|
| 1 | +--- |
| 2 | +title: Hello World! |
| 3 | +author: Roni Rosenfeld and Ryan Tibshirani |
| 4 | +date: 2020-08-10 |
| 5 | +tags: |
| 6 | + - COVIDcast |
| 7 | +authors: |
| 8 | + - roni |
| 9 | + - ryan |
| 10 | +heroImage: blog-lg-img_hello-world.jpg |
| 11 | +heroImageThumb: blog-thumb-img_hello-world.jpg |
| 12 | +summary: | |
| 13 | + Hello from the Delphi research group at Carnegie Mellon University! |
| 14 | + We're a group of faculty, students, and staff, based primarily out of CMU |
| 15 | + together with strong collaborators from other universities and industry. |
| 16 | + Our group was founded in 2012 to advance the theory and practice of epidemic |
| 17 | + forecasting. Since March 2020, we have refocused efforts towards helping combat |
| 18 | + the COVID-19 pandemic, by supporting informed decision-making at federal, state, |
| 19 | + and local levels of government and in the healthcare sector. Until now, we've |
| 20 | + been pretty “heads down” with our work, and slow to communicate what we've been |
| 21 | + up to. But at last ... Delphi finally has a blog! This first post serves as an |
| 22 | + introduction of sorts. Future posts will dive deeper into our various projects. |
| 23 | +output: |
| 24 | + blogdown::html_page: |
| 25 | + toc: true |
| 26 | +--- |
| 27 | + |
| 28 | +Hello from the Delphi research group at Carnegie Mellon University! |
| 29 | +We're a group of faculty, students, and staff, based primarily out of CMU |
| 30 | +together with strong collaborators from other universities and industry. |
| 31 | +Our group was founded in 2012 to advance the theory and practice of epidemic |
| 32 | +forecasting. Since March 2020, we have refocused efforts towards helping combat |
| 33 | +the COVID-19 pandemic, by supporting informed decision-making at federal, state, |
| 34 | +and local levels of government and in the healthcare sector. Until now, we've |
| 35 | +been pretty "heads down" with our work, and slow to communicate what we've been |
| 36 | +up to. But at last ... Delphi finally has a blog! This first post serves as an |
| 37 | +introduction of sorts. Future posts will dive deeper into our various projects. |
| 38 | + |
| 39 | +## A Little Bit About Us |
| 40 | + |
| 41 | +We (Roni and Ryan) are co-leads and co-founders of the Delphi group. |
| 42 | +When we started the group back in 2012, |
| 43 | +our mission was to develop the theory |
| 44 | +and practice of epidemiological forecasting, with a long-term |
| 45 | +vision of seeing this technology become as universally accepted |
| 46 | +and useful as weather forecasting is today. |
| 47 | +In the following years, we developed both nowcasting and forecasting models, |
| 48 | +focusing primarily on seasonal influenza in the US. |
| 49 | +Our forecasting system is a perennial top finisher |
| 50 | +in the Centers for Disease Control's annual forecasting challenges. |
| 51 | +In 2019, we were named one of the CDC's two |
| 52 | +Centers of Excellence for Influenza Forecasting. |
| 53 | + |
| 54 | +Back when we formed the group in 2012, |
| 55 | +the "original four" members were David Farrow, Logan Brooks, Roni, and Ryan. |
| 56 | +By early 2020, pre-pandemic, we'd grown to about 7-8 members in size, |
| 57 | +and were happily and steadily making progress on flu forecasting and nowcasting, |
| 58 | +and making headway into dengue and norovirus. |
| 59 | +To learn about the work our group has done in the past, |
| 60 | +including some papers we've written, software tools we've built, |
| 61 | +and the real-time epidemiological data server we've been deploying since 2016, |
| 62 | +check out `r blogdown::shortcode_html("reflink", "/", "our website")`. |
| 63 | + |
| 64 | +When the COVID-19 pandemic arrived, we focused all our attention on it. |
| 65 | +Our Delphi team |
| 66 | +quickly grew to 30+ members, and is still growing. |
| 67 | +Most of our members are volunteers, |
| 68 | +drawing talent from within CMU and several other universities, |
| 69 | +including Stanford, UC Davis, and USC, and also from industry. |
| 70 | +The pace has been intense and dizzying at times, |
| 71 | +and we're infinitely grateful for the contributions |
| 72 | +and commitment of all our new members---we'd be nowhere without them. |
| 73 | +See `r blogdown::shortcode_html("reflink", "team", "here")` |
| 74 | +for a list of the Delphi team members. |
| 75 | + |
| 76 | +With new members comes a new breadth of expertise: |
| 77 | +our expertise now covers statistical modeling, computation, |
| 78 | +systems engineering, user interfaces, visualization, |
| 79 | +digital epidemiology, and epidemic forecasting. |
| 80 | +We've been extremely fortunate to have developed |
| 81 | +several new relationships with stakeholders and supporters in government, |
| 82 | +the healthcare sector, and the tech industry; and of course, |
| 83 | +we're proud and grateful for our existing relationships, |
| 84 | +especially our longstanding relationship with the CDC. |
| 85 | + |
| 86 | +## Our Mission and Strategy |
| 87 | + |
| 88 | +While Delphi's long-term mission remains to advance the theory and practice of |
| 89 | +epidemic forecasting, since March 2020 our goals are to help combat the COVID-19 |
| 90 | +pandemic and save lives and livelihoods. |
| 91 | +We aim to support informed decision-making at federal, |
| 92 | +state, and local levels of government and in the healthcare sector. |
| 93 | +Whenever possible, we strive to make our work useful |
| 94 | +to the private and public sectors, other researchers, |
| 95 | +the press, and the general public. |
| 96 | + |
| 97 | +Our strategy: |
| 98 | + |
| 99 | +1. Improve pandemic situational awareness and understanding |
| 100 | + by providing comprehensive, geographically-detailed, |
| 101 | + and continuously-updated indicators of pandemic activity and its impact, |
| 102 | + helping to make meaning out of the pandemic information deluge. |
| 103 | + |
| 104 | +2. Support local, state, and federal governments' |
| 105 | + ongoing decision-making in their attempts to balance |
| 106 | + public health concerns with economic preservation, |
| 107 | + by providing validated, verifiable, localized, |
| 108 | + short-term forecasts of epidemic spread and healthcare demand, |
| 109 | + under any assumed level of the local population's mobility and distancing |
| 110 | + behavior. |
| 111 | + |
| 112 | +3. Analyze and demonstrate the impacts of governments' |
| 113 | + past tightening or loosening of mitigation measures |
| 114 | + (e.g., opening or closing schools or businesses, |
| 115 | + imposing or lifting bans on gatherings, shelter in place orders, etc.) |
| 116 | + on a population's mobility and distancing behavior. |
| 117 | + |
| 118 | +4. Engage continuously with our target users to communicate |
| 119 | + our findings, and inform our directions and priorities. |
| 120 | + |
| 121 | +5. Make our products useful and accessible to other |
| 122 | + researchers and tool developers, to amplify their impact. |
| 123 | + To that end, we'll continue to make everything we invent or produce |
| 124 | + publicly and freely available as soon as possible |
| 125 | + and to the greatest degree allowed, including models, |
| 126 | + algorithms, software, tools, estimates, and forecasts. |
| 127 | + |
| 128 | +## What We've Been Up To |
| 129 | + |
| 130 | +So what have we been doing since March? |
| 131 | +We've been working towards our strategic goals above, |
| 132 | +and have made good progress on some of them. |
| 133 | + |
| 134 | +Here's a quick summary: |
| 135 | + |
| 136 | +- We've built a number of new indicators of COVID-19 activity. |
| 137 | + These are fine-grained geographically |
| 138 | + (most of them are available at the US county level) |
| 139 | + and temporally (all of them updated daily). |
| 140 | + They are designed to shed light on the current picture of COVID in the US, |
| 141 | + beyond the typical publicly-available metrics like confirmed cases and deaths. |
| 142 | + |
| 143 | +- Some of our indicators are based on massive-scale surveys that we’re running |
| 144 | + through partnerships with Facebook and Google, and others are based on |
| 145 | + aggregated counts from massive medical claims data sets through partners like |
| 146 | + Change Healthcare. |
| 147 | + |
| 148 | +- We've built a `r blogdown::shortcode_html("apireflink", "api/covidcast.html", "public API")`, |
| 149 | + and `r blogdown::shortcode_html("apireflink", "api/covidcast_clients.html", "R and Python packages")`, |
| 150 | + to serve our indicators to researchers and the public. This API provides new |
| 151 | + data daily. |
| 152 | + |
| 153 | +- We've built `r blogdown::shortcode_html("reflink", "covidcast", "interactive maps and graphics")` to |
| 154 | + display our indicators, and better inform the public and decision-makers. |
| 155 | + |
| 156 | +- We've developed forecasts of the future spread of the pandemic, |
| 157 | + validated them prospectively, and started submitting them to CDC. |
| 158 | + |
| 159 | +In subsequent posts, we'll dive deeper into many of these projects, and touch on |
| 160 | +what's next for us (a lot... in many ways, we feel that our most important work |
| 161 | +is still ahead of us). Until next time! |
0 commit comments