{"id":669339,"date":"2020-07-14T08:00:25","date_gmt":"2020-07-14T15:00:25","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-group&p=669339"},"modified":"2022-12-22T06:53:19","modified_gmt":"2022-12-22T14:53:19","slug":"studies-in-pandemic-preparedness","status":"publish","type":"msr-group","link":"https:\/\/www.microsoft.com\/en-us\/research\/collaboration\/studies-in-pandemic-preparedness\/","title":{"rendered":"Studies in Pandemic Preparedness"},"content":{"rendered":"
\n\t
\n\t\t
\n\t\t\t\"microscopic\t\t<\/div>\n\t\t\n\t\t
\n\t\t\t\n\t\t\t
\n\t\t\t\t\n\t\t\t\t
\n\t\t\t\t\t\n\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t\n\n

Studies in Pandemic Preparedness<\/h1>\n\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t<\/div>\n\t<\/div>\n<\/section>\n\n\n\n\n\n

Microsoft is committed to pushing the boundaries of technology to improve and positively influence all aspects of society. COVID-19 has mobilized communities around the world to work together on a range of critical issues required to understand and respond to the current pandemic. Scientific research is vital to both this near and long-term pandemic preparedness. To help tackle the COVID-19 pandemic, we are increasing the pace and scale of our efforts by deepening our global academic collaborations to help address the current situation and better prepare for future pandemics.<\/p>\n\n\n\n

\u201cBringing together people with expertise, creativity, and passion is the best path forward for addressing difficult challenges with COVID-19. Our efforts today will also bolster our ability to detect and mitigate future pandemics. I am inspired by the many ways the scientific community has aligned to share critical information, new findings, and important insights. These collaborations are essential and will lead the way to vital breakthroughs<\/em>.\u201d<\/p>Eric Horvitz, Chief Scientific Officer, Microsoft<\/cite><\/blockquote>\n\n\n\n

These collaborative projects with academia span topics including infection prevention and control, treatment and diagnostics, mental health, and return to work. This program is a joint initiative supported by between Microsoft’s Chief Scientific Officer<\/a>, Microsoft Research<\/a>, and Microsoft AI for Health<\/a>. <\/p>\n\n\n\n

\n
Explore the research projects<\/a><\/div>\n<\/div>\n\n\n\n
<\/div>\n\n\n\n
\n\t\n\t
\n\t\t
\n\t\t\t

Royal Society Science of COVID conference<\/h4>\n\n\n\n

Highlights from the Studies in Pandemic Preparedness program were presented at the Royal Society Science of COVID Conference (opens in new tab)<\/span><\/a> 30-31 March 2022.<\/p>\n\n\n\n

\n
Watch now<\/a><\/div>\n<\/div>\t\t<\/div>\n\t<\/div>\n\n\t<\/div>\n\n\n\n

Our collaborators<\/h3>\n\n\n\n

We are working closely with the following research collaborators:<\/p>\n\n\n\n

\n
\n
\"The<\/figure>\n<\/div>\n\n\n\n
\n
\"Baylor<\/figure>\n<\/div>\n\n\n\n
\n
\"InBios<\/figure>\n<\/div>\n<\/div>\n\n\n\n
\n
\n
\"Johns<\/figure>\n<\/div>\n\n\n\n
\n
\"Stanford<\/figure>\n<\/div>\n\n\n\n
\n
\"Swansea<\/figure>\n<\/div>\n<\/div>\n\n\n\n
\n
\n
\"Tufts<\/figure>\n<\/div>\n\n\n\n
\n
\"University<\/figure>\n<\/div>\n\n\n\n
\n
\"University<\/figure>\n<\/div>\n<\/div>\n\n\n\n
\n
\n
\"UCL<\/figure>\n<\/div>\n\n\n\n
\n
\"University<\/figure>\n<\/div>\n\n\n\n
\n
\"University<\/figure>\n<\/div>\n<\/div>\n\n\n\n
\n
\n
\"University<\/figure>\n<\/div>\n\n\n\n
\n
\"University<\/figure>\n<\/div>\n\n\n\n
\n
\"University<\/figure>\n<\/div>\n<\/div>\n\n\n\n
\n
Explore the research projects<\/a><\/div>\n<\/div>\n\n\n\n
\n\t\n\t
\n\t\t
\n\t\t\t

Societal Resilience<\/h2>\n\n\n\n

Building a more resilient society through mission-driven research and applied technology<\/strong><\/p>\n\n\n\n

\n
Learn more<\/a><\/div>\n<\/div>\t\t<\/div>\n\t<\/div>\n\n\t\"light<\/div>\n\n\n\n\n\n

These collaborative projects with academia span topics including infection prevention and control, treatment and diagnostics, mental health, and return to work.<\/p>\n\n\n\n

<\/div>\n\n\n\n

Infection Prevention and Control Projects<\/h3>\n\n\n\n\n\n

University of Oxford:<\/strong> Deirdre Hollingsworth
MIT:<\/strong> Chris Rackauckas
Alan Turing Institute:<\/strong> Andrew Duncan (Imperial College London)
Swansea University:<\/strong> Michael Gravenor, Biagio Lucini
University of Cambridge:<\/strong> Ronojoy Adhikari
University of Warwick:<\/strong> Sebastian Vollmer (Alan Turing Institute)
Royal Society Rapid Assistance in Modelling the Pandemic (RAMP):<\/strong> Graeme Ackland (University of Edinburgh), Kostas Kavoussanakis (University of Edinburgh), Mike Cates (University of Cambridge)
Microsoft:<\/strong> Simon Frost, Tim Carroll (Azure HPC), Rolf Harms (Cloud & AI), Grace Huynh<\/a>, Oege de Moor (GitHub), Aqeel Siddiqui (GitHub), Kenji Takeda<\/a>, Miah Wander<\/a>, Jenny Ye (Cloud & AI)<\/p>\n\n\n\n

Mathematical modeling of infectious disease transmission is an important tool in forecasting future trends of pandemics, such as COVID-19, and in evaluating \u2018what if<\/em>\u2019 projections of different interventions. Yet, different models tend to give different results. To draw robust conclusions from modeling, it is important to consider multiple models, which can be facilitated by expanding the modeling community, as models tend to reflect the people that develop them, and by bringing models and model developers together to compare and contrast models. The aims of this project are: to expand and empower the next generation of models and modelers; to act as an incubator for a platform for comparisons of infectious disease models, both in terms of computers and people; and hence enable swift and more robust policy decisions for current and future pandemics.<\/p>\n\n\n\n\n\n

University of Washington:<\/strong> Shwetak Patel, Luis Ceze
Microsoft:<\/strong> Jonathan Lester<\/a>, Bichlien Nguyen<\/a>, Karin Strauss<\/a>, Mike Reddy, Asta Roseway<\/a>, Mike Sinclair<\/a><\/p>\n\n\n\n

The Molecular Information Systems Lab (MISL) and the UbiComp Lab at the University of Washington in collaboration with Microsoft Research are exploring hybrid molecular-electronic systems, with a particular focus on DNA computing as a hybrid approach for detecting COVID-19 in the environment. Detection of the virus has been demonstrated in city wastewater samples. As such, it has been proposed that environmental sampling locations, such as wastewater or air filtration systems, could be informative indicators for continuous monitoring to help cities stay ahead of community-wide COVID-19 spread. Towards that end, the objective of this research is to develop environmental sampling and pathogen detection techniques for air filtration systems on public transit networks, such as buses and subway systems. These techniques include safe sample extraction, qPCR, DNA sequencing, real-time nanopore protocols, and molecular computation. This offers the potential to detect the virus in buses, airliners and buildings, enabling new avenues for viral spread monitoring for a safer and better-informed pandemic response.<\/p>\n\n\n\n\n\n

University of Washington:<\/strong> Shwetak Patel, Luis Ceze
Microsoft:<\/strong> Bichlien Nguyen<\/a>, Jonathan Lester<\/a>, Karin Strauss<\/a>, Mike Reddy, Asta Roseway<\/a>, Mike Sinclair<\/a><\/p>\n\n\n\n

There is currently a shortage of medical-grade face masks worldwide, and that shortage is likely to continue for the duration of the pandemic. As a result, the CDC is providing guidance on how people can create homemade face masks to mitigate the spread of COVID-19. There have been many recent efforts to evaluate the filtration quality of various everyday materials. The objective of this research is to leverage the sensors embedded on commodity smartphones to assess the filtration capability and breathability of homemade face masks. This effort will culminate in a mobile app that can be used to help end-users create better homemade masks, which would significantly limit the risk of virus transmission as social distancing measures begin to relax worldwide.<\/p>\n\n\n\n\n\n

Johns Hopkins University:<\/strong> Anton Dahbura
Royal Society Rapid Assistance in Modelling the Pandemic (RAMP) & University of Edinburgh:<\/strong>\u202fGraeme Ackland, Steven Carlysle-Davies, Kostas Kavoussanakis
University of Exeter:<\/strong> Peter Challenor, Michael Dunne
University of Oxford:<\/strong>\u202fDeirdre Hollingsworth, Emma Davis, Jasmina Panovska Griffiths, Andreia Vasconcelos
University of Washington:<\/strong>\u202fDavid Smith, Sean Wu
The Wilson Centre:<\/strong> Alex Long
Microsoft:<\/strong> Xiaoji Chen, Simon Frost, Eric Horvitz<\/a>, Kenji Takeda<\/a>,\u202fJessica Young<\/p>\n\n\n\n

Computer-based epidemiological models are important tools to help policymakers make important public health decisions. There are, however, challenges in both determining model uncertainty and communicating this uncertainty effectively to policymakers and the public. This project brings together a multi-disciplinary team of academic and Microsoft researchers to develop, demonstrate, and deploy novel methods to better quantify and explain uncertainty in epidemiological models. The goal is to help improve public health decision-making and communication at national, regional, and local levels to effectively respond to COVID-19, future pandemics, and vector-borne epidemics via Microsoft Premonition.<\/p>\n\n\n\n\n\n

<\/div>\n\n\n\n

Treatment and Diagnostics Projects<\/h3>\n\n\n\n\n\n

University College London Hospitals NHS Foundation Trust (UCLH) and University College London (UCL): <\/strong>Joseph Jacob, Sam Janes, Daniel Alexander, Geoff Parker, Jerry Brown, Arjun Nair, Paul Taylor, David Hawkes, Marc Lipman, Joanna Porter, John Hurst, Nick McNally, Bryan Williams
Microsoft:<\/strong> Javier Alvarez-Valle<\/a>, Usman Munir<\/a>, Melissa Bristow<\/a>, Melanie Bernhardt<\/a>, Anton Schwaighofer<\/a>, Jay Nanavati<\/a>, David Carter<\/a>, Ozan Oktay<\/a>, Shruthi Bannur<\/a>, Hannah Murfet<\/a>, Kenji Takeda<\/a>, Aditya Nori<\/a><\/p>\n\n\n\n

Microsoft Research Project InnerEye team<\/a> in Cambridge (UK) is supporting UCLH and UCL to help identify vulnerable patients who are not currently covered by \u201cshielding<\/a>\u201d guidelines. Additionally, UCLH and UCL will help identify patients for whom shielding may not be necessary to avoid economic hardship imposed by unnecessary shielding in future Covid-19 outbreaks. UCLH and UCL will evaluate pre-Covid-19 imaging and clinical data to predict endpoints and speed up development of imaging biomarkers. They will look at cardiac computed tomography (CT) and use quantitative techniques to identify pre-Covid-19 cardiac imaging metrics that link to the likelihood of severe Covid-19 infection and cardiac events. The InnerEye<\/a> team will support UCLH and UCL, who will be developing models for the Prognosis of COVID-19 using CT imaging. This work is supported by the National Institute for Health Research Biomedical Research Centre at UCLH.<\/p>\n\n\n\n\n\n

University Hospitals Birmingham NHS Foundation Trust:<\/strong> Shazad Ashraf, George Gkoutos, Andrew Beggs, Kal Natarajan, Elizabeth Sapey, Alastair Denniston, Sharan Wadhwani
Microsoft:<\/strong> Javier Alvarez-Valle<\/a>, Melissa Bristow<\/a>, Melanie Bernhardt<\/a>, Anton Schwaighofer<\/a>, Jay Nanavati<\/a>, David Carter<\/a>, Ozan Oktay<\/a>, Shruthi Bannur<\/a>, Junaid Bajwa<\/a>, Usman Munir<\/a>, Hannah Murfet<\/a>, Kenji Takeda<\/a>, Aditya Nori<\/a><\/p>\n\n\n\n

Microsoft Research\u2019s Project InnerEye team<\/a> in Cambridge (UK) is working with University Hospitals Birmingham NHS Foundation Trust to develop deep learning models to analyze anonymized chest X-Rays and chest computed tomography (CT) scans to assist clinicians in determining disease severity, aid decision making, and improve our understanding of the disease. The aim is to improve the objective determination of disease severity by classifying and quantifying lesions in the lungs. This could help provide additional information for making a prognosis, assisting in the management of both hospitalized patients and their longer-term health needs. By quantifying disease progression, the model may aid hospitals in making decisions about resource deployment. It will also help to build our knowledge of the disease.<\/p>\n\n\n\n\n\n

Stanford University:<\/strong> Allison Koenecke, Ruoxuan Xiong, Susan Athey
Johns Hopkins University:<\/strong> Mike Powell, Joshua T. Vogelstein
Microsoft:<\/strong> Weiwei Yang, Emre Kiciman<\/a>, Chris White<\/a><\/p>\n\n\n\n

Prospective randomized clinical trials are the most reliable way of ascertaining the causal effect of a treatment on patient outcomes. However, trial prioritization for both institutions and individuals remains a complex problem due to limited numbers of highly heterogeneous patients. This project will conduct federated retrospective analyses designed to assess the benefit of off-label drug use by pooling multiple disparate databases, to help prioritize and guide subsequent initiation and recruitment of randomized clinical trials. This will include evaluating the impact of the target drugs on patient outcomes from diseases similar to COVID-19, such as pneumonia or acute respiratory distress, generating artificial datasets using generative adversarial networks to asses performance of methods when \u2018ground truth\u2019 is known, applying the best methods to analyze the effect of the target drugs on the outcomes of COVID-19 patients across hospital systems, and using the results to evaluate the potential of these drugs and suggest guidelines for clinical trials.<\/p>\n\n\n\n\n\n

University of Washington:<\/strong> Jesse Erasmus, Deborah Fuller
Tufts University:<\/strong> Charles Shoemaker
InBios, Inc:<\/strong> Syamal Raychaudhuri
Microsoft:<\/strong> Grace Huynh<\/a><\/p>\n\n\n\n

During an emerging infectious disease outbreak, rapid deployment of an effective and highly specific therapeutic can dramatically alter the course of the epidemic. One gold standard approach is to use monoclonal antibody (mAb) therapy. Traditional approaches can take months or years to execute. The protein production and purification process can take months to develop and optimize and the result may not be a true representation of the target pathogen. In this project we propose a method to accelerate the production of high-quality mAb therapeutics by eliminating the need to produce recombinant protein antigens. To screen and identify candidate mAbs without a protein antigen, we propose here a sequence-based bioinformatic and machine-learning approach to rapidly identify candidate mAbs. If successful, this method will significantly accelerate the discovery and delivery of high quality mAb therapeutics specifically for treatment of SARS-CoV-2 infection, and establish a platform that can be applicable in future outbreaks of infectious diseases. In addition, our bioinformatic approach can easily be extended to other forms of monoclonal antibody therapy, including treatments for cancer, autoimmune diseases, and neurological conditions such as Alzheimer\u2019s disease.<\/p>\n\n\n\n\n\n

<\/div>\n\n\n\n

Mental Health Implications and Return to Work Projects<\/h3>\n\n\n\n\n\n

Rice University:<\/strong> Akane Sano, Fred Oswald
Baylor College of Medicine:<\/strong> Nidal Moukaddam
Microsoft:<\/strong> Mary Czerwinski<\/a>, Daniel McDuff<\/a>, Shamsi Iqbal<\/a><\/p>\n\n\n\n

This project aims to develop an intelligent agent for a wide range of workers during the COVID-19 pandemic, to manage both their work- and life-related activities. We will leverage intelligent agents, ubiquitous and affective computing, and machine learning that analyze users\u2019 behavior and emotion, to maintain or even improve their productivity and wellbeing in comparison with work and life outside of the pandemic. We will ask the following research questions: (1) how has the pandemic changed people\u2019s work and lifestyle patterns and how have people managed to maintain their productivity and wellbeing? (2) can a personalized conversational agent help users manage their work\/personal tasks and schedules better to improve their productivity and wellbeing during the pandemic and the transition phase of return to quasi-normal? (3) can the agent identify regular and irregular behavioral patterns related to productivity and wellbeing and usefully intervene on the user\u2019s behalf?<\/p>\n\n\n\n\n\n

Johns Hopkins University:<\/strong> Avanti Athreya, Youngser Park, Carey Priebe
Microsoft:<\/strong> Carolyn Buractaon<\/a>, Nick Caurvina, Darren Edge<\/a>, Jonathan Larson<\/a>, Chris White<\/a><\/p>\n\n\n\n

With more than 30 million claims to unemployment this year due to the pandemic, organizations are desperate to figure out how to get people back to work. This project will use network machine learning on enterprise communication and collaboration data to evaluate and detect network resilience in a post-pandemic world. This research is aimed at providing better guidance as many companies transition from \u201cwork-from-home\u201d back to their workplaces. For example, given a large tech company, can we determine which groups\u2019 productivity would benefit the most from returning to the office first vs which groups might be more resilient to working from home? This research will contribute network machine learning that will better inform companies on how to operate.<\/p>\n\n\n\n\n\n

<\/div>\n\n\n","protected":false},"excerpt":{"rendered":"

These collaborative projects with academia span topics including infection prevention and control, treatment and diagnostics, allocation of resources, mental health implications and return to work.<\/p>\n","protected":false},"featured_media":669432,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"msr_group_start":"","footnotes":""},"research-area":[13553],"msr-group-type":[243721],"msr-locale":[268875],"msr-impact-theme":[],"class_list":["post-669339","msr-group","type-msr-group","status-publish","has-post-thumbnail","hentry","msr-research-area-medical-health-genomics","msr-group-type-collaboration","msr-locale-en_us"],"msr_group_start":"","msr_detailed_description":"","msr_further_details":"","msr_hero_images":[],"msr_research_lab":[199561,199565],"related-researchers":[{"type":"user_nicename","display_name":"Eric Horvitz","user_id":32033,"people_section":"Program organizers","alias":"horvitz"},{"type":"user_nicename","display_name":"Kenji Takeda","user_id":32522,"people_section":"Program organizers","alias":"kenjitak"},{"type":"user_nicename","display_name":"Roy Zimmermann","user_id":33456,"people_section":"Program organizers","alias":"royz"},{"type":"user_nicename","display_name":"Javier Alvarez-Valle","user_id":32137,"people_section":"Researchers and engineers","alias":"jaalvare"},{"type":"user_nicename","display_name":"Shruthi Bannur","user_id":39213,"people_section":"Researchers and engineers","alias":"shbannur"},{"type":"user_nicename","display_name":"Melissa Bristow","user_id":38727,"people_section":"Researchers and engineers","alias":"mebristo"},{"type":"guest","display_name":"Tim Carroll","user_id":669363,"people_section":"Researchers and engineers","alias":""},{"type":"user_nicename","display_name":"Mary Czerwinski","user_id":32824,"people_section":"Researchers and engineers","alias":"marycz"},{"type":"user_nicename","display_name":"Darren Edge","user_id":31509,"people_section":"Researchers and engineers","alias":"daedge"},{"type":"user_nicename","display_name":"Simon Frost","user_id":38233,"people_section":"Researchers and engineers","alias":"sdwfrost"},{"type":"guest","display_name":"Rolf Harms","user_id":669366,"people_section":"Researchers and engineers","alias":""},{"type":"user_nicename","display_name":"Shamsi Iqbal","user_id":33592,"people_section":"Researchers and engineers","alias":"shamsi"},{"type":"user_nicename","display_name":"Emre Kiciman","user_id":31739,"people_section":"Researchers and engineers","alias":"emrek"},{"type":"user_nicename","display_name":"Jonathan Larson","user_id":32385,"people_section":"Researchers and engineers","alias":"jolarso"},{"type":"user_nicename","display_name":"Jonathan Lester","user_id":32333,"people_section":"Researchers and engineers","alias":"jlester"},{"type":"guest","display_name":"Oege de Moor","user_id":669369,"people_section":"Researchers and engineers","alias":""},{"type":"user_nicename","display_name":"Hannah Richardson (nee Murfet)","user_id":37703,"people_section":"Researchers and engineers","alias":"hamurfet"},{"type":"user_nicename","display_name":"Bichlien Nguyen","user_id":35942,"people_section":"Researchers and engineers","alias":"bnguy"},{"type":"user_nicename","display_name":"Aditya Nori","user_id":30829,"people_section":"Researchers and engineers","alias":"adityan"},{"type":"user_nicename","display_name":"Asta Roseway","user_id":31130,"people_section":"Researchers and engineers","alias":"astar"},{"type":"user_nicename","display_name":"Anton Schwaighofer","user_id":31059,"people_section":"Researchers and engineers","alias":"antonsc"},{"type":"guest","display_name":"Aqeel Siddiqui","user_id":669372,"people_section":"Researchers and engineers","alias":""},{"type":"user_nicename","display_name":"Karin Strauss","user_id":32587,"people_section":"Researchers and engineers","alias":"kstrauss"},{"type":"user_nicename","display_name":"Jeremiah (Miah) Wander","user_id":32896,"people_section":"Researchers and engineers","alias":"miah"},{"type":"user_nicename","display_name":"Chris White","user_id":31442,"people_section":"Researchers and engineers","alias":"chwh"},{"type":"guest","display_name":"Jenny Ye","user_id":669375,"people_section":"Researchers and engineers","alias":""}],"related-publications":[641829,738628,759400,811444],"related-downloads":[],"related-videos":[],"related-projects":[],"related-events":[],"related-opportunities":[],"related-posts":[],"tab-content":[{"id":0,"name":"Projects","content":"These collaborative projects with academia span topics including infection prevention and control, treatment and diagnostics, mental health, and return to work.\r\n

Infection Prevention and Control Projects<\/h3>\r\n[accordion]\r\n\r\n[panel header=\"Building an Open Platform for Pandemic Modeling\"]\r\n

University of Oxford:<\/strong> Deirdre Hollingsworth<\/p>\r\n

MIT:<\/strong> Chris Rackauckas<\/p>\r\n

Alan Turing Institute:<\/strong> Andrew Duncan (Imperial College London)<\/p>\r\n

Swansea University:<\/strong> Michael Gravenor, and Biagio Lucini<\/p>\r\n

University of Cambridge:<\/strong> Ronojoy Adhikari<\/p>\r\n

University of Warwick:<\/strong> Sebastian Vollmer(Alan Turing Institute)<\/p>\r\n

Royal Society Rapid Assistance in Modelling the Pandemic (RAMP):<\/strong> Graeme Ackland (University of Edinburgh), and Mike Cates (University of Cambridge)<\/p>\r\nMicrosoft:<\/strong> Simon Frost, Tim Carroll (Azure HPC), Rolf Harms (Cloud & AI), Grace Huynh<\/a>, Oege de Moor (GitHub), Aqeel Siddiqui (GitHub), Kenji Takeda<\/a>, Miah Wander<\/a>, and Jenny Ye (Cloud & AI)\r\n\r\nMathematical modeling of infectious disease transmission is an important tool in forecasting future trends of pandemics, such as COVID-19, and in evaluating \u2018what if<\/em>\u2019 projections of different interventions. Yet, different models tend to give different results. To draw robust conclusions from modeling, it is important to consider multiple models, which can be facilitated by expanding the modeling community, as models tend to reflect the people that develop them, and by bringing models and model developers together to compare and contrast models. The aims of this project are: to expand and empower the next generation of models and modelers; to act as an incubator for a platform for comparisons of infectious disease models, both in terms of computers and people; and hence enable swift and more robust policy decisions for current and future pandemics.\r\n[\/panel]\r\n\r\n[panel header=\"Sensing and Metagenomics for Continuous Viral\/Antibody Monitoring\"]\r\n

University of Washington:<\/strong> Shwetak Patel and Luis Ceze<\/p>\r\nMicrosoft:<\/strong> Jonathan Lester<\/a>, Bichlien Nguyen<\/a>, Karin Strauss<\/a>, Mike Reddy, Asta Roseway<\/a>, and Mike Sinclair<\/a>\r\n\r\nThe Molecular Information Systems Lab (MISL) and the UbiComp Lab at the University of Washington in collaboration with Microsoft Research are exploring hybrid molecular-electronic systems, with a particular focus on DNA computing as a hybrid approach for detecting COVID-19 in the environment. Detection of the virus has been demonstrated in city wastewater samples. As such, it has been proposed that environmental sampling locations, such as wastewater or air filtration systems, could be informative indicators for continuous monitoring to help cities stay ahead of community-wide COVID-19 spread. Towards that end, the objective of this research is to develop environmental sampling and pathogen detection techniques for air filtration systems on public transit networks, such as buses and subway systems. These techniques include safe sample extraction, qPCR, DNA sequencing, real-time nanopore protocols, and molecular computation. This offers the potential to detect the virus in buses, airliners and buildings, enabling new avenues for viral spread monitoring for a safer and better-informed pandemic response.\r\n\r\n[\/panel]\r\n\r\n[panel header=\"Commodity Smartphones for Face Mask Filtration Capability, Breathability, and Fit\"]\r\n

University of Washington:<\/strong> Shwetak Patel and Luis Ceze<\/p>\r\nMicrosoft:<\/strong> Bichlien Nguyen<\/a>, Jonathan Lester<\/a>, Karin Strauss<\/a>, Mike Reddy, Asta Roseway<\/a>, and Mike Sinclair<\/a>\r\n\r\nThere is currently a shortage of medical-grade face masks worldwide, and that shortage is likely to continue for the duration of the pandemic. As a result, the CDC is providing guidance on how people can create homemade face masks to mitigate the spread of COVID-19. There have been many recent efforts to evaluate the filtration quality of various everyday materials. The objective of this research is to leverage the sensors embedded on commodity smartphones to assess the filtration capability and breathability of homemade face masks. This effort will culminate in a mobile app that can be used to help end-users create better homemade masks, which would significantly limit the risk of virus transmission as social distancing measures begin to relax worldwide.\r\n[\/panel]\r\n\r\n \r\n\r\n[\/accordion]\r\n

Treatment and Diagnostics Projects<\/h3>\r\n[accordion]\r\n\r\n[panel header=\"Delineating Impact of COVID-19 Infection in High-Risk Populations\"]\r\n\r\nUniversity College London Hospitals NHS Foundation Trust (UCLH) and University College London (UCL): <\/strong>Joseph Jacob, Sam Janes, Daniel Alexander, Geoff Parker, Jerry Brown, Arjun Nair, Paul Taylor, David Hawkes, Marc Lipman, Joanna Porter,\u00a0 John Hurst, Nick McNally, and Bryan Williams\r\n\r\nMicrosoft:<\/strong> Javier Alvarez-Valle<\/a>, Usman Munir<\/a>, Melissa Bristow<\/a>, Melanie Bernhardt<\/a>, Anton Schwaighofer<\/a>, Jay Nanavati<\/a>, David Carter<\/a>, Ozan Oktay<\/a>, Shruthi Bannur<\/a>, Hannah Murfet<\/a>, Kenji Takeda<\/a>, and Aditya Nori<\/a>\r\n\r\nMicrosoft Research Project InnerEye team<\/a> in Cambridge (UK) is supporting UCLH and UCL to help identify vulnerable patients who are not currently covered by \u201cshielding<\/a>\u201d guidelines. Additionally, UCLH and UCL will help identify patients for whom shielding may not be necessary to avoid economic hardship imposed by unnecessary shielding in future Covid-19 outbreaks. UCLH and UCL will evaluate pre-Covid-19 imaging and clinical data to predict endpoints and speed up development of imaging biomarkers. They will look at cardiac computed tomography (CT) and use quantitative techniques to identify pre-Covid-19 cardiac imaging metrics that link to the likelihood of severe Covid-19 infection and cardiac events. The InnerEye<\/a> team will support UCLH and UCL, who will be developing models for the Prognosis of COVID-19 using CT imaging. This work is supported by the National Institute for Health Research Biomedical Research Centre at UCLH.\r\n\r\n[\/panel]\r\n\r\n[panel header=\"Prognosis of COVID-19 using Deep Learning Models of Chest X-Rays and Chest Computed Tomography (CT) imaging\"]\r\n

University Hospitals Birmingham NHS Foundation Trust:<\/strong> Shazad Ashraf, George Gkoutos, Andrew Beggs, Kal Natarajan, Elizabeth Sapey, Alastair Denniston, and Sharan Wadhwani<\/p>\r\nMicrosoft:<\/strong> Javier Alvarez-Valle<\/a>, \u00a0Melissa Bristow<\/a>, Melanie Bernhardt<\/a>, Anton Schwaighofer<\/a>, Jay Nanavati<\/a>, David Carter<\/a>, Ozan Oktay<\/a>, Shruthi Bannur<\/a>, Junaid Bajwa<\/a>, Usman Munir<\/a>, Hannah Murfet<\/a>, and Kenji Takeda<\/a>, and Aditya Nori<\/a>\r\n\r\nMicrosoft Research\u2019s Project InnerEye team<\/a> in Cambridge (UK) is working with University Hospitals Birmingham NHS Foundation Trust to develop deep learning models to analyze anonymized chest X-Rays and chest computed tomography (CT) scans to assist clinicians in determining disease severity, aid decision making, and improve our understanding of the disease. The aim is to improve the objective determination of disease severity by classifying and quantifying lesions in the lungs. This could help provide additional information for making a prognosis, assisting in the management of both hospitalized patients and their longer-term health needs.\u00a0 By quantifying disease progression, the model may aid hospitals in making decisions about resource deployment. It will also help to build our knowledge of the disease.\r\n\r\n[\/panel]\r\n\r\n[panel header=\"Federated Causal Inference for Multi-site Real-World Evidence & Clinical Trial Analysis\"]\r\n

Stanford University:<\/strong> Allison Koenecke, Ruoxuan Xiong, and Susan Athey<\/p>\r\n

Johns Hopkins University:<\/strong> Mike Powell, and Joshua T. Vogelstein<\/p>\r\nMicrosoft:<\/strong> Weiwei Yang, Emre Kiciman<\/a>, and Chris White<\/a>\r\n\r\nProspective randomized clinical trials are the most reliable way of ascertaining the causal effect of a treatment on patient outcomes. However, trial prioritization for both institutions and individuals remains a complex problem due to limited numbers of highly heterogeneous patients.\u00a0This project will conduct federated retrospective analyses designed to assess the benefit of off-label drug use by pooling multiple disparate databases, to help prioritize and guide subsequent initiation and recruitment of randomized clinical trials. This will include evaluating the impact of the target drugs on patient outcomes from diseases similar to COVID-19, such as pneumonia or acute respiratory distress, generating artificial datasets using generative adversarial networks to asses performance of methods when \u2018ground truth\u2019 is known, applying the best methods to analyze the effect of the target drugs on the outcomes of COVID-19 patients across hospital systems, and using the results to evaluate the potential of these drugs and suggest guidelines for clinical trials.\r\n\r\n[\/panel]\r\n[panel header=\"Development of a Rapid-Response Antibody Therapeutic Pipeline for Emerging Infectious Diseases\"]\r\n

University of Washington:<\/strong> Jesse Erasmus and Deborah Fuller<\/p>\r\n

Tufts University:<\/strong> Charles Shoemaker<\/p>\r\n

InBios, Inc:<\/strong> Syamal Raychaudhuri<\/p>\r\nMicrosoft:<\/strong> Grace Huynh<\/a>\r\n\r\nDuring an emerging infectious disease outbreak, rapid deployment of an effective and highly specific therapeutic can dramatically alter the course of the epidemic. One gold standard approach is to use monoclonal antibody (mAb) therapy. Traditional approaches can take months or years to execute. The protein production and purification process can take months to develop and optimize and the result may not be a true representation of the target pathogen. In this project we propose a method to accelerate the production of high-quality mAb therapeutics by eliminating the need to produce recombinant protein antigens. To screen and identify candidate mAbs without a protein antigen, we propose here a sequence-based bioinformatic and machine-learning approach to rapidly identify candidate mAbs. If successful, this method will significantly accelerate the discovery and delivery of high quality mAb therapeutics specifically for treatment of SARS-CoV-2 infection, and establish a platform that can be applicable in future outbreaks of infectious diseases. In addition, our bioinformatic approach can easily be extended to other forms of monoclonal antibody therapy, including treatments for cancer, autoimmune diseases, and neurological conditions such as Alzheimer\u2019s disease.\r\n[\/panel]\r\n\r\n[\/accordion]\r\n

Mental Health Implications and Return to Work Projects<\/h3>\r\n[accordion]\r\n\r\n[panel header=\"Design and Evaluation of Intelligent Agent Prototypes for Assistance with COVID-19 Work and Lifestyle Disruptions\"]\r\n

Rice University:<\/strong> Akane Sano and Fred Oswald<\/p>\r\n

Baylor College of Medicine:<\/strong> Nidal Moukaddam<\/p>\r\nMicrosoft:<\/strong> Mary Czerwinski<\/a>, Daniel McDuff<\/a>, and Shamsi Iqbal<\/a>\r\n\r\nThis project aims to develop an intelligent agent for a wide range of workers during the COVID-19 pandemic, to manage both their work- and life-related activities. We will leverage intelligent agents, ubiquitous and affective computing, and machine learning that analyze users\u2019 behavior and emotion, to maintain or even improve their productivity and wellbeing in comparison with work and life outside of the pandemic.\u00a0We will ask the following research questions: (1) how has the pandemic changed people\u2019s work and lifestyle patterns and how have people managed to maintain their productivity and wellbeing?\u00a0 (2) can a personalized conversational agent help users manage their work\/personal tasks and schedules better to improve their productivity and wellbeing during the pandemic and the transition phase of return to quasi-normal? (3) can the agent identify regular and irregular behavioral patterns related to productivity and wellbeing and usefully intervene on the user\u2019s behalf?\r\n\r\n[\/panel]\r\n\r\n[panel header=\"Organizational Dynamics to Enable Post-Pandemic Return to Work\"]\r\n

Johns Hopkins University:<\/strong> Avanti Athreya, Youngser Park, and Carey Priebe<\/p>\r\nMicrosoft:<\/strong> Carolyn Buractaon<\/a>, Nick Caurvina, Darren Edge<\/a>, Jonathan Larson<\/a>, and Chris White<\/a>\r\n\r\nWith more than 30 million claims to unemployment this year due to the pandemic, organizations are desperate to figure out how to get people back to work. This project will use network machine learning on enterprise communication and collaboration data to evaluate and detect network resilience in a post-pandemic world. This research is aimed at providing better guidance as many companies transition from \u201cwork-from-home\u201d back to their workplaces. For example, given a large tech company, can we determine which groups\u2019 productivity would benefit the most from returning to the office first vs which groups might be more resilient to working from home? This research will contribute network machine learning that will better inform companies on how to operate.\r\n[\/panel]\r\n\r\n[\/accordion]"}],"msr_impact_theme":[],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group\/669339"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-group"}],"version-history":[{"count":76,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group\/669339\/revisions"}],"predecessor-version":[{"id":971178,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group\/669339\/revisions\/971178"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/669432"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=669339"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=669339"},{"taxonomy":"msr-group-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group-type?post=669339"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=669339"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=669339"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}