A novel dictionary learning-based approach for Ultrasound Elastography denoising

Yihua Song; Chen Ge; Ningning Song; Meili Deng; Yihua Song; Chen Ge; Ningning Song; Meili Deng

doi:10.3934/mbe.2022537

Mathematical Biosciences and Engineering

2022, Volume 19, Issue 11: 11533-11543. doi: 10.3934/mbe.2022537

Previous Article Next Article

Research article Special Issues

A novel dictionary learning-based approach for Ultrasound Elastography denoising

1.
School of Artificial Intelligence and Information Technology, Nanjing University of Chinese Medicine, Nanjing 210023, China
2.
Shandong Vocational and Technical University of Engineering, Jinan 250200, China
3.
Nanjing First Hospital, Nanjing 210000, China
4.
China United Network Communications Corporation, Nanjing 210000, China

Academic Editor: Simon James Fong

Received: 08 June 2022 Revised: 25 July 2022 Accepted: 11 August 2022 Published: 11 August 2022

Ultrasound Elastography is a late-model Ultrasound imaging technique mainly used to diagnose tumors and diffusion diseases that can't be detected by traditional Ultrasound imaging. However, artifact noise, speckle noise, low contrast and low signal-to-noise ratio in images make disease diagnosing a challenging task. Medical images denoising, as the first step in the follow-up processing of medical images, has been concerned by many people. With the widespread use of deep learning technique in the research field, dictionary learning method are once again receiving attention. Dictionary learning, as a traditional machine learning method, requires less sample size, has high training efficiency, and can describe images well. In this work, we present a novel strategy based on K-clustering with singular value decomposition (K-SVD) and principal component analysis (PCA) to reduce noise in Ultrasound Elastography images. At this stage of dictionary training, we implement a PCA method to transform the way dictionary atoms are updated in K-SVD. Finally, we reconstructed the image based on the dictionary atoms and sparse coefficients to obtain the denoised image. We applied the presented method on datasets of clinical Ultrasound Elastography images of lung cancer from Nanjing First Hospital, and compared the results of the presented method and the original method. The experimental results of subjective and objective evaluation demonstrated that presented approach reached a satisfactory denoising effect and this research provides a new technical reference for computer aided diagnosis.

Keywords:

Citation: Yihua Song, Chen Ge, Ningning Song, Meili Deng. A novel dictionary learning-based approach for Ultrasound Elastography denoising[J]. Mathematical Biosciences and Engineering, 2022, 19(11): 11533-11543. doi: 10.3934/mbe.2022537

Related Papers:

[1]	Ali Moussaoui, El Hadi Zerga . Transmission dynamics of COVID-19 in Algeria: The impact of physical distancing and face masks. AIMS Public Health, 2020, 7(4): 816-827. doi: 10.3934/publichealth.2020063
[2]	Saina Abolmaali, Samira Shirzaei . A comparative study of SIR Model, Linear Regression, Logistic Function and ARIMA Model for forecasting COVID-19 cases. AIMS Public Health, 2021, 8(4): 598-613. doi: 10.3934/publichealth.2021048
[3]	Musyoka Kinyili, Justin B Munyakazi, Abdulaziz YA Mukhtar . Mathematical modeling and impact analysis of the use of COVID Alert SA app. AIMS Public Health, 2022, 9(1): 106-128. doi: 10.3934/publichealth.2022009
[4]	Pooja Yadav, Shah Jahan, Kottakkaran Sooppy Nisar . Analysis of fractal-fractional Alzheimer's disease mathematical model in sense of Caputo derivative. AIMS Public Health, 2024, 11(2): 399-419. doi: 10.3934/publichealth.2024020
[5]	Ahmed A Mohsen, Hassan Fadhil AL-Husseiny, Xueyong Zhou, Khalid Hattaf . Global stability of COVID-19 model involving the quarantine strategy and media coverage effects. AIMS Public Health, 2020, 7(3): 587-605. doi: 10.3934/publichealth.2020047
[6]	Mehreen Tariq, Margaret Haworth-Brockman, Seyed M Moghadas . Ten years of Pan-InfORM: modelling research for public health in Canada. AIMS Public Health, 2021, 8(2): 265-274. doi: 10.3934/publichealth.2021020
[7]	Kottakkaran Sooppy Nisar, Muhammad Wajahat Anjum, Muhammad Asif Zahoor Raja, Muhammad Shoaib . Recurrent neural network for the dynamics of Zika virus spreading. AIMS Public Health, 2024, 11(2): 432-458. doi: 10.3934/publichealth.2024022
[8]	Shahid Ahmed, Shah Jahan, Kamal Shah, Thabet Abdeljawad . On mathematical modelling of measles disease via collocation approach. AIMS Public Health, 2024, 11(2): 628-653. doi: 10.3934/publichealth.2024032
[9]	Rafat Zreiq, Souad Kamel, Sahbi Boubaker, Asma A Al-Shammary, Fahad D Algahtani, Fares Alshammari . Generalized Richards model for predicting COVID-19 dynamics in Saudi Arabia based on particle swarm optimization Algorithm. AIMS Public Health, 2020, 7(4): 828-843. doi: 10.3934/publichealth.2020064
[10]	Mario Coccia . Sources, diffusion and prediction in COVID-19 pandemic: lessons learned to face next health emergency. AIMS Public Health, 2023, 10(1): 145-168. doi: 10.3934/publichealth.2023012

Abstract

1. Introduction

A mathematical model is a helpful tool to recognize the conduct of an infection when it starts to affect the community and it is useful to analyze under what conditions it can be screened out or to be continued [1]. A virus is known as infectious when any disease is transferred from one person to another via different ways of transmission like droplets generated when an infected person coughs, sneezes, or exhales, or direct contact with another human, water, or any physical product. To analyze this type of transmission we need some authentic mathematical tools in which few of them are difference equations, initial conditions, working parameters, and statistical estimation. In this new era, new mathematical techniques give us more updated and reliable tools to understand many diseases or infections in epidemiology and even give us updated strategies to control disease or infection in different and suitable conditions [2].

From all of the viruses, the COVID-19 is gradually becoming a watershed pandemic in the antiquity of the planet. COVID-19 is an abbreviation of Coronavirus disease which started in 2019. In December of 2019, the first case of COVID-19 was observed in Wuhan, the city of China [3]. The common symptoms of COVID-19 are loss of smell and taste, fever, dry cough, shortening of breath, fatigue, muscle, and joint pain, phlegm production, sore throat, headache, and chills, these symptoms vary from person to person. The most common incubation period ranges from 1 to 12 days. COVID-19 spreads by physical interaction between individuals. Use of masks, sanitizer, and having a distance of 2 m between individuals results in minimizing the spread of the virus up to much extent [4]. These vaccines played a bold role in minimizing the spread of COVID-19. The main focused area of this spread is the working area, schools, offices, markets, and other open circles [5],[6].

Fractional derivative was originated in 1695. If we describe the list of fractional derivatives then it is divided into two types. Caputo, Riemann-Liouville, and Katugampola [7] are fractional derivatives with the singular kernel. Caputo-Fabrizio(exponential) [8] and ABC(Mittag-Leffler) [9] are fractional derivatives without singular kernels. Fractional calculus has very vast application properties in our daily life. It is being used in chemical, biological, physical, finance, pharmaceutical, engineering [10],[11], and many other fields [12]–[14]. FFD is mostly used because it gives a realistic way of representation of our model and hence we have used this same for representing our COVID-19 epidemics [15]–[18]. A time-fractional compartmental model for the COVID-19 pandemic [21] and classical SIR model for COVID-19 in United States is study in [22]. The COVID-19 pandemic (caused by SARS-CoV-2) has introduced significant challenges for accurate prediction of population morbidity and mortality by traditional variable-based methods of estimation. Challenges to modeling include inadequate viral physiology comprehension and fluctuating definitions of positivity between national-to-international data. This paper proposes that accurate forecasting of COVID-19 caseload may be best preformed non-perimetrically, by vector autoregressive (VAR) of verifiable data regionally [23]. Fundamental properties of the new generalized fractional derivatives in the sense of Caputo and RiemannLiouville are rigorously studied and its related work [24]–[26]. COVID-19 Decision-Making System (CDMS) was developed to study disease transmission in [27]. The change in atmospheric pollution from a public lockdown in Greece introduced to curb the spread of the COVID-19 is examined based on ground-based and satellite observations and some related issues in [28]–[30].

2. Basic concepts

The fractional-order derivative of AB in Reimann Liouville-Caputo sense (ABC) [19] is given by

$_{γ}^{A B C} D_{t}^{γ} {f (t)} = \frac{A B (γ)}{m - γ} \int_{γ}^{t} \frac{d^{m}}{d w^{m}} f (w) E_{γ} [- γ \frac{{(t - w)}^{γ}}{m - γ}] d w, m - 1 < γ < m$ (1)

where E_γ is the Mittag-Leffler function and AB(γ) is a normalization function and AB(0) = AB(1) = 1. The Laplace transform of above is given by

$[_{γ}^{A B C} D_{t}^{γ} f (t)] (s) = \frac{A B (γ)}{1 - γ} \frac{s^{γ} L [f (t)] (s) - s^{γ - 1} f (0)}{s^{γ} + \frac{γ}{1 - γ}}$ (2)

with the aid of sumudu transformation, we get

$S T [_{γ}^{A B C} D_{t}^{γ} f (t)] (s) = \frac{B (γ)}{1 - γ} (γ Γ (γ + 1) E_{γ} (- \frac{1}{1 - γ} ν^{γ})) \times [S T (f (t)) - f (0)]$ (3)

The ABC fractional integral of order γ of a function f(t) is given by

$_{γ}^{A B C} I_{t}^{γ} {f (t)} = \frac{1 - (γ)}{B - γ} f (t) + \frac{(γ)}{B (γ) Γ (γ)} \int_{γ}^{t} f (s) (t - s)^{γ - 1} d s$ (4)

3. Materials and methods

In this section, consider the improved SEIR model given in [20] having compartments SEIQRPD, where S represents the number of uninfected individuals, E represents infected individuals at the time t but still in incubation period (without clinical symptoms and low infectivity), I represents the number of infected individuals at the time t (with obvious clinical symptoms), Q represents the number of individuals who have been diagnosed and isolated at the time t, R represents the number of recovered individuals at the time t, P represents the number of susceptible individuals who are not exposed to the external environment at the time t and D represents the number of death cases at time t.

$\begin{matrix} _{0}^{A B C} D_{t}^{γ} S (t) = - β_{1} (t) a_{1} (S (t)) b_{1} (I (t)) - β_{2} a_{2} (S (t)) b_{2} (E (t)) - ρ S (t), \\ _{0}^{A B C} D_{t}^{γ} E (t) = β_{1} (t) a_{1} (S (t)) b_{1} (I (t)) + β_{2} a_{2} (S (t)) b_{2} (E (t)) - ϵ E (t), \\ _{0}^{A B C} D_{t}^{γ} I (t) = ϵ E (t) - δ I (t), \\ _{0}^{A B C} D_{t}^{γ} Q (t) = δ I (t) - (λ (t) + κ (t)) Q (t), \\ _{0}^{A B C} D_{t}^{γ} R (t) = λ (t) Q (t), \\ _{0}^{A B C} D_{t}^{γ} P (t) = ρ S (t) \\ _{0}^{A B C} D_{t}^{γ} D (t) = κ (t) Q (t) \end{matrix}$ (5)

here $β_{1} (t) = σ_{1} e x p (- σ_{2} t), λ (t) = λ_{1} (1 - e x p (λ_{2} t))$ and $κ (t) = κ_{1} e x p (- κ_{2} t)$ $σ_{1}, σ_{2}, λ_{1}, λ_{2}, κ_{1}$ and κ₂ are the parameters which are all positive, here simulation is used by the σ affect of government control. It should be emphasized that the protection rate ρ for susceptible individuals also reflects the intensity of government control [18]. $_{0}^{A B C} D_{t}^{γ}$ , is the ABC sense fractional derivative with $0 < γ \leq 1$ . The initial conditions of the system Eq 5 are:

$\begin{matrix} S_{0} (t) = S (0), E_{0} (t) = E (0), I_{0} (t) = I (0), Q_{0} (t) = Q (0) \\ R_{0} (t) = R (0), P_{0} (t) = P (0), D_{0} (t) = D (0) \end{matrix}$ (6)

applying ST operator on both sides, we get

$\begin{matrix} O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ}) [S T (S (t)) - S (0)] \\ = S T [- β_{1} (t) a_{1} (S (t)) b_{1} (I (t)) - β_{2} a_{2} (S (t)) b_{2} (E (t)) - ρ S (t)], \\ O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ}) [S T (E (t)) - E (0)] \\ = S T [β_{1} (t) a_{1} (S (t)) b_{1} (I (t)) + β_{2} a_{2} (S (t)) b_{2} (E (t)) - ϵ E (t)], \\ O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ}) [S T (I (t)) - I (0)] \\ = S T [ϵ E (t) - δ I (t)], \\ O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ}) [S T (Q (t)) - Q (0)] \\ = S T [δ I (t) - (λ (t) + κ (t)) Q (t)], \\ O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ}) [S T (R (t)) - R (0)] \\ = S T [λ (t) Q (t)], \\ O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ}) [S T (P (t)) - P (0)] \\ = S T [ρ S (t)], \\ O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ}) [S T (D (t)) - D (0)] \\ = S T [κ (t) Q (t)] \end{matrix}$ (7)

where $O_{γ} = \frac{B (γ) γ Γ (γ + 1)}{1 - γ}$ system Eq 7 becomes

$\begin{matrix} S T [S (t)] = S (0) + \frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [- β_{1} (t) a_{1} (S (t)) b_{1} (I (t)) - β_{2} a_{2} (S (t)) b_{2} (E (t)) - ρ S (t)], \\ S T [E (t)] = E (0) + \frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [β_{1} (t) a_{1} (S (t)) b_{1} (I (t)) + β_{2} a_{2} (S (t)) b_{2} (E (t)) - ϵ E (t)], \\ S T [I (t)] = I (0) + \frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [ϵ E (t) - δ I (t)], \\ S T [Q (t)] = Q (0) + \frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [δ I (t) - (λ (t) + κ (t)) Q (t)], \\ S T [R (t)] = R (0) + \frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [λ (t) Q (t)], \\ S T [P (t)] = P (0) + \frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [ρ S (t)], \\ S T [D (t)] = D (0) + \frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [κ (t) Q (t)] \end{matrix}$ (8)

taking inverse Sumudu Transform on both sides, we get

$\begin{matrix} S (t) = S (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [- β_{1} (t) a_{1} (S (t)) b_{1} (I (t)) - β_{2} a_{2} (S (t)) b_{2} (E (t)) - ρ S (t)]}, \\ E (t) = E (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [β_{1} (t) a_{1} (S (t)) b_{1} (I (t)) + β_{2} a_{2} (S (t)) b_{2} (E (t)) - ϵ E (t)]}, \\ I (t) = I (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [ϵ E (t) - δ I (t)]}, \\ Q (t) = Q (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [δ I (t) - (λ (t) + κ (t)) Q (t)]}, \\ R (t) = R (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [λ (t) Q (t)]}, \\ P (t) = P (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [ρ S (t)]}, \\ D (t) = D (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [κ (t) Q (t)]} \end{matrix}$ (9)

Therefore, the following is obtained

$\begin{matrix} S_{(m + 1)} (t) = S_{m} (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [- β_{1} a_{1} (S_{m} (t)) b_{1} (I_{m} (t)) - β_{2} a_{2} (S_{m} (t)) b_{2} (E_{m} (t)) - ρ S_{m} (t)]}, \\ E_{(m + 1)} (t) = E_{m} (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [β_{1} a_{1} (S_{m} (t)) b_{1} (I_{m} (t)) + β_{2} a_{2} (S_{m} (t)) b_{2} (E_{m} (t)) - ϵ E_{m} (t)]}, \\ I_{(m + 1)} (t) = I_{m} (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [ϵ E_{m} (t) - δ I_{m} (t)]}, \\ Q_{(m + 1)} (t) = Q_{m} (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [δ I_{m} (t) - (λ + κ) Q_{m} (t)]}, \\ R_{(m + 1)} (t) = R_{m} (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [λ Q_{m} (t)]}, \\ P_{(m + 1)} (t) = P_{m} (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [ρ S_{m} (t)]}, \\ D_{(m + 1)} (t) = D_{m} (0) + S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [κ Q_{m} (t)]} \end{matrix}$ (10)

Let's consider Eq 10, and then we get

$\begin{array}{l} S (t) = \underset{m \to \infty}{l i m} S_{m} (t); E (t) = \underset{m \to \infty}{l i m} E_{m} (t); I (t) = \underset{m \to \infty}{l i m} I_{m} (t); \\ Q (t) = \underset{m \to \infty}{l i m} Q_{m} (t); R (t) = \underset{m \to \infty}{l i m} R_{m} (t); P (t) = \underset{m \to \infty}{l i m} P_{m} (t)); D (t) = \underset{m \to \infty}{l i m} D_{m} (t) \end{array}$ (11)

Theorem 3.1: Let $(X, | . |)$ be a Banach space and H a self-map of X satisfying

$∥ H_{r} - H_{x} ∥ \leq θ ∥ X - H_{r} ∥ + θ ∥ r - x ∥$ (12)

for all $r, x \in X$ , where $0 \leq θ < 1$ . Assume that H is Pichard H-stable

Let us consider Eq 10, and we obtain

$\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})}$ (13)

the above equation is associated with the fractional Lagrange multiplier.

Proof

Define K be a self-map is given by

$\begin{matrix} K [S_{(m + 1)} (t)] = S_{(m + 1)} (t) = S_{m} (0) + S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [- β_{1} a_{1} (S_{m} (t)) b_{1} (I_{m} (t)) - β_{2} a_{2} (S_{m} (t)) b_{2} (E_{m} (t)) - ρ S_{m} (t)] \\ K [E_{(m + 1)} (t)] = E_{(m + 1)} (t) = E_{m} (0) + S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [β_{1} a_{1} (S_{m} (t)) b_{1} (I_{m} (t)) + β_{2} a_{2} (S_{m} (t)) b_{2} (E_{m} (t)) - ϵ E_{m} (t)] \\ K [I_{(m + 1)} (t)] = I_{(m + 1)} (t) = I_{m} (0) + S T^{- 1} [\frac{1 -}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [ϵ E_{m} (t) - δ I_{m} (t)] \\ K [Q_{(m + 1)} (t)] = Q_{(m + 1)} (t) = Q_{m} (0) + S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [δ I_{m} (t) - (λ + κ) Q_{m} (t)] \\ K [R_{(m + 1)} (t)] = R_{(m + 1)} (t) = R_{m} (0) + S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [λ Q_{m} (t)] \\ K [P_{(m + 1)} (t)] = P_{(m + 1)} (t) = P_{m} (0) + S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [ρ S_{m} (t)] \\ K [D_{(m + 1)} (t)] = D_{(m + 1)} (t) = D_{m} (0) + S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [κ Q_{m} (t)] \end{matrix}$ (14)

Applying the properties of the norm and triangular inequality, we get

$\begin{matrix} ∥ K [S_{m} (t)] - K [S_{n} (t)] ∥ \leq ∥ S_{m} (t) - S_{n} (t) ∥ + ∥ S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [- β_{1} a_{1} S_{m} (t) b_{1} (I_{m} (t)) - β_{2} a_{2} (S_{m} (t)) b_{2} (E_{m} (t)) - ρ S_{m} (t)]} \\ - S T^{- 1} {\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [- β_{1} a_{1} (S_{n} (t)) b_{1} (I_{n} (t)) - β_{2} a_{2} (S_{n} (t)) b_{2} (E_{n} (t)) - ρ S_{n} (t)]} ∥, \\ ∥ K [E_{m} (t)] - K [E_{n} (t)] ∥ \leq ∥ E_{m} (t) - E_{n} (t) ∥ + ∥ S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [β_{1} a_{1} (S_{m} (t)) b_{1} (I_{m} (t)) + β_{2} a_{2} (S_{m} (t)) b_{2} (E_{m} (t)) - ϵ E_{m} (t)]} \\ - S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [β_{1} a_{1} (S_{n} (t)) b_{1} (I_{n} (t)) + β_{2} a_{2} (S_{n} (t)) b_{2} (E_{n} (t)) - ϵ E_{n} (t)]} ∥, \\ ∥ K [I_{m} (t)] - K [I_{n} (t)] ∥ \leq ∥ I_{m} (t) - I_{n} (t) ∥ + ∥ S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [ϵ (E_{m} (t)) - δ (I_{m} (t)))]} - S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [ϵ E_{n} (t) - δ I_{n} (t)]} ∥, \\ ∥ K [Q_{m} (t)] - K [Q_{n} (t)] ∥ \leq ∥ Q_{m} (t) - Q_{n} (t) ∥ + ∥ S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [δ I_{m} (t) - (λ + κ) Q_{m} (t))]} - S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [δ I_{n} (t) - (λ + κ) Q_{n} (t)]} ∥ \\ ∥ K [R_{m} (t)] - K [R_{n} (t)] ∥ \leq ∥ R_{m} (t) - R_{n} (t) ∥ + ∥ S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [λ (Q_{m} (t))]} - S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [λ Q_{n} (t)]} ∥, \\ ∥ K [P_{m} (t)] - K [P_{n} (t)] ∥ \leq ∥ P_{m} (t) - P_{n} (t) ∥ + ∥ S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [ρ S_{m} (t)]} - S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [ρ S_{n} (t)]} ∥, \\ ∥ K [D_{m} (t)] - K [D_{n} (t)] ∥ \leq ∥ D_{m} (t) - D_{n} (t) ∥ + ∥ S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [κ Q_{m} (t)]} - S T^{- 1} [\frac{1}{O_{γ} E_{γ} (- \frac{1}{1 - γ} ω^{γ})} \\ \times S T [κ Q_{n} (t)]} ∥ \end{matrix}$ (15)

K fulfills the conditions associated with theorem 3.1 when

$θ = (0,0,0,0,0,0,0), θ = (\begin{array}{l} ∥ S_{m} (t) - S_{n} (t) ∥ \times ∥ - S_{m} (t) + S_{n} (t) ∥ \\ - ∥ β_{1} a_{1} (S_{m} (t) - S_{n} (t)) b_{1} (I_{m} (t) - I_{n} (t)) ∥ \\ - ∥ β_{2} a_{2} (S_{m} (t) - S_{n} (t)) b_{2} (E_{m} (t) - E_{n} (t)) ∥ \\ - ∥ ρ (S_{m} (t) - S_{n} (t)) ∥ \\ \times ∥ E_{m} (t) - E_{n} (t) ∥ \times ∥ - E_{m} (t) + E_{n} (t) ∥ \\ + ∥ β_{1} a_{1} (S_{m} (t) - S_{n} (t)) b_{2} (E_{m} (t) - E_{n} (t)) ∥ \\ - ∥ ϵ (E_{m} (t) - E_{n} (t)) ∥ \\ \times ∥ I_{m} (t) - I_{n} (t) ∥ \times ∥ - I_{m} (t) + I_{n} (t) ∥ \\ + ∥ ϵ (E_{m} (t) - E_{n} (t)) ∥ - δ ∥ I_{m} (t) - I_{n} (t) ∥ \\ \times ∥ Q_{m} (t) - Q_{n} (t) ∥ \times ∥ - Q_{m} (t) + Q_{n} (t) ∥ \\ + ∥ δ (I_{m} (t) - I_{n} (t)) ∥ - (λ (t) + κ (t)) ∥ Q_{m} (t) - Q_{n} (t) ∥ \\ \times ∥ R_{m} (t) - R_{n} (t) ∥ \times ∥ - R_{m} (t) + R_{n} (t) ∥ \\ + ∥ λ (Q_{m} (t) - Q_{n} (t)) ∥ \\ \times ∥ P_{m} (t) - P_{n} (t) ∥ - ∥ - P_{m} (t) + P_{n} (t) ∥ \\ + ∥ ρ (S_{m} (t) - S_{n} (t)) ∥ \\ \times ∥ D_{m} (t) - D_{n} (t) ∥ - ∥ - D_{m} (t) + D_{n} (t) ∥ \\ + ∥ κ (Q_{m} (t) - Q_{n} (t)) ∥ \end{array}$ (16)

and we add that K is Picard K-stable.

4. Numerical results and discussion

In this section, consider the numerical simulations of the proposed scheme using the ABC technique for the fractional-order COVID-19 model. Figure 1 shows the simulation S(t) represents the number of uninfected individuals. Shows a deep decreasing curve till point (20, 0.25) and then becomes constant and reduced to zero at (100, 0). Figure 2 shows the simulation E(t) of infected individuals but still is in the incubation period (without clinical symptoms and low infectivity). Figure 3 I(t) which represents the number of infected individuals. Here the graph shows a rapid increase (10, 9) and then decrease rapidly with the same rate and then it becomes constant at (100, 0). Figure 4 represents the number of individuals who have been diagnosed and isolated. Figures 5 and 6 shows the simulation of recovered individuals and those not exposed to the external environment respectively. Figure 7 shows the simulation D(t), which represents the death due to increasing or decreasing the infection rate of COVID-19 in society. It can be easily observed from all figures the solution will converge to steady-state and lie in the bounded domain by decreasing the fractional value. Moreover, it has been demonstrated that physical processes are better well described using the derivatives of fractional order which are more accurate and reliable in comparison with the classical-order derivatives. Moreover, it can be seen from all figures that tell that all infected individual comes zero after a few days due to the quarantine effect. The behavior of the dynamics obtained for different instances of fractional-order was shown in the form of numerical results has been reported.

Figure 1. Simulation of S(t) at the time t with parametric value of γ with ABC.

DownLoad: Full-Size Img PowerPoint

Figure 2. Simulation of E(t) at the time t with parametric value of γ with ABC.

DownLoad: Full-Size Img PowerPoint

Figure 3. Simulation of I(t) at the time t with parametric value of γ with ABC.

DownLoad: Full-Size Img PowerPoint

Figure 4. Simulation of Q(t) at the time t with parametric value of γ with ABC.

DownLoad: Full-Size Img PowerPoint

Figure 5. Simulation of R(t) at the time t with parametric value of γ with ABC.

DownLoad: Full-Size Img PowerPoint

Figure 6. Simulation of P(t) at the time t with parametric value of γ with ABC.

DownLoad: Full-Size Img PowerPoint

Figure 7. Simulation of D(t) at the time t with parametric value of γ with ABC.

DownLoad: Full-Size Img PowerPoint

5. Conclusions

We consider the COVID-19 model with fractional operator for this work to check the dynamical behavior of infection of disease in society. In this regard, ABC derivative gave a realistic approach to analyze the effect of disseise during Quarantine which will be helpful for such type of epidemic. The existence and unique solution of the fractional-order model was made with the help of fixed point theory and iterative method. Numerical simulation has been made to check the actual behavior of the COVID-19 effect during quarantine which shows that infected individuals start decreasing after a few days. Such kind of results are very helpful for planning, decision-making, and developing control strategies to overcome the effect of COVID-19 in society.

References

[1]	C. G. Wang, W. Liu, D. Q. Liu, Noise suppression method for ultrasound strain imaging based on coded excitation, Appl. Res. Comput., 30 (2013), 1596-1600. https://doi.org/10.3969/j.jssn.1001-3695.2013.05.083 doi: 10.3969/j.jssn.1001-3695.2013.05.083
[2]	Y. He, S. Cao, H. Zhang, H. Sun, L. Lu, Dynamic PET image denoising with deep learning-based joint filtering, IEEE Access, 9 (2021), 41998-42012. https://doi.org/10.1109/ACCESS.2021.3064926 doi: 10.1109/ACCESS.2021.3064926
[3]	P. Liu, M. D. E. Basha, Y. Li, Y, Xiao, P. C. Sanelli, R. Fang, Deep evolutionary networks with expedited genetic algorithms for medical image denoising, Med. Image Anal., 54 (2019), 306-315. https://doi.org/10.1016/j.media.2019.03.004 doi: 10.1016/j.media.2019.03.004
[4]	C. Broaddus, A. Krull, M. Weigert, U. Schmidt, G. Myers, Removing structured noise with self-supervised blind-spot networks, in IEEE 17th International Symposium on Biomedical Imaging (ISBI), IEEE, (2020), 159-163. https://doi.org/10.1109/ISBI45749.2020.9098336
[5]	M. Green, E. M. Marom, E. Konen, N. Kiryati, A. Mayer, Learning real noise for ultra-low dose lung CT denoising, in Patch-Based Techniques in Medical Imaging Patch-MI 2018 (eds. W. Bai, G. Sanroma, G. Wu, B. Munsell, Y. Zhan, P. Coupé), Springer, Cham, (2018), 3-11. https://doi.org/10.1007/978-3-030-00500-9_1
[6]	L. Tao, C. Zhu, G. Xiang, Y. Li, H. Jia, X. Xie, Llcnn: A convolutional neural network for low-light image enhancement, in 2017 IEEE Visual Communications and Image Processing (VCIP), IEEE, (2018), 1-4. https://doi.org/10.1109/VCIP.2017.8305143
[7]	D. Wu, H. Ren, Q. Li, Self-supervised dynamic CT perfusion image denoising with deep neural networks, IEEE Trans. Radiat. Plasma Med. Sci., 5 (2021), 350-361. https://doi.org/10.48550/arXiv.2005.09766 doi: 10.48550/arXiv.2005.09766
[8]	A. Ouahabi, Signal and Image Multiresolution Analysis, John Wiley & Sons, 2012. https://doi.org/10.1002/9781118568767
[9]	A. Ouahabi, A review of wavelet denoising in medical imaging, in 2013 8th International Workshop on Systems, Signal Processing and their Applications (WoSSPA), IEEE, (2013), 19-26. https://doi.org/10.1109/WoSSPA.2013.6602330
[10]	H. Jomaa, R. Mabrouk, N. Khlifa, F. Morain-Nicolier, Denoising of dynamic pet images using a multi-scale transform and non-local means filter, Biomed. Signal Process. Control, 41 (2017), 69-80. https://doi.org/10.1016/j.bspc.2017.11.002 doi: 10.1016/j.bspc.2017.11.002
[11]	A. Gupta, V. Bhateja, A. Srivastava, A. Gupta, S. C. Satapathy, Speckle noise suppression in Ultrasound images by using an improved non-local mean filter, in Soft Computing and Signal Processing, Springer, Singapore, (2019), 13-19. https://doi.org/10.1007/978-981-13-3393-4_2
[12]	F. Baselice, G. Ferraioli, V. Pascazio, A. Sorriso, Denoising of MR images using Kolmogorov-Smirnov distance in a non local framework, Magn. Reson. Imaging, 57 (2019), 176-193. https://doi.org/10.1016/j.mri.2018.11.022 doi: 10.1016/j.mri.2018.11.022
[13]	M. Xu, X. Xie, An efficient feature-preserving PDE algorithm for image denoising based on a spatial-fractional anisotropic diffusion equation, preprint, arXiv: 2101.01496.
[14]	H. Wang, S. Cao, K. Jiang, H. Wang, Q. Zhang, Seismic data denoising for complex structure using BM3D and local similarity, J. Appl. Geophys., 170 (2019), 103759. https://doi.org/10.1016/j.jappgeo.2019.04.018 doi: 10.1016/j.jappgeo.2019.04.018
[15]	C. Feng, D. Zhao, M. Huang, Image segmentation using CUDA accelerated non-local means denoising and bias correction embedded fuzzy c-means (BCEFCM), Signal Process., 122 (2016), 164-189. https://doi.org/10.1016/j.sigpro.2015.12.007 doi: 10.1016/j.sigpro.2015.12.007
[16]	C. Feng, M. Huang, D. Zhao, Segmentation of longitudinal brain MR images using bias correction embedded fuzzy c-means with non-locally spatio-temporal regularization, J. Visual Commun. Image Represent., 38 (2016), 517-529. https://doi.org/10.1016/j.jvcir.2016.03.027 doi: 10.1016/j.jvcir.2016.03.027
[17]	C. Feng, W. Li, J. Hu, K. Yu, D. Zhao, BCEFCM_S: Bias correction embedded fuzzy c-means with spatial constraint to segment multiple spectral images with intensity inhomogeneities and noises, Signal Process., 168 (2020), 107347. https://doi.org/10.1016/j.sigpro.2019.107347 doi: 10.1016/j.sigpro.2019.107347
[18]	S. Valiollahzadeh, H. Firouzi, M. Babaie-Zadeh, C. Jutten, Image denoising using sparse representations, in International Conference on Independent Component Analysis and Signal Separation, (2009), 557-564. https://doi.org/10.1007/978-3-642-00599-2_70
[19]	H. R. Shahdoosti, S. M. Hazavei, A new compressive sensing based image denoising method using block-matching and sparse representations over learned dictionaries, Multimedia Tools Appl., 78 (2018), 12561-12582. https://doi.org/10.1007/s11042-018-6818-3 doi: 10.1007/s11042-018-6818-3
[20]	F. I. Miertoiu, B. Dumitrescu, Sparse representation and denoising for images affected by generalized Gaussian noise, U.P.B. Sci. Bull., Ser. C, 84 (2022), 75-86.
[21]	L. Nasser, T. Boudier, A novel generic dictionary-based denoising method for improving noisy and densely packed nuclei segmentation in 3D time-lapse fluorescence microscopy images, Sci. Rep., 9 (2019), 1-13. https://doi.org/10.1038/s41598-019-41683-3 doi: 10.1038/s41598-019-41683-3
[22]	H. Haneche, A. Ouahabi, B. Boudraa, New mobile communication system design for Rayleigh environments based on compressed sensing-source coding, IET Commun., 13 (2019), 2375-2385. https://doi.org/10.1049/iet-com.2018.5348 doi: 10.1049/iet-com.2018.5348
[23]	A. E. Mahdaoui, A. Ouahabi, M. S. Moulay, Image denoising using a compressive sensing approach based on regularization constraints, Sensors, 22 (2022), 2199. https://doi.org/10.3390/s22062199 doi: 10.3390/s22062199
[24]	H. Zhu, L. Han, R. Chen, Seismic data denoising method combining principal component analysis and dictionary learning, Global Geol., 39 (2020), 656-662. https://doi.org/10.3969/j.issn.1004-5589.2020.03.015 doi: 10.3969/j.issn.1004-5589.2020.03.015

This article has been cited by:

Changjin Xu, Muhammad Farman, Ali Hasan, Ali Akgül, Mohammed Zakarya, Wedad Albalawi, Choonkil Park, Lyapunov stability and wave analysis of Covid-19 omicron variant of real data with fractional operator, 2022, 61, 11100168, 11787, 10.1016/j.aej.2022.05.025

Reader Comments

Your name:*

Email:*
© 2022 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)