Fuzzy Cloud Based COVID-19 Diagnosis Assistant for identifying affected cases globally using MCDM.

The COVID-19, Coronavirus Disease 2019, emerged as a hazardous disease that led to many causalities across the world. Early detection of COVID-19 in patients and proper treatment along with awareness can help to contain COVID-19. Proposed Fuzzy Cloud-Based (FCB) COVID-19 Diagnosis Assistant aims to identify the patients as confirmed, suspects, or suspicious of COVID-19. It categorized the patients into four categories as mild, moderate, severe, or critical. As patients register themselves online on the FCB COVID-19 DA in real-time, it creates the database for the same. This database helps to improve diagnostic accuracy as it contains the latest updates from real-world cases data. A team of doctors, experts, consultants are integrated with the FCB COVID-19 DA for better consultation and prevention. The ultimate aim of this proposed theory of FCB COVID-19 DA is to take control of COVID-19 pandemic and de-accelerate its rate of transmission among the society.

Introduction

The COVID-19 formerly known as 2019-nCoV, severe acute respiratory syndrome coronavirus 2, was first found in Wuhan in late 2019 as a case of untraceable pneumonia [1], [2]. The World Health Organization named 2019-nCoV as COVID-19 (Coronavirus disease 2019) on February 11, 2020 [6]. COVID-19 was neither detected nor reported in humans before [3], [4].

The reported cases of COVID-19 reached 74,167,013 as on December 17, 2020 [22]. COVID-19 is a communicable disease which transmitted through close contact with the infected person or even with the infected surface. As per the reports, it is highly contagious and period of its incubation can be 2 weeks or more than that [7]. It infects the lung. Severely infected patients faced problem in respiration which leads to acute respiratory distress syndrome and finally caused death [8], [9]. Hence it is necessary to detect this problem at the early stages so that its expansion can be controlled. Current viral nucleic testing of COVID-19 is having only 30% − 50% of accuracy. Due to this fact, many suspected or suspicious cases got unnoticed and undetectable which are left out from treatment [10]. Initially, “Diagnosis and Treatment Scheme for Pneumonia of COVID-19 (Interim Version 5)” [11], was adopted by National Health Commission of China, which was based on the chest imaging while proposing Version 6 [12]. Despite of adopting the aforesaid proposed Version 6, few cases still remains untouched or misdiagnosed due to different level of diagnosis through different doctors in different regions. Also, the detection of COVID-19 goes untraceable in its initial stage. Bernheim A et al. [13] revealed that detection of nucleic acid test can be done by observing the changes in computed tomography (CT). It is much more important to diagnose the suspects and isolate them to prevent the further dispersion of the infection among the society.

In this paper, we are proposing FCB COVID-19 DA (Fuzzy Cloud Based COVID-19 Diagnosis Assistant) based on Li Bai et al. [23]. The FCB COVID-19 DA will diagnose and helps in early detection of COVID-19 and also provide the treatment based on the condition of the patient. This will be really helpful for the suspects who are not able to visit any hospital or clinic. It also prevents unnecessary visits to hospitals in such hazardous scenario. With the help of the proposed method patients will be able to diagnose themselves, because all the relevant consultation will be easily identified by the proposed system.

Rest of the parts of this paper are as follows (See Fig. 1 ): Section 2 gives overview of cloud computing, Section 3 explains Fuzzy Set Theory, Section 4 contains related work and proposed methodology is given by Section 5 and finally, Section 6 will have conclusions and research directions.

Fig. 1

Paper Organization.

Overview of cloud computing

The processing of heterogeneous data and delivery of computing services like storage, databases, networking, software, analytics, computing power, and intelligence via the remote servers hosted on the internet is called cloud computing [25], [26] . This brings many advantages, including data ubiquity, flexibility of access, and resilience. The cloud computing, upsurges the capabilities of the hardware resources by optimal and shared utilization. The innovative technology enables businesses and enterprises to store applications, develop software solutions, manage business infrastructure and utilize the platform as per their expectation using the tools and development options available. The convenience of the cloud platform is to reduce the dependency, investment, and liabilities of on-premise business infrastructure [27], [28]. Paper also introduces the benefits of cloud computing (as shown by Fig. 2 ) in the market related to scalability and flexibility with existing software solutions which can be overcome using cloud computing platform and gives brief highlights of the advantages using the cloud platforms under COVID-19 pandemic.

Fig. 2

Cloud Computing Benefits.

The worldwide COVID-19 widespread has constrained us to reevaluate how we work, how we learn, offer assistance, lock-in, and socialize. Cloud computing has played a significant part in empowering businesses and governments to rapidly apply arrangements to reply to the emergency and keep up coherence. “Cloud computing represents the catalyst and the enabler of the important technological shift that was already well underway before COVID-19,” said by Doreen Bogdan-Martin, Executive of ITU’s Telecommunication Development Bureau (BDT) during a current cloud for COVID-19 response webinar. “And it’s likely to be key to business resilience in the aftermath of the pandemic.” “Every time we use an App on a smartphone or hit important Web App, chances are high that the back-end that powers this App is a Cloud. The Cloud is becoming the new invisible power thar drives many of the IT systems and Apps that we consume on a daily basis. Every day, we touch multiple Clouds multiple times without even noticing,” said Nasser Kettani, Rapporteur of ITU-D Study Group 1 Question 3/1. We are using cloud computing concept in our proposed method because cloud-based system is based on pay as go price model which means there is no limitation for space while using cloud computing.

Fuzzy set theory

Fuzzy set theory was developed by Lutfi. A. Zadeh in the year 1965. In order to use vague values, we use fuzzy logic. The range lies in fuzzy logic is [0-1]. There are several types of fuzzy set theory like as trapezoidal fuzzy number, bell shaped fuzzy number, Gaussian fuzzy number, Triangular Fuzzy Number (TFN) [29], [30]. In our proposed method we are using triangular fuzzy number to calculate the weight of questionnaires based on DM’s. Our proposed method will provide the optimal treatment to the COVID-19 victim at home through fuzzy assessment calculation by TFN. In fuzzy system, we can only provide the linguistic values like (0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1) to each question based on DM’s expertise. Once we get the linguistic values from DM’s, then we apply TFN to each DM’s decision, and calculate the weight accordingly. MCDM (Multi Criterion Decision Making) method could be a sub-discipline of operations investigate that unequivocally assesses numerous conflicting criteria in choice-making (both in existence and in settings such as commerce, government, and medication). In our proposed method, we are using fuzzy set theory for identifying the vague vales from the questionnaires.

Related work

The detection of COVID-19 disease at early stage is very much important to contain its dispersion. Researches has been done for the same and many more are still going on.

Halgurd S. Maghdid et al. [33] proposed a tool based on artificial intelligence to diagnose Coronavirus COVID-19. An AI-enabled framework has been proposed which uses the embedded sensors of a smartphone to detect COVID-19 disease.

Li Bai et al. [23] proposed “COVID-19 Intelligent Diagnosis and Treatment Assistant Program (nCapp)” which is based on Internet of things (IoT). It collects all the information from suspected patients on a cloud via questionnaires. It analyzes the information and then diagnosis is done automatically and the result is generated as confirmed, suspected or suspicious of COVID-19 disease. Ophir Gozes et al. [34] proposed a model for patients monitoring by AI based automated CT image analysis tool.

There is no method proposed based on fuzzy set theory till date. In our proposed method, we used MCDM method for the detection of COVID-19 and also provide the treatment based on the condition of the patient. This will be really helpful for the suspects who are not able to visit any hospital or clinic. It also prevents unnecessary visits to hospitals in such hazardous scenario.

This section presents a cloud based COVID-19 diagnosis assistant (FCB COVID-19 DA) for the treatment of any patient with recommender systems. The proposed method is presented in the following (see Fig. 3 )

Fig 3

Proposed Methodology.

Register into FCB COVID-19 DA.

Consultation Based on Questionnaire by the DM’s

Identify patient for different regions.

Identify the level of patient.

Identification of treatment category wise.

Calculate fuzzy assessment for each category.

Treatment and protection negotiation and prioritization.

User Registration

Registration will be done by a new patient using FCB COVID-19 DA by giving a modern username and comparing password. The patient then creates a public–private key pair with KMS (Key Management Systems) in FCB COVID-19 DA is overseen by means of an OAuth2 stream utilizing the asset proprietor password credentials grant (see Algorithm 1). Therefore, the first step of our proposed method is to get registered into FCB COVID-19 DA.

Consultation based on Questionnaire by DM’s in FCB Covid-19 DA software

Once the patient is successfully registered into FCB COVID-19 DA, the page will show questions (see Algorithm2), and the patient chooses the reply button, which exchange the information back to the cloud online. Consultation Based on Questionnaires by the DM’s evaluated fuzzy assessment for each category and assigning scores to questionnaires according to DM’s. After getting the cores from each DM’s we have to calculate the weight for each DM’s and find the ranking by sum of weights. Based on the testing by the different DM’s scores, we will provide the optimal treatment to the patients.

Identify patients for different regions

Patients identification is an important activity to elicit patients based on category of patients. For this first the patients have to register into the proposed system i.e., FCB COVID-19 DA.

Identify the level of patient after getting tested by the DM’s

Once the testing done by the different DM’s, we will display the chart category wise to the registered patient, so that he/she can start their treatment with the help of different domain experts at home (See Algorithm 4).

Identification of treatment category wise

In this step, we have to provide the treatment to COVID-19 victim category wise. Relevant guidelines about COVID-19, determination and treatment details, master addresses, research papers, and links are provided. Furthermore, FCB COVID-19 DA can also be used by visualization techniques. After the establishment of the conclusion, the case will be naturally detailed and transmitted to the Cloud. The proposed method also makes specialists and patients communicate in an increased reality way, hence diminishes cross contamination.

Calculate fuzzy assessment for each category through questionnaires

From Table 1 , we identify, after evaluating of weights (with the help of triangular fuzzy number) for Questionnaires the DM’s will provide the optimal solution of COVID-19 through FCB COVID-19 DA. I used the rough data for the same.

Table 1

Fuzzy assessment of decision makers for Questionnaires [31], [32]

Questions	DM1	DM2	DM3	DM4	DM5	WEIGHTS
Q1	0.1	0.9	0.8	1	0.8	0.833
Q2	0.5	0.9	0.6	0.1	0.5	0.550
Q3	0.9	0.9	1	0.8	1	0.833
Q4	0.5	0.9	0.8	0.7	0.4	0.600
Q5	0.9	0.5	0.1	0.4	0.7	0.550
Q6	0.9	0.5	0.7	0.7	1	0.741
Q7	0.9	0.9	1	0.6	0.5	0.741
Q8	1	1	0.5	0.9	1	0.825
Q9	0.9	0.5	0.1	1	0.8	0.641
Q10	0.9	1	0	0.8	0.1	0.558

Treatment and protection negotiation and prioritization

We designed 10 system generated easy-to-use questionnaires for profound mining and intelligent processing for the registered patients. The treatment recommendation and diagnosis are naturally produced and transmitted to specialists (DM’s) or domain specialists for reference. Domain experts and physicians can use FCB COVID-19 DA software to participate according to their needs [35], [36], [37], [38], [39], [40], [41].

Conclusion

In this paper, we proposed a method based on Fuzzy Cloud Based COVID-19 Diagnosis Assistant. The proposed method includes the following steps: register into the proposed method, consultation based on questionnaires by the DM’s, identify patient for different regions, level identification of patient, category wise treatment identification, calculate fuzzy assessment for each category, negotiation and prioritization for treatment and protection. FCB COVID-19 DA helps in distinguishing volunteer specialists for master discussion (help in inspection the outpatient time of attentive specialists in each healing center), online interview (master discussion of volunteers, as it were restricted to the questions of the specialists teaching), and address preparing, counting preparing direct agreement, conclusion and treatment innovation, science instruction, and master gathering. In our proposed method, we used MCDM method for the detection of COVID-19 and also provide the treatment based on the condition of the patient. This will be really helpful for the suspects who are not able to visit any hospital or clinic. It also prevents unnecessary visits to hospitals in such hazardous scenario. With the help of the proposed method patients will be able to diagnose themselves, because all the relevant consultation will be easily identified by the proposed system. Future research directions include the following:

To expand the proposed strategy by utilizing Multi-Criteria Decision-Making strategies like AHP, TOPSIS, MAUT, CBR, DEA, SMART, Goal Programming, ELECTRE, PROMETHEE, SAW etc.; and to design a hybrid system by utilizing an effective strategy for mining recurrence thing sets.

To show the comparative study between different diagnosis center for getting cure from COVID-19.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Algorithm 1
1. User → Patient   : Username, Password
2. Patent   :   : kpub, kpriv ← generatekeyPair()
EncPassword(kpriv)
Salt ← generateSalt()
Hash (Password, Salt)
3. Patient → KMS :   Username, kpub, EncPassword(kpriv)
(Hash (Password, Salt), Salt
4. KMS → Patient :   Oauth key, Refresh Token

Algorithm 2
1. User → Patient     : Start Consultation
2. Q1 ← copd     : Range [0–1]
3. Q2 ← dry cough   : Range [0–1]
4. Q3 ← fever     : Range [0–1]
5. Q4 ← presence of weakness : Range [0–1]
6. Q5 ← difficulty in breathing : Range [0–1]
7. Q6 ← respiratory rate   : Range [0–1]
8. Q7 ← oxygen saturation : Range [0–1]
9. Q8 ← WBC availability : Range [0–1]
10. Q9 ← CRP elevation   : Range [0–1]
11. Q10 ← lung texture abnormality : Range [0–1]
12. Patient ← End
13. DM’s ← Start testing based on expertise

Algorithm 3
1. User → Patient   : Identification
2. Patient ← Registration   : To access FCB COVID-19 DA
3. Registration ← Based on KMS   : Identified patient’s region wise

Algorithm 4
1. User → Patient
2. If patient rangei ε (0–0.2) → Cat1(Mild)
3. Else if patient rangei ε (0.3–0.5) → Cat2(Moderate)
4. Else if patient rangei ε (0.6–0.8) → Cat3(Severe) :
5. Else patient rangei ε (0.9–1) → Cat4(Critical)

Algorithm 5
1. User → Patient : Treatment
2. Patient Cat1 ← Mild : The clinical symptoms are slight, and there is no sign of pneumonia on CT images
3. Patient Cat2 ← Moderate : patients with pneumonia on CT images do not meet the criteria of severe and critical cases
4. Patient Cat3 ← Severe : patients with pneumonia on CT images meet one of the following criteria (respiratory rate>=0.6, resting oxygen saturation<=0.8, oxygenation index<=0.7.
5. Patient Cat4 ← Critical : patients with pneumonia on CT images meet one of the following criteria (respiratory failure requiring mechanical ventilation>=0.9, shock = 1, other organ failures = 1, requiring intensive care unit treatment = 1, and oxygenation index = 1.