidimensional examination criteria that were extracted from real data, as well as the determination of how important these criteria are.

3. The Proposed Multidimensional Framework for COVID-19 Patients’ Prioritization

This section presents a detailed description and an overview of the steps for the proposed multidimensional framework for COVID-19 patients’ prioritization based the most urgent patients. The final output of this proposed framework is patients’ prioritization ranking based on multidimensional medical sources’ criteria used in monitoring COVID-19 patients. The CRITIC weighting method is used to provide criteria weight, while the VIKOR is used for COVID-19 patients’ prioritization ranking. All components of our study are illustrated in the methodology framework in Figure 1.

3.1. Construction of Multidimensional Evaluation Matrix (MEM)

The Multidimensional evaluation matrix contains an important element of the proposed framework for COVID-19 patients’ prioritization. It consists of a list of COVID-19 patients, who are considered to be alternatives, and medical sources, which are considered to be the evaluation criteria. The real dataset that was used in our proposed study was extracted from References [8,48]. The total number of people infected with the COVID-19 virus was 78; they were diagnosed under the supervision of specialized doctors and were distributed into Al-Aziziyah Hospital in Wasit Governorate of Iraq. This dataset contains five examination criteria for COVID-19 patients, namely demographic, laboratory findings, vital signs, symptoms, and chronic conditions. Each criterion represents a single dimension in our proposed MEM, as well as MEF. Moreover, each one of these five examination criteria is categorized into sub-criteria, giving a total of 25 sub-criteria. Based on the selected sub-criteria from another study [48], we selected only 15 out of the 25 sub-criteria as the basis for our prioritization process as shown in Figure 2. For instance, the demographic dimension consists of the age of the patients, and the laboratory-findings dimension consists of four sub-criteria, which are lymphocyte count, C-reactive protein (mg/L), urea (mmol/L), and creatinine (µmol/L). The vital-signs dimension consists of two sub-criteria, namely oximetry saturation (%), and body temperature (°C), while the symptoms dimension of the patients consists of five sub-criteria, which are pleuritic chest pain, nasal congestion, cough, lost sense of smell (1/0), and lost sense of taste (1/0). Lastly the chronic-conditions dimension consists of three sub-criteria, which are heart disease, diabetes disease, and cancer. In addition, to summarize the abovementioned scenario, the evaluation criteria in the proposed MEM consist of the main and sub-criteria, which are used to monitor COVID-19 patients from five perspectives. Therefore, a doctor can benchmark the COVID-19 patients according to their urgent situation in consideration of their medical condition. Table 2 illustrates the proposed Multidimensional Evaluation Matrix (MEM) for COVID-19 patients.

3.2. Hybridization of CRITIC and VIKOR

The proposed benchmarking method and evaluation methodology were developed by using the techniques of MCDM. Based on the hybridization of the CRITIC weighting method and VIKOR ranking method, this methodology was developed for weighting, ranking, and selecting the most and least critical conditions COVID-19 patients in the proposed matrix for the doctors. The following steps are discussed below. CRITIC and VIKOR are the appropriate methods for ranking and benchmarking of COVID-19 patients. The CRITIC method is used to overcome problems relevant to (1) conflict and trade-off and (2) the encountered evaluation of multi-criteria in the suggested evaluation matrix. VIKOR is often proposed to implement criteria ranking and (3) the value of relational criteria to the suggested matrix of decisions. Consequently, it is justified to combine the CRITIC weighting method and VIKOR ranking method to rank the COVID-19 patients and benchmark them.

3.2.1. Weighting of COVID-19 Examination Criteria Based on CRITIC Method

The CRITIC (CRiteria Importance Through Intercriteria Correlation) method is one of the weighting methods which determines objective weights for criteria. The method was initially proposed by Diakoulaki et al. (1995) [49]. The method was used to derive criterion weights since it tries to determine objective weights of relative relevance in MCDM problems. The proposed approach is based on an analytical examination of the evaluation matrix in order to extract all of the information included in the evaluation criteria. In other words, objective weights are calculated by measuring each evaluation criterion’s intrinsic information. The procedure of determining criteria weights in this method incorporates both the criterion’s standard deviation and its association with other criteria.
An initial evaluation matrix is given in Equation (1), which contains n criteria and m alternatives, where xij denotes the performance of the ith alternative in relation to the jth criterion. The following notations are used to calculate the weight of the jth criteria Wj: the quantity of information included in the jth criterion is denoted by Cj, the standard deviation of the jth criterion is denoted by σ j, and the correlation coefficient between the jth and kth criteria is denoted by rjk.
The CRITIC method’s calculating stages are shown below based on these notations [50]:
1st step: The X evaluation matrix is created. It compares the performance of several alternatives based on numerous criteria:
X = x i j m n = [ X 11 X 21 X 31 X 12 X 1 n X 22 X 2 n X 32 X 3 n X m 1           X m 2 X m n ]   .         ( i = 1 , 2 , , m   a n d   j = 1 , 2 , .. , n ) .  
The performance value of the ith alternative on the jth criterion is represented by xij.
2nd step: The following equation is used to normalize the evaluation matrix:
r i j = x i j x j m i n x j m a x x j m i n .  
3rd step: Using Equation (3), compute the correlation between criterion pairs:
ρ j k = i = 1 m ( r i j r ¯ j ) ( r i k r ¯ k ) i=1 m ( r i j r ¯ j ) 2 i=1 m ( r i k r ¯ k ) 2
4th step:Using Equations (4) and (5), compute the weights of the criterion (5).
c j = σ j k = 1 n ( 1 ρ j k )
w j = c j / k = 1 n c k
where i = 1, 2, …, m; and j, k = 1, 2, …, n.

3.2.2. Ranking of COVID-19 Severe Patients Based on VIKOR Method

Typically, during the decision-making process, the weighted matrix serves as the starting point for ranking the alternatives. The prior section covers how to obtain the weights that are assigned to each criterion in the EM, giving the weighted EM. During this process, the COVID-19 patients are evaluated and eventually ranked based on the weighted EM. The ranking procedure is as follows:
1st step: Define the best f i * . value and worst f i value for all criteria functions:
f i * = m a x j f i j ,   f i = m i n j f i j
where f i j . is the ith criterion function value for the xi of alternative. An ideal positive solution increases the benefit criterion while minimizing the cost criteria, whereas the cost criterion is maximized by the ideal negative solution while minimizing the benefit criterion.
2nd step: Using the entropy approach, compute the weights of each criterion. A weight set w = w 1 ,   w 2 ,   w 3   , , w j , ,   w n from a decision-maker is accommodated in the EM. This set is equal to 1. The following formula can be used to compute the resulting matrix:
W M = w i f i * f i j f i * f i   .
A weighted EM serves as the output of this step.
3rd step: Using the following equations, calculate the values of Sj and Rj (j = 1, 2, 3, …m and i = 1, 2, 3, … n):
S j = i = 1 n w i f i * f i j f i * f i  
R j = max i w i   f i * f i j f i * f i  
where Sj and Rj denote the measure of the utility and regret for the alternative fi, and wi specifies the relative weights of the criterion.
4th step: Using the following relation, calculate the values of Qj ( j = ( 1 , 2 , , J ) ):
Q j = v ( S j S * ) S S * + ( 1 v ) ( R j R * ) R R *
where S * = min j S j   , S = max j S j   , R * = min j R j   , R = max j R j   , and v is defined as the strategy weight of “the majority of criteria” (or “the maximum group utility”) and v = 0.5.
5th step: The alternatives can be performed in this step by sorting the values and obtaining the values of S, R, and Q in ascending order. The lowest value represents the best performance.
6th step: The alternative ( a ) is provided as a solution alternative and ranked as the best solution based on the minimal Q measure.
7th step: Propose alternative A1, which is ranked the best by the metric Q (minimum) if the following two requirements are met, as a compromise solution:
  • C1. Acceptable advantage:
Q ( A 2 ) Q ( A 1 ) D Q
where A2 represents the alternative ranked second in the list by Q, DQ = 1/(m − 1), and m is the number of alternatives.
C2. Acceptable stability in decision-making:
According to this condition, the alternative A1, which is best-ranked by Q, must also be the best-ranked by S or/and R.
A set of compromise solutions are suggested if one of the conditions is not met, which consists of the following: alternatives A1 and A2 if only condition C2 is not satisfied; or alternatives A1, A2, …, AM if condition C1 is not satisfied. AM is determined by the relation Q(AM) − Q(A1) < DQ for maximum M (the positions of these alternatives are “in closeness”).

4. Results and Discussions

This section presents the results and discussion of the prioritization framework of the COVID-19 patients based on the CRITIC weighting and VIKOR ranking methods. Section 4.1 describes how the MEM data results were generated from the 78 COVID-19 patients. In Section 4.2, the CRITIC results are discussed, and the VIKOR method results are presented and discussed in detail in Section 4.3.

4.1. MEM Results

Table 3 displays the structure of the proposed EM. The top two rows of the table show the five main medical-source criteria, and the second row from the top shows the fifteen sub-criteria, while the first column from the left shows the patients categorized according to their severity; the COVID-19 patients in the list are considered as alternatives in the EM. The values in the EM indicate all the diagnosis data for a total number of 78 people who were infected with the COVID-19 virus and were diagnosed under the supervision of specialized doctors from the Wasit Governorate of Iraq. The 78 patients are divided into three main groups based on the patient’s severity, which was assessed and categorized as being severe, moderate, and mild by doctors according to the Iraqi hospitals’ criteria at admission.

4.2. CRITIC Weighting Results

In this section, we discuss the use of the CRITIC method for finding the weights of each criterion. First, the normalized evaluation matrix was determined based on Equation (2); the normalized evaluation matrix is shown in Appendix A. The last row of Appendix A shows the values of standard deviations for all criteria. The values of the correlation coefficient were then calculated, and these are shown in Appendix B. Moreover, the quantity of info (Cj) is shown in Appendix C. Finally, the criteria weights in Table 4 were determined by using Equations (4) and (5).
Based on the result shown in Table 4, Appendix A, Appendix B and Appendix C, and Figure 3, the gained values of Wj and weights of the CRITIC method for the fifteen sub-criteria were obtained by using Equations (4) and (5). Heart disease achieved the maximum CRITIC weight, at 0.118, followed by cough and nasal congestion, with CRITIC weight 0.113 and 0.107, respectively. On the other hand, creatinine achieved the minimum CRITIC weight, at 0.023. Based on the results, the medical source criteria with the highest CRITIC weights are considered to be the most important criteria, whilst the criteria with the lowest CRITIC weight are of least importance.

4.3. Ranking Results based on VIKOR Method

This subsection discusses the results of the 78 COVID-19 patients ranking based on the weighted EM. The weighted EM can be obtained by using Equations (5) and (6) and is shown in Table 5. Notably, we considered the age of the patients, lymphocyte count, C-reactive protein (mg/L), urea (mmol/L), creatinine (µmol/L), body temperature (°C), pleuritic chest pain, nasal congestion, cough, lost sense of smell (1/0), lost sense of taste (1/0), heart disease, diabetes disease, and cancer as benefit criteria, while oximetry saturation (%) is considered as cost criteria. Then Equations (7) and (8) were used to calculate alternative distances from the positive ideal solutions and negative ideal solutions. Last, based on Equation (9), the Qi values for the 78 COVID-19 patients were calculated. Hence, using VIKOR principles, the 78 COVID-19 patients were ranked based on their urgent situation in consideration of their medical condition. Table 5 shows the result of the weighted EM, while Table 6 displays the final result of VIKOR ranking.
As shown in Table 6, the first five patients, namely Patients 8, 3, 9, 59, and 1, were the most urgent cases that required the highest priority among the other 78 patients, and they obtained minimum values in terms of Qi, with 0, 0.116101, 0.125431, 0.136293, and 0.152904, respectively. Therefore, Patient 8 is ranked as the highest priority, and Patients 3, 9, 37, and 1 are next highest priority. Therefore, a rank based on Table 6 was the final outcome of the VIKOR ranking, and the validation processes are based on it.

5. Evaluation and Sensitivity Analysis

Numerous authors in their research emphasized that a sensitivity analysis in multi-criteria problems is an indispensable step to confirm the robustness of the obtained solutions [51,52]. Some authors [53,54] suggest checking the robustness of solutions in MCDA problems by changing the input parameters of the model. According to Reference [55], the changing process could be on weight coefficients of the criteria or other parameters into decision-making method. Therefore, our study followed three scenarios to show the robustness of proposed framework in the prioritization of the most critical COVID-19 cases:
First: A compromise solution is determined from the alternative that has the best ranking by the measure Q (minimum) if the following two conditions are satisfied:
Condition 1: Acceptable advantage:
D Q = 1 ( m 1 ) = 1 ( 78 1 ) = 1 77 = 0.012987013
Q ( p a t i e n t   3 ) Q ( p a t i e n t   8 ) = 0.116101 0 = 0.116101
The value of Q (Patient 3) − Q (Patient 8) ≤ DQ. Thus, the acceptable advantage conditions are satisfied.
Condition 2: Acceptable stability in decision-making.
The best ranking results, Q, are for Patient 8, obtaining the lowest Q, R, and S values with 0, 0.034901, and 0.076941, respectively, as shown in Table 6. So, it can be proved that the condition of acceptable stability in decision-making is satisfied.
Based on the abovementioned details, the two conditions mentioned in Step 7 are satisfied. This can conclude that Patient 8 exhibited the most urgent case and received the highest priority level.
Second: In accordance to Reference [56], changing the value of V in the VIKOR method was used to examine the rationality and stability of the proposed framework. Due to the high number of patients in our study, we chose only the first ten patients for the sensitivity analysis as a baseline to verify the ranking of the proposed framework. Furthermore, the V value changed nine times, from 0.1 to 0.9, and was compared to the baseline rank, as shown in Figure 4.
As stated before, this work used the disparity of V values to validate that none of them impacts the final rank for patients, as shown in Figure 4. The ranking orders of ten patients are the same, and Patient 8 is still the most critical case. Thus, the analysis study can confirm that the results acquired by using the proposed model are reliable and effective. Furthermore, the proposed framework is not sensitive to a change of VIKOR parameters.
Third: The sensitive analysis based on changing sub-criteria coefficient weights is the third method that was used to verify the proposed framework’s ranking order. In this analysis, we have changed the weight by adding an identified percentage to the original weights. Specifically, all sub-criteria have 10%, 20%, and 30% extra weight than the base weight, as shown in Figure 5.
Based on Figure 5 analysis results, it can be seen there is a high variation into ranking of Patients 2, 4, 6, and 7 but with more high priority rate than base rank. While each of Patients 3, 5, and 9 have very less variation into ranking order with one more high-ranking step. However, Patients 1 and 8 have the same ranking order for three analysis scenarios. Furthermore, still the Patient 8 is the most critical case no matter how the weights change proving the top-ranked patients achieves extraordinarily in different circumstances. It can be verified that the top-ranked-patients results from using the Multidimensional Examination Framework for the prioritization of COVID-19 severe patients is robust and effective. This study can distinguish the priorities of COVID-19 severe patients more easily, and thus help medical staff and decision-makers evaluate and identify the most critical COVID-19 case.

6. Conclusions

This study proposed a framework for prioritization of COVID-19 severe patients that have different combination examination criteria that were extracted from different resources. Apart from the existing literature, the MEF has not only considered one dimension of examination factors but multidimensional examination criteria for the prioritization of COVID-19 severe patients such as demographic, laboratory findings, vital signs, symptoms, and chronic conditions.
The five main criteria and fifteen sub-criteria were included in the construction of the multidimensional evaluation matrix. The most effective criteria and sub-criteria for the prioritization of severe COVID-19 patients were identified through given weights by the CRITIC method. Accordingly, COVID-19 patients with different examination criteria were prioritized by using the VIKOR method. From a pool of 78 patients, the proposed intelligent framework succeeded in prioritizing most of the COVID-19 severe patients with different aspects of obtained real data. However, in this study, the authors confirmed several limitations that may be considered for future work. For example, the weighting-based objective method may not consider all environmental conditions; therefore, this may affect the real importance for the examination criteria. Fifteen out of twenty-five sub-criteria were selected based on the Gini Information method for the prioritization of COVID-19 severe patients.
The proposed framework can help health authorities prioritize patients by considering the results not only from a single dimension but from multidimensions. Thus, more complex scenarios are presented for medical staff in order to prioritize the most critical COVID-19 patients. This situation could be even more difficult specifically when critical resources such as beds and ventilators (respirators) in ICUs and medical staff are becoming scarce. Besides that, the proposed methodology can aid healthcare organizations in providing the desired intensive-care services and handling patients’ medical problems. Furthermore, it can help them distinguish among health conditions for large-scale admission process.
However, the number of selected sub-criteria is varied and restricted according to the used mathematical model. Thus, this may affect the selection of the most common examination criteria that consistent of theory and practical domains. When the condition of COVID-19 patients changes, dynamic, more accurate weighting and ranking methods are needed. However, weighting methods that are based on an objective context lack consideration for the changes in a patient’s health circumstance, thus affecting the definition of important examination criteria. Therefore, such methods may not present very accurate ranking for most critical COVID-19 cases. In future work, authors should implement a more robust weighting method based on the integration of subjective and objective contexts to tackle the mentioned issues.

Author Contributions

Conceptualization, M.A.M., M.J.A.-I., A.S.A. and M.A.A.A.-Q.; methodology, K.H.A., M.N.A.-M., A.M.D. and M.A.M.; software, K.H.A. and M.N.A.-M.; validation, K.H.A., M.N.A.-M., A.M.D. and M.A.M.; formal analysis, K.H.A., M.N.A.-M., A.M.D., M.A.M., M.J.A.-I., A.S.A.-W. and M.A.A.A.-Q.; data curation, A.M.D. and M.A.M.; writing—original draft preparation, K.H.A., M.N.A.-M., A.M.D. and M.A.M.; writing—review and editing, K.H.A., M.N.A.-M., A.M.D., M.A.M., M.J.A.-I., A.S.A., A.S.A.-W. and M.A.A.A.-Q.; funding acquisition, A.S.A.-W. All authors have read and agreed to the published version of the manuscript.


This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

A real dataset was used in our proposed study that was extracted from References [8,39].


Special thanks to Al-Mustaqbal University College for the research support.

Conflicts of Interest

The authors declare no conflict of interest.

Appendix A

Table A1. The normalized MEM based on the CRITIC method.
Table A1. The normalized MEM based on the CRITIC method.
DemographicLaboratory FindingsVital SignsSymptomsChronic Conditions
PatientsAgeLymphocyte CountC-reactive Protein Urea Creatinine Oximetry Saturation Body TemperaturePleuritic Chest PainNasal CongestionCoughLost Sense of Smell Lost Sense of Taste Heart DiseaseDiabetes DiseaseCancer
Sigma (standard deviation)0.2320.2010.3820.1230.1130.1990.2310.5020.4970.4950.2880.2680.5020.4390.247

Appendix B

Table A2. The correlation coefficient values of the criteria.
Table A2. The correlation coefficient values of the criteria.
AgeLymphocyte CountC-reactive ProteinUrea Creatinine Oximetry Saturation Body Temperature Pleuritic Chest PainNasal CongestionCoughLost Sense of Smell Lost Sense of Taste Heart DiseaseDiabetes DiseaseCancer
Lymphocyte count0.0911.0000.3710.0440.084−0.1890.0280.066−0.0380.1250.0030.1500.0080.3780.218
C-reactive protein 0.3970.2851.0000.3620.232−0.6520.5080.3390.3250.1950.2270.1270.2040.2450.334
Urea 0.4190.0440.3621.0000.919−0.4570.3060.2570.2710.1710.0880.0600.0910.3630.630
Creatinine 0.1990.0840.2320.9191.0000.2840.1990.1660.1970.1310.0500.050−0.0640.2840.548
Oximetry saturation 0.5580.1890.6520.4570.2841.0000.5840.3740.3590.1690.1870.0970.2100.1940.518
Body temperature 0.4250.0280.5080.3060.1990.5841.0000.4010.2770.0690.1580.0930.1960.2390.288
Pleuritic chest pain0.4710.0660.3390.2570.1660.3740.4011.0000.1160.2210.1590.1190.2900.2490.242
Nasal congestion0.318−0.0380.3250.2710.1970.3590.2770.1161.000−0.024−0.003−0.0450.2480.2700.200
Lost sense of smell 0.0550.0030.2270.0880.0500.1870.1580.159−0.0030.2851.0000.1500.0400.1320.218
Lost sense of taste −0.1330.1500.1270.0600.0500.0970.0930.119−0.0450.1500.7511.000−0.159−0.0210.082
Heart Disease0.5120.0080.2040.091−0.0640.2100.1960.2900.2480.040−0.159−0.2151.000−0.0510.076
Diabetes Disease0.3780.3780.2450.3630.2840.1940.2390.2490.2700.132−0.021−0.0510.3991.0000.073

Appendix C

Table A3. The quantity of information (Cj).
Table A3. The quantity of information (Cj).
Medical SourceSigma SumCj
Lymphocyte count0.20113.0322.621
C-reactive protein 0.38210.8734.148
Urea 0.12310.4751.293
Creatinine 0.11310.7211.213
Oximetry saturation 0.1999.1671.827
Body temperature 0.23110.2292.359
Pleuritic chest pain0.50210.5305.284
Nasal congestion0.49711.5295.733
Lost sense of smell 0.28812.2513.524
Lost sense of taste 0.26812.6783.400
Heart Disease0.50212.6156.329
Diabetes Disease0.43910.8674.776


  1. Mohammed, M.A.; Al-Khateeb, B.; Yousif, M.; Mostafa, S.A.; Kadry, S.; Abdulkareem, K.H.; Garcia-Zapirain, B. Novel Crow Swarm Optimization Algorithm and Selection Approach for Optimal Deep Learning COVID-19 Diagnostic Model. Comput. Intell. Neurosci. 2022, 2022, 1307944. [Google Scholar] [CrossRef]
  2. Saeed, M.; Ahsan, M.; Saeed, M.H.; Rahman, A.U.; Mehmood, A.; Mohammed, M.A.; Jaber, M.M.; Damaševičius, R. An Optimized Decision Support Model for COVID-19 Diagnostics Based on Complex Fuzzy Hypersoft Mapping. Mathematics 2022, 10, 2472. [Google Scholar] [CrossRef]
  3. Allioui, H.; Mohammed, M.A.; Benameur, N.; Al-Khateeb, B.; Abdulkareem, K.H.; Garcia-Zapirain, B.; Damaševičius, R.; Maskeliūnas, R. A Multi-Agent Deep Reinforcement Learning Approach for Enhancement of COVID-19 CT Image Segmentation. J. Pers. Med. 2022, 12, 309. [Google Scholar] [CrossRef] [PubMed]
  4. Albahli, S.; Algsham, A.; Aeraj, S.; Alsaeed, M.; Alrashed, M.; Rauf, H.T.; Arif, M.; Mohammed, M.A. COVID-19 Public Sentiment Insights: A Text Mining Approach to the Gulf Countries. Comput. Mater. Contin. 2021, 67, 1613–1627. [Google Scholar] [CrossRef]
  5. Harun, M.A.A.; Hossain, M.M.; Bari, M.A.; Rubel, N.A.S.; Karim, M.E.; Siddiquee, N.; Hossain, M.D.; Sultana, F.; Taous, A.; Islam, A.K.M.M.; et al. Pulse Oximetry Is Essential in Home Management of Elderly COVID-19 Patients. Bangladesh J. Otorhinolaryngol. 2020, 26, 55–67. [Google Scholar] [CrossRef]
  6. Chavez, S.; Long, B.; Koyfman, A.; Liang, S.Y. Coronavirus Disease (COVID-19): A Primer for Emergency Physicians. Am. J. Emerg. Med. 2021, 44, 220–229. [Google Scholar] [CrossRef]
  7. Guo, H.; Zhou, Y.; Liu, X.; Tan, J. The Impact of the COVID-19 Epidemic on the Utilization of Emergency Dental Services. J. Dent. Sci. 2020, 15, 564–567. [Google Scholar] [CrossRef]
  8. Dinar, A.M.; Raheem, E.A.; Abdulkareem, K.H.; Mohammed, M.A.; Oleiwie, M.G.; Zayr, F.H.; Al-Boridi, O.; Al-Mhiqani, M.N.; Al-Andoli, M.N. Towards Automated Multiclass Severity Prediction Approach for COVID-19 Infections Based on Combinations of Clinical Data. Mob. Inf. Syst. 2022, 2022, 7675925. [Google Scholar] [CrossRef]
  9. Chen, N.; Zhou, M.; Dong, X.; Qu, J.; Gong, F.; Han, Y.; Qiu, Y.; Wang, J.; Liu, Y.; Wei, Y.; et al. Epidemiological and Clinical Characteristics of 99 Cases of 2019 Novel Coronavirus Pneumonia in Wuhan, China: A Descriptive Study. Lancet 2020, 395, 507–513. [Google Scholar] [CrossRef]
  10. Dong, E.; Du, H.; Gardner, L. An Interactive Web-Based Dashboard to Track COVID-19 in Real Time. Lancet Infect. Dis. 2020, 20, 533–534. [Google Scholar] [CrossRef]
  11. Bai, Y.; Yao, L.; Wei, T.; Tian, F.; Jin, D.-Y.; Chen, L.; Wang, M. Presumed Asymptomatic Carrier Transmission of COVID-19. JAMA 2020, 323, 1406–1407. [Google Scholar] [CrossRef]
  12. Huang, C.; Wang, Y.; Li, X.; Ren, L.; Zhao, J.; Hu, Y.; Zhang, L.; Fan, G.; Xu, J.; Gu, X.; et al. Clinical Features of Patients Infected with 2019 Novel Coronavirus in Wuhan, China. Lancet 2020, 395, 497–506. [Google Scholar] [CrossRef]
  13. Montgomery, A.A.; Fahey, T. How Do Patients’ Treatment Preferences Compare with Those of Clinicians? Qual. Heal. Care 2001, 10, i39–i43. [Google Scholar] [CrossRef] [PubMed]
  14. Lee, C.N.; Hultman, C.S.; Sepucha, K. Do Patients and Providers Agree About the Most Important Facts and Goals for Breast Reconstruction Decisions? Ann. Plast. Surg. 2010, 64, 563–566. [Google Scholar] [CrossRef] [PubMed]
  15. Petrilli, C.M.; Jones, S.A.; Yang, J.; Rajagopalan, H.; O’Donnell, L.; Chernyak, Y.; Tobin, K.A.; Cerfolio, R.J.; Francois, F.; Horwitz, L.I. Factors Associated with Hospital Admission and Critical Illness among 5279 People with Coronavirus Disease 2019 in New York City: Prospective Cohort Study. BMJ 2020, 369, m1966. [Google Scholar] [CrossRef]
  16. Jin, X.; Pang, B.; Zhang, J.; Liu, Q.; Yang, Z.; Feng, J.; Liu, X.; Zhang, L.; Wang, B.; Huang, Y.; et al. Core Outcome Set for Clinical Trials on Coronavirus Disease 2019 (COS-COVID). Engineering 2020, 6, 1147–1152. [Google Scholar] [CrossRef]
  17. Quiroz, J.C.; Feng, Y.-Z.; Cheng, Z.-Y.; Rezazadegan, D.; Chen, P.-K.; Lin, Q.-T.; Qian, L.; Liu, X.-F.; Berkovsky, S.; Coiera, E.; et al. Development and Validation of a Machine Learning Approach for Automated Severity Assessment of COVID-19 Based on Clinical and Imaging Data: Retrospective Study. JMIR Med. Inf. 2021, 9, e24572. [Google Scholar] [CrossRef]
  18. Albahri, A.S.; Al-Obaidi, J.R.; Zaidan, A.A.; Albahri, O.S.; Hamid, R.A.; Zaidan, B.B.; Alamoodi, A.H.; Hashim, M. Multi-Biological Laboratory Examination Framework for the Prioritization of Patients with COVID-19 Based on Integrated AHP and Group VIKOR Methods. Int. J. Inf. Technol. Decis. Mak. 2020, 19, 1247–1269. [Google Scholar] [CrossRef]
  19. Mohammed, R.T.; Zaidan, A.A.; Yaakob, R.; Sharef, N.M.; Abdullah, R.H.; Zaidan, B.B.; Albahri, O.S.; Abdulkareem, K.H. Determining Importance of Many-Objective Optimisation Competitive Algorithms Evaluation Criteria Based on a Novel Fuzzy-Weighted Zero-Inconsistency Method. Int. J. Inf. Technol. Decis. Mak. 2021, 21, 195–241. [Google Scholar] [CrossRef]
  20. Albahri, A.S.; Hamid, R.A.; Albahri, O.S.; Zaidan, A.A. Detection-Based Prioritisation: Framework of Multi-Laboratory Characteristics for Asymptomatic COVID-19 Carriers Based on Integrated Entropy–TOPSIS Methods. Artif. Intell. Med. 2021, 111, 101983. [Google Scholar] [CrossRef]
  21. Al-Mhiqani, M.N.; Ahmad, R.; Abidin, Z.Z.; Abdulkareem, K.H.; Mohammed, M.A.; Gupta, D.; Shankar, K. A New Intelligent Multilayer Framework for Insider Threat Detection. Comput. Electr. Eng. 2022, 97, 1075977. [Google Scholar] [CrossRef]
  22. Opricovic, S.; Tzeng, G.-H. Compromise Solution by MCDM Methods: A Comparative Analysis of VIKOR and TOPSIS. Eur. J. Oper. Res. 2004, 156, 445–455. [Google Scholar] [CrossRef]
  23. Opricovic, S.; Tzeng, G.-H. Extended VIKOR Method in Comparison with Outranking Methods. Eur. J. Oper. Res. 2007, 178, 514–529. [Google Scholar] [CrossRef]
  24. Mohammed, M.A.; Abdulkareem, K.H.; Al-Waisy, A.S.; Mostafa, S.A.; Al-Fahdawi, S.; Dinar, A.M.; Alhakami, W.; BAZ, A.; Al-Mhiqani, M.N.; Alhakami, H.; et al. Benchmarking Methodology for Selection of Optimal COVID-19 Diagnostic Model Based on Entropy and TOPSIS Methods. IEEE Access 2020, 8, 99115–99131. [Google Scholar] [CrossRef]
  25. Žižović, M.; Pamucar, D.; Miljković, B.; Karan, A. Multiple-Criteria Evaluation Model for Medical Professionals Assigned to Temporary SARS-CoV-2 Hospitals. Decis. Mak. Appl. Manag. Eng. 2021, 4, 153–173. [Google Scholar] [CrossRef]
  26. May, R.M. Uses and Abuses of Mathematics in Biology. Science 2004, 303, 790–793. [Google Scholar] [CrossRef]
  27. Ghosh, I.; Datta Chaudhuri, T. FEB-Stacking and FEB-DNN Models for Stock Trend Prediction: A Performance Analysis for Pre and Post Covid-19 Periods. Decis. Mak. Appl. Manag. Eng. 2021, 4, 51–84. [Google Scholar] [CrossRef]
  28. Kayapinar Kaya, S. Evaluation of the Effect of COVID-19 on Countries’ Sustainable Development Level: A Comparative MCDM Framework. Oper. Res. Eng. Sci. Theory Appl. 2020, 3, 101–122. [Google Scholar] [CrossRef]
  29. Emanuel, E.J.; Persad, G.; Upshur, R.; Thome, B.; Parker, M.; Glickman, A.; Zhang, C.; Boyle, C.; Smith, M.; Phillips, J.P. Fair Allocation of Scarce Medical Resources in the Time of COVID-19. N. Engl. J. Med. 2020, 382, 2049–2055. [Google Scholar] [CrossRef]
  30. Özkan, B.; Özceylan, E.; Kabak, M.; Dikmen, A.U. Evaluation of Criteria and COVID-19 Patients for Intensive Care Unit Admission in the Era of Pandemic: A Multi-Criteria Decision Making Approach. Comput. Methods Programs Biomed. 2021, 209, 106348. [Google Scholar] [CrossRef]
  31. Parsons, L.M.; Ákos, S.; Cristina, G.; Evan, L.; Paramasivan, C.N.; Alash’le, A.; Steven, S.; Giorgio, R.; John, N. Laboratory Diagnosis of Tuberculosis in Resource-Poor Countries: Challenges and Opportunities. Clin. Microbiol. Rev. 2011, 24, 314–350. [Google Scholar] [CrossRef] [PubMed]
  32. Chen, H.; Guo, J.; Wang, C.; Luo, F.; Yu, X.; Zhang, W.; Li, J.; Zhao, D.; Xu, D.; Gong, Q.; et al. Clinical Characteristics and Intrauterine Vertical Transmission Potential of COVID-19 Infection in Nine Pregnant Women: A Retrospective Review of Medical Records. Lancet 2020, 395, 809–815. [Google Scholar] [CrossRef]
  33. Abdel-Basst, M.; Mohamed, R.; Elhoseny, M. A Model for the Effective COVID-19 Identification in Uncertainty Environment Using Primary Symptoms and CT Scans. Health Inform. J. 2020, 26, 3088–3105. [Google Scholar] [CrossRef]
  34. Majumder, P.; Biswas, P.; Majumder, S. Application of New Topsis Approach to Identify the Most Significant Risk Factor and Continuous Monitoring of Death of COVID-19. Electron. J. Gen. Med. 2020, 17, em234. [Google Scholar] [CrossRef]
  35. Singh, R.; Avikal, S. COVID-19: A Decision-Making Approach for Prioritization of Preventive Activities. Int. J. Healthc. Manag. 2020, 13, 257–262. [Google Scholar] [CrossRef]
  36. Wang, X.; Pan, Y.; Zhang, D.; Chen, L.; Jia, L.; Li, X.; Yang, P.; Wang, Q.; Macintyre, C.R. Basic Epidemiological Parameter Values from Data of Real-World in Mega-Cities: The Characteristics of COVID-19 in Beijing, China. BMC Infect. Dis. 2020, 20, 526. [Google Scholar] [CrossRef] [PubMed]
  37. Artigaud, S.; Gauthier, O.; Pichereau, V. Identifying Differentially Expressed Proteins in Two-Dimensional Electrophoresis Experiments: Inputs from Transcriptomics Statistical Tools. Bioinformatics 2013, 29, 2729–2734. [Google Scholar] [CrossRef]
  38. Hodges, G.; Pallisgaard, J.; Schjerning Olsen, A.-M.; McGettigan, P.; Andersen, M.; Krogager, M.; Kragholm, K.; Køber, L.; Gislason, G.H.; Torp-Pedersen, C.; et al. Association between Biomarkers and COVID-19 Severity and Mortality: A Nationwide Danish Cohort Study. BMJ Open 2020, 10, e041295. [Google Scholar] [CrossRef] [PubMed]
  39. Khan, M.; Shah, N.; Mushtaq, H.; Jehanzeb, V. Profiling Laboratory Biomarkers Associated with COVID-19 Disease Progression: A Single-Center Experience. Int. J. Microbiol. 2021, 2021, 6643333. [Google Scholar] [CrossRef]
  40. Liang, W.; Liang, H.; Ou, L.; Chen, B.; Chen, A.; Li, C.; Li, Y.; Guan, W.; Sang, L.; Lu, J.; et al. Development and Validation of a Clinical Risk Score to Predict the Occurrence of Critical Illness in Hospitalized Patients With COVID-19. JAMA Intern. Med. 2020, 180, 1081–1089. [Google Scholar] [CrossRef]
  41. Yan, L.; Zhang, H.-T.; Goncalves, J.; Xiao, Y.; Wang, M.; Guo, Y.; Sun, C.; Tang, X.; Jing, L.; Zhang, M.; et al. An Interpretable Mortality Prediction Model for COVID-19 Patients. Nat. Mach. Intell. 2020, 2, 283–288. [Google Scholar] [CrossRef]
  42. Levy, T.J.; Richardson, S.; Coppa, K.; Barnaby, D.P.; McGinn, T.; Becker, L.B.; Davidson, K.W.; Cohen, S.L.; Hirsch, J.S.; Zanos, T.P.; et al. Development and Validation of a Survival Calculator for Hospitalized Patients with COVID-19. medRxiv 2020. preprint. [Google Scholar] [CrossRef]
  43. Mejía-Vilet, J.M.; Córdova-Sánchez, B.M.; Fernández-Camargo, D.A.; Méndez-Pérez, R.A.; Morales-Buenrostro, L.E.; Hernández-Gilsoul, T. A Risk Score to Predict Admission to the Intensive Care Unit in Patients with COVID-19: The ABC-GOALS Score. Salud Publica Mex. 2020, 63, 1–11. [Google Scholar] [CrossRef] [PubMed]
  44. Xie, J.; Covassin, N.; Fan, Z.; Singh, P.; Gao, W.; Li, G.; Kara, T.; Somers, V.K. Association Between Hypoxemia and Mortality in Patients With COVID-19. Mayo Clin. Proc. 2020, 95, 1138–1147. [Google Scholar] [CrossRef] [PubMed]
  45. Zhou, F.; Yu, T.; Du, R.; Fan, G.; Liu, Y.; Liu, Z.; Xiang, J.; Wang, Y.; Song, B.; Gu, X.; et al. Clinical Course and Risk Factors for Mortality of Adult Inpatients with COVID-19 in Wuhan, China: A Retrospective Cohort Study. Lancet 2020, 395, 1054–1062. [Google Scholar] [CrossRef]
  46. Lippi, G.; Plebani, M. Laboratory Abnormalities in Patients with COVID-2019 Infection. Clin. Chem. Lab. Med. (CCLM) 2020, 58, 1131–1134. [Google Scholar] [CrossRef] [PubMed]
  47. Lassau, N.; Ammari, S.; Chouzenoux, E.; Gortais, H.; Herent, P.; Devilder, M.; Soliman, S.; Meyrignac, O.; Talabard, M.-P.; Lamarque, J.-P.; et al. Integrating Deep Learning CT-Scan Model, Biological and Clinical Variables to Predict Severity of COVID-19 Patients. Nat. Commun. 2021, 12, 634. [Google Scholar] [CrossRef]
  48. Hameed Abdulkareem, K.; Awad Mutlag, A.; Musa Dinar, A.; Frnda, J.; Abed Mohammed, M.; Hasan Zayr, F.; Lakhan, A.; Kadry, S.; Ali Khattak, H.; Nedoma, J. Smart Healthcare System for Severity Prediction and Critical Tasks Management of COVID-19 Patients in IoT-Fog Computing Environments. Comput. Intell. Neurosci. 2022, 2022, 5012962. [Google Scholar] [CrossRef]
  49. Diakoulaki, D.; Mavrotas, G.; Papayannakis, L. Determining Objective Weights in Multiple Criteria Problems: The Critic Method. Comput. Oper. Res. 1995, 22, 763–770. [Google Scholar] [CrossRef]
  50. Jahan, A.; Mustapha, F.; Sapuan, S.M.; Ismail, M.Y.; Bahraminasab, M. A Framework for Weighting of Criteria in Ranking Stage of Material Selection Process. Int. J. Adv. Manuf. Technol. 2012, 58, 411–420. [Google Scholar] [CrossRef]
  51. Saaty, T.L.; Ergu, D. When Is a Decision-Making Method Trustworthy? Criteria for Evaluating Multi-Criteria Decision-Making Methods. Int. J. Inf. Technol. Decis. Mak. 2015, 14, 1171–1187. [Google Scholar] [CrossRef]
  52. Stević, Ž.; Pamučar, D.; Puška, A.; Chatterjee, P. Sustainable Supplier Selection in Healthcare Industries Using a New MCDM Method: Measurement of Alternatives and Ranking According to COmpromise Solution (MARCOS). Comput. Ind. Eng. 2020, 140, 106231. [Google Scholar] [CrossRef]
  53. Mukhametzyanov, I.; Pamucar, D. A Sensitivity Analysis in MCDM Problems: A Statistical Approach. Decis. Mak. Appl. Manag. Eng. 2018, 1, 51–80. [Google Scholar] [CrossRef]
  54. Stewart, T.J.; French, S.; Rios, J. Integrating Multicriteria Decision Analysis and Scenario Planning—Review and Extension. Omega 2013, 41, 679–688. [Google Scholar] [CrossRef]
  55. Pamucar, D.; Yazdani, M.; Montero-Simo, M.J.; Araque-Padilla, R.A.; Mohammed, A. Multi-Criteria Decision Analysis towards Robust Service Quality Measurement. Expert Syst. Appl. 2021, 170, 114508. [Google Scholar] [CrossRef]
  56. Zhao, H.; Li, N. Optimal Siting of Charging Stations for Electric Vehicles Based on Fuzzy Delphi and Hybrid Multi-Criteria Decision Making Approaches from an Extended Sustainability Perspective. Energies 2016, 9, 270. [Google Scholar] [CrossRef] [Green Version]
Figure 1. Multidimensional Examination Framework for the prioritization of COVID-19 severe patients.
Figure 1. Multidimensional Examination Framework for the prioritization of COVID-19 severe patients.
Bioengineering 09 00457 g001
Figure 2. Main and sub-criteria for examination of COVID-19 patients.
Figure 2. Main and sub-criteria for examination of COVID-19 patients.
Bioengineering 09 00457 g002
Figure 3. COVID-19 criteria weights.
Figure 3. COVID-19 criteria weights.
Bioengineering 09 00457 g003
Figure 4. Sensitivity analysis of V value for 10 patients.
Figure 4. Sensitivity analysis of V value for 10 patients.
Bioengineering 09 00457 g004
Figure 5. Sensitivity analysis on changing criteria weights by adding (10%, 20%, and 30%).
Figure 5. Sensitivity analysis on changing criteria weights by adding (10%, 20%, and 30%).
Bioengineering 09 00457 g005
Table 1. Related works on COVID-19 identification and prioritization.
Table 1. Related works on COVID-19 identification and prioritization.
Ref.Aim of the StudyCase StudyCriteria UsedMultidimensional
[33]Differentiate between
COVID-19 and other four viral chest diseases under uncertain environment using the viruses primary
symptoms and CT scans
Based on other cases studies Primary symptoms and CT scansNo
[18]COVID-19 patient
prioritization dependent on
their health conditions
Real data of 6 patients from the literature Laboratory characteristicsNo
of asymptomatic carriers
Real data from the literature were extended to 56, using simulated dataLaboratory characteristicsNo
[24]Benchmarking Methodology for Selection of
Optimal COVID-19 Diagnostic Model
Real data (publicly available)CT scans dataNo
[34]Identify and select the significant risk factor and continuous monitoring of death due to COVID-19 Real data (publicly available)Confirmed and death cases number of COVID-19No
[35]Identify some of the suggested activities not to be performed during the pandemic period Real data (publicly available)WHO GuidelinesNo
ProposedCOVID-19 patient
prioritization based on multidimensional criteria
Real data of 78 patients (self-collected or own dataset)(General factors and comorbidities),
(multi-clinical characteristics with SpO2 sensor) and (multi-biological characteristics)
Table 2. Proposed Multidimensional Evaluation Matrix (MEM) for COVID-19 prioritization.
Table 2. Proposed Multidimensional Evaluation Matrix (MEM) for COVID-19 prioritization.
DemographicLaboratory FindingsVital SignsSymptomsChronic Conditions
PatientsAgeLymphocyte CountC-reactive Protein (mg/L)Urea (mmol/L)Creatinine (µmol/L)Oximetry Saturation (%)Body Temperature (°C)Pleuritic Chest PainNasal CongestionCoughLost Sense of Smell (1/0)Lost Sense of Taste (1/0)Heart DiseaseDiabetes DiseaseCancer
P1AV (P1/Ts)LCV (P1/Ts)CrPV (P1/Ts)UrV (P1/Ts)CrV (P1/Ts)OSV (P1/Ts)BTV (P1/Ts)PCPV (P1/Ts)NCV (P1/Ts)CoV (P1/Ts)LSV (P1/Ts)LTV (P1/Ts)HDV (P1/Ts)DDV (P1/Ts)CaV (P1/Ts)
P2AV (P2/Ts)LCV (P2/Ts)CrPV (P2/Ts)UrV (P2/Ts)CrV (P2/Ts)OSV (P2/Ts)BTV (P2/Ts)PCPV (P2/Ts)NCV (P2/Ts)CoV (P2/Ts)LSV (P2/Ts)LTV (P2/Ts)HDV (P2/Ts)DDV (P2/Ts)CaV (P2/Ts)
P3AV (P3/Ts)LCV (P3/Ts)CrPV (P3/Ts)UrV (P3/Ts)CrV (P3/Ts)OSV (P3/Ts)BTV (P3/Ts)PCPV (P3/Ts)NCV (P3/Ts)CoV (P3/Ts)LSV (P3/Ts)LTV (P3/Ts)HDV (P3/Ts)DDV (P3/Ts)CaV (P3/Ts)
PnAV (Pn/Ts)LCV (Pn/Ts)CrPV (Pn/Ts)UrV (Pn/Ts)CrV (Pn/Ts)OSV (Pn/Ts)BTV (Pn/Ts)PCPV (Pn/Ts)NCV (Pn/Ts)CoV (Pn/Ts)LSV (Pn/Ts)LTV (Pn/Ts)HDV (Pn/Ts)DDV (Pn/Ts)CaV (Pn/Ts)
Table 3. MEM data.
Table 3. MEM data.
DemographicLaboratory FindingsVital SignsSymptomsChronic Conditions
PatientsAgeLymphocyte CountC-reactive Protein Urea Creatinine Oximetry Saturation Body Temperature Pleuritic Chest PainNasal CongestionCoughLost Sense of Smell Lost Sense of Taste Heart DiseaseDiabetes DiseaseCancer
Table 4. The criteria weights’ values.
Table 4. The criteria weights’ values.
Calculate Weight, WWj
Lymphocyte count0.049
C-reactive protein 0.077
Urea 0.024
Creatinine 0.023
Oximetry saturation 0.034
Body temperature 0.044
Pleuritic chest pain0.099
Nasal congestion0.107
Lost sense of smell 0.066
Lost sense of taste 0.063
Heart Disease0.118
Diabetes Disease0.089
Table 5. Weighted MEM of COVID-19 patients.
Table 5. Weighted MEM of COVID-19 patients.
DemographicLaboratory FindingsVital SignsSymptomsChronic Conditions
PatientsAgeLymphocyte CountC-reactive ProteinUreaCreatinineOximetry SaturationBody TemperaturePleuritic Chest PainNasal CongestionCoughLost Sense of SmellLost Sense of TasteHeart DiseaseDiabetes DiseaseCancer
Table 6. Final ranking of COIVID-19 patients.
Table 6. Final ranking of COIVID-19 patients.
Patient No. SRQRankPatient No.SRQRank
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Abdulkareem, K.H.; Al-Mhiqani, M.N.; Dinar, A.M.; Mohammed, M.A.; Al-Imari, M.J.; Al-Waisy, A.S.; Alghawli, A.S.; Al-Qaness, M.A.A. MEF: Multidimensional Examination Framework for Prioritization of COVID-19 Severe Patients and Promote Precision Medicine Based on Hybrid Multi-Criteria Decision-Making Approaches. Bioengineering 2022, 9, 457.

AMA Style

Abdulkareem KH, Al-Mhiqani MN, Dinar AM, Mohammed MA, Al-Imari MJ, Al-Waisy AS, Alghawli AS, Al-Qaness MAA. MEF: Multidimensional Examination Framework for Prioritization of COVID-19 Severe Patients and Promote Precision Medicine Based on Hybrid Multi-Criteria Decision-Making Approaches. Bioengineering. 2022; 9(9):457.

Chicago/Turabian Style

Abdulkareem, Karrar Hameed, Mohammed Nasser Al-Mhiqani, Ahmed M. Dinar, Mazin Abed Mohammed, Mustafa Jawad Al-Imari, Alaa S. Al-Waisy, Abed Saif Alghawli, and Mohammed A. A. Al-Qaness. 2022. "MEF: Multidimensional Examination Framework for Prioritization of COVID-19 Severe Patients and Promote Precision Medicine Based on Hybrid Multi-Criteria Decision-Making Approaches" Bioengineering 9, no. 9: 457.

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop