Making Use of Machine Learning Algorithms for Multimodal Equipment to Assist in COVID-19's Assessment

Marco Imbalzano*
 
*Correspondence: Marco Imbalzano, Department of Electrical, Computer and Biomedical Engineering, University of Pavia, 27100 Pavia, Italy, Email: marcoimbalzano625@gmail.com

Author info »

Introduction

The brand-new Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) is responsible for COVID-19, a well-known contagious condition. Similar to the UK variant (B.1.1.7), Brazilian variants, South Africa variant (B.1.325), omicron (B.1.1.529), originally discovered in Africa, ihu (B.1.640.2), originally discovered in France, the most recent mongrel variant deltacron (AY.4/ BA.1), which is a combination of the variants delta and omicron, and the newer recombination of variants BA.1 and BA.2 of the The Center for Systems Science and Engineering (CSSE) at John Hopkins University (JHU) has provided sanctioned data as of June 2, 2022, indicating that COVID- 19 has affected more than 531 million people worldwide and killed nearly 6.3 million people, including more than 800 Brazilians, 100 Canadians, and 600 Ecuadorians [1].

Description

It is a fact that widespread vaccination—up to 90 in some countries— has prevented or reduced the effects of this infection, dramatically reducing the number of deaths. However, recent lockdowns in Shanghai and Beijing demonstrate that the COVID-19 pandemic is far from over. Additionally, there are only 60 people worldwide who have been vaccinated (some nations have vaccinated less than 10 percent of their populations), and the daily death toll currently exceeds 3000 [2].

Fever or chills, cough, shortness of breath or difficulty breathing, headache, muscle or body pangs, dizziness or fatigue, sore throat, traffic or watery nose, new loss of smell or taste, nausea, puking, diarrhoea, abdominal pain or anorexia, confusion or bloodied knowledge, and rash are among the COVID- 19 symptoms that vary depending on the SARS-CoV- 2 variant. The Centers for Disease Control and Prevention (CDC) says that people who have the omicron variant, which affects 99.8 percent of people worldwide, can present with symptoms similar to those of earlier variants [3].

Asymptomatic or characteristic, these infected individuals can range from mild to severe to critical. Age, smoking, and comorbidities (such as diabetes, hypertension, cardiovascular complaint, rotundity, habitual lung complaint, and order complaint) are risk factors that increase the likelihood of developing the severe and critical interpretation of the complaint. The most effective method for detecting SARS/CoV 2 infection is reverse transcriptase polymerase chain reaction (RT-PCR); Nonetheless, its high cost restricts access in countries like Ecuador and Brazil (where this test costs between $45 and $65 and more than $100 in Canada and other nations). Additionally, if there is a high demand, test results may be delayed by a few days. RT-PCR is only more reliable when the sample is obtained up to three days after the infection.

Specific measures, such as essential individual styles for asymptomatic and characteristic individual discovery using Artificial Intelligence (AI), are still required to reduce the spread of the epidemic due to the high transmission rate of the omicron variant—which is significantly more advanced than the earlier bones. It is important to note that between 40 and 45 percent of people who have COVID-19 are asymptomatic. This is a big problem for helping the virus spread because people who are similar to them keep passing it on without realizing it. Cough, heart rate variability, blood pressure, body temperature, and oxygen achromatism position are just a few examples of biomedical signals and labels that have been used by researchers around the world to develop low-cost wearable biases and stoner-friendly mobile applications to describe COVID-19 symptoms [4].

Even though these biomedical data aren't enough to prove COVID-19 infection, they could be used as a webbing tool for telemedicine and remote monitoring. For instance, despite the fact that the sound of a forced cough is sufficient to indicate COVID-19 infection, another study estimated that only 59 percent of COVID-19 infected individuals have a dry cough. On the other hand, in asymptomatic individuals, heart rate variability is yet another distinct indicator of potential contagion infection. Fever is another symptom that affects 99 of characteristic individuals with COVID-19, but it does not affect asymptomatic bones. Body temperature is measured to check for fever in individuals. The drop in oxygen achromatism position in the blood (shortened to SpO2 — supplemental capillary oxygen achromatism) is another sign of COVID-19, and if it falls below 95, this could lead to serious health issues; Consequently, it must be covered on a regular basis. Still, other respiratory illnesses like the flu and cold also lower the SpO2 position between 90 and 95 without causing major health problems.

The evaluation of respiratory conditions by means of artificial intelligence (AI) based on biomedical data is relatively new. For example, a methodical review of a workshop that used a variety of AI techniques, such as Logistic Retrogression (LR), Deep Learning (DL), Least Absolute Loss and Selection Operation (Lariat), Random Forest, Bracket and Retrogression Trees (wain), Support Vector Machine (SVM), fuzzy sense, and k-Nearest Neighbor (KNN), among others, to address the opinion of pneumonia using a number of biomedical signals (such as the Because there is a lot of overlap between the symptoms of COVID-19 and other respiratory conditions, the study found that AI could help reduce the number of patients who are misdiagnosed with the virus. In a public crowdsourced Coswara dataset consisting of coughing, breathing, and sustained vowel phonation, as well as one to twenty sounds recorded on a smartphone, samples of individual sounds were used to infer COVID-19 infection. In another study, two subsets of the Cambridge COVID- 19 Sound database were used to infer COVID-19 through coughing sounds and speech using data from the INTERSPEECH 2021 Computational Paralinguistics (ComPaRe) challenge. The COVID- 19 Cough Sub-Challenge (CCS) is the first subset, with cough sounds from 725 audio recordings. The COVID- 19 Speech Sub-Challenge (CSS) is the second subset, with only speech sounds from 893 audio recordings.

Conclusion

In another study, 477 hand-crafted features, including zero-crossing, spectral centroid, and Mel frequency Cepstral Portions (MFCCs), were used to correctly classify healthy and COVID-19 sounds from a crowdsourced dataset of respiratory sounds. Using Origianal double Patterns (LBPs) and Haralick's system as the point birth styles, an audio texture analysis was carried out on three distinct signal modalities of COVID-19 sounds—cough, breath, and speech signal. Another study used biomedical data (body temperature, heart rate, and SpO2) from 1085 quarantined healthy and unhealthy individuals collected through a wearable device to infer COVID- 19 infections, in contrast to cough sounds [5].

Acknowledgement

None.

Conflict of Interest

None.

References

  1. Torezani, Renata, Pedro Henrique Dias dos Santos, Greiciane Gaburro Paneto and Teodiano Freire Bastos-Filho. "Molecular dynamics of the COVID-19 pandemic in Espirito Santo (Brazil) and border States." Rev Inst Med Trop Sao Paulo 64 (2022).

    2. Google Scholar, Crossref, Indexed at

  2. Oliveira, Beatriz Araujo, Lea Campos de Oliveira, Ester Cerdeira Sabino and Thelma Suely Okay. "SARS-CoV-2 and the COVID-19 disease: A mini review on diagnostic methods." Rev Inst Med Trop Sao Paulo 62 (2020).

    3. Google Scholar, Crossref, Indexed at

  3. Oran, Daniel P and Eric J. Topol. "Prevalence of asymptomatic SARS-CoV-2 infection: A narrative review." Ann Intern Med 173 (2020): 362-367.

    4. Google Scholar, Crossref, Indexed at

  4. Wang, Xin Huang, Mengyuan Xu, Tengfei He and Zhenhua Zhang. "The establishment of reference sequence for SARS‐CoV‐2 and variation analysis." J Med Virol 92 (2020): 667-674

    5. Google Scholar, Crossref, Indexed at

  5. Sabino, Ester C, Myuki AE Crispim, Nelson A. Fraiji and Rafael HM Pereira, et al. "Resurgence of COVID-19 in Manaus, Brazil, despite high seroprevalence." Lancet 397 (2021): 452-455.

    6. Google Scholar, Crossref, Indexed at

Author Info

1Department of Electrical, Computer and Biomedical Engineering, University of Pavia, 27100 Pavia, Italy
 

Received: 26-Sep-2022, Manuscript No. jbbs-23-87910; Editor assigned: 28-Sep-2022, Pre QC No. P-87910; Reviewed: 12-Oct-2022, QC No. Q-87910; Revised: 18-Oct-2022, Rev Manuscript No. R-87910; Published: 26-Oct-2022, DOI: 10.37421/2155-9538.2022.12.325

Citation: Imbalzano, Marco. “Making Use of Machine Learning Algorithms for Multimodal Equipment to Assist in COVID-19's Assessment.” J Bioengineer & Biomedical Sci 12 (2022): 325.

Copyright: © 2022 Imbalzano M. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.