{"id":2811,"date":"2020-09-10T17:18:40","date_gmt":"2020-09-10T21:18:40","guid":{"rendered":"https:\/\/blogs.mathworks.com\/headlines\/?p=2811"},"modified":"2020-09-11T13:38:24","modified_gmt":"2020-09-11T17:38:24","slug":"researchers-around-the-globe-tackle-covid-their-results-and-code","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/headlines\/2020\/09\/10\/researchers-around-the-globe-tackle-covid-their-results-and-code\/","title":{"rendered":"Researchers around the globe tackle COVID: their results and code"},"content":{"rendered":"

Never have so many in the research community pivoted to focus on the same problem \u2013 from numerous angles \u2013 so quickly. COVID-19 has united researchers around the world to help better understand the virus, how it affects our bodies and economies, how best to contain it, treat it, and protect others from catching it.<\/span><\/p>\n

With hopes of combatting the pandemic and learning how to help those affected by it, countless researchers are sharing access to their work and code and the MathWorks Research Support Team has assembled\u00a0a collection of research<\/a> papers, blogs, code, and simulation apps related to COVID-19.\u00a0Work shared by commercial firms,\u00a0especially in the area of ventilator design<\/a>, is also hosted on the page. Here is a sampling of the work contained on the page:<\/span><\/p>\n

Does COVID-19 change your brain?<\/span><\/h2>\n

A team of radiologists examined 60 recovered COVID-19 patients and 39 people who never contracted the disease. The results were published in\u00a0The Lancet<\/em>.<\/span><\/p>\n

The researchers found neurological symptoms were present in 55% of COVID-19 patients, suggesting potential long-term neurological effects from the virus. The areas of the brain affected correlate to memory loss and the loss of the sense of smell.<\/span><\/p>\n

\"\"

The regions with relative higher mean values (density) in the COVID-19 group were marked as red, and the regions with relatively lower mean values in the COVID-19 group were marked as blue. Image credit: Lu et.al<\/span><\/p><\/div><\/p>\n

 <\/p>\n

The team used\u00a0SPM-12<\/a>, a free, downloadable, community-developed MATLAB toolbox.<\/span><\/p>\n

How many people need to receive a vaccine to stop the epidemic spread?<\/span><\/h2>\n

Herd immunity is a form of indirect protection from infectious disease that occurs when a sufficient percentage of a population has become immune to an infection, preventing epidemic growth. Herd immunity thresholds (HITs) are used to determine targets for vaccination coverage targets since herd immunity can be reached through vaccination or previous infection.<\/span><\/p>\n

Researchers from the University of Oxford, the Universidade de S\u00e3o Paulo, the University of Edinburgh, the Universidade do Porto, and the University of Strathclyde, Glasgow recently published a paper on medRxiv<\/a> showing the percent of individuals that need to be vaccinated, or previously infected by the virus, varies based on individual susceptibility and connectivity. As we\u2019ve seen throughout the pandemic, areas with both high connectivity and individuals with higher susceptibility have seen greater spread or R0,\u00a0<\/sub>the average number of cases attributed to each infected individual.<\/span><\/p>\n

\"\"

Herd immunity calculations with varying susceptibility (top) and connectivity (bottom) for four countries. The solid line shows the expected HIT for COVID-19, while the dashed lines show the final sizes for unmitigated epidemics. Image credit: Aguas et. al<\/span><\/p><\/div><\/p>\n

 <\/p>\n

The code for this study is available on GitHub<\/a>.<\/span><\/p>\n

In New York City, are the wealthy more likely to receive a COVID test?<\/span><\/h2>\n

In a paper, COVID-19: Testing Inequality in New York City<\/em>, researchers from the National Bureau of Economic Research examined the impact of income on test accessibility, and surprisingly found that the access was quite equal.<\/span><\/p>\n

The study found, \u201cThe ten percent of the city’s population living in the richest zip codes received 11 percent of the Covid-19 tests and 29 percent of the city’s income. The ten percent of the city’s population living in the poorest zip codes received 10 percent of the tests but only 4 percent of the city’s income.\u201d<\/span><\/p>\n

In other words, the testing rates roughly matched the size of the population in both the highest income and lowest income zip codes. There was, however, a marked difference in the rates in positive test results between the two groups. Check out the COVID-19 Research and Development page<\/a> to see how those rates differed.<\/span><\/div>\n

The data used in this investigation along with the MATLAB code to replicate the tables and figures are available here<\/a>.<\/span><\/p>\n

Can PET\/CT imaging diagnose COVID-19 complications?<\/span><\/h2>\n

We\u2019ve repeatedly been told that not all patients have the same COVID symptoms. And when there is a suspected case of COVID, it can be difficult to determine how severe the infection is.<\/span><\/p>\n

The article, Value of PET\/CT and MATLAB in Detection of COVID-19 in an Oncology Patient – Case Report<\/em>, shared a case study of a patient. They were able to quantitatively show the severity of the infection in the patient\u2019s lungs.<\/span><\/p>\n

\"\"

CT image of the 28 years old male patient and the related pixelated CT map using MATLAB, which together with the various density levels of the infected lung demonstrates quantitatively severity of the disease. Image credit: Masoomi et. al.<\/span><\/p><\/div><\/p>\n

 <\/p>\n

Making Custom 3D Printed Masks<\/span><\/h2>\n

Need a well-fitting mask? Dr. Kevin Moerman, a lecturer at the National University of Ireland Galway and a research affiliate at MIT Media Lab\u2019s Biomechanics group, has shared an automated subject-specific mask design system to create a custom 3D printed mask.<\/span><\/p>\n

\n

The algorithm requires a scanned face or some basic landmarks. The landmarks are used to generate a mask curve. I used the red points but other landmarks can be coded [2 of n] pic.twitter.com\/vSwzoD2p3f<\/a><\/p>\n

— Kevin M Moerm\u24d0n (@KMMoerman) April 5, 2020<\/a><\/p><\/blockquote>\n