LINKS:
https://github.com/EXYNOS-999
https://devpost.com/EXYNOS-999
www.notrishabh.co
DEVPOST PROJECTS:10
GOVT PROPOSALS: 1 FRENCH GOVT.
DEPLOYED: N
OTHER PROJECTS: 5
OUTIDE/ ONGOING: 3
PROJECTS:
http://darcie.me/
https://devpost.com/software/darcie-virtual-assistant-for-city-services
https://covid-basic-needs.github.io/JS-frontend/#
https://www.nokidhungry.org/
Team: Akeem Seymens Max S Yaakov Bressler
The Problem:
The covid-19 pandemic is wreaking medical and financial devastation. There are many who need help and don’t know where to find it.
Between schools closing and social distancing, foodbanks are facing more challenges with providing help than usual. People are offering help – but it’s hard to get the word out, especially when their time is spent securing limited resources. Our Solution:
Darcie solves this problem by consolidation the wealth of information of services, namely food, taking the burden off the food providers. Darcie’s sleek front-end is an intuitive tool for unlocking tens of thousands of live data sources which actively provide up to date information on various local services. Partnerships:
We’re not in this alone. Our team is partnering with several non-for-profits in the shared goal of maintaining up to date shared databases for resources. As an example, No Kid Hungry uses data we provide for their national text-support operation. See it in Action:
Call Darcie (toll-free 1-844-839-4334) at any time to find services near you. Alternatively, find help with our interactive resource map. Next Steps:
Our team is developing compatibility for our resources to be accessed by Google Assistant. We're also focused on scaling our data for more service options, locations in the US, and around the globe (we are almost ready to launch in India!).
https://docs.google.com/spreadsheets/d/1XMsJJIduO6yIGEo1VySXoz9YwbgtEL63-siNS_Q/edit#gid=0
https://airtable.com/tblTSuESZFXKltAFz/viwiUNgzIDEEhC1lP?blocks=hide
One-Stop-Shop Public Resource for Telehealth Services, Anywhere in the World.
TEAM:
Rafah Ali
Kanupriya Agarwal, MD
Physician Entrepreneur | Founder and CEO | WHO Digital Health Roster of Experts | TEDMED 2020 Research Scholar Karan Kataria
Inspiration
As a physician entrepreneur working in digital health, I realized that even I didn't know where to find all the available telehealth resources in one place. Therefore, felt a need to create a public, global, and exhaustive resource with accurate information on teleconsults services. During COVID19, this becomes an important must-know, must-have, must-do initiative for the greater good. What it does
Live directory to capture Telehealth providers from around the world, and their information accessible to the global community. One-stop-shop for finding access to healthcare providers digitally via Telehealth - removes redundancy of misinformation/lack of disposable information to public as a large-scale mass initiative. How we built it
We started with a Google spreadsheet, and later converted it into an Airtable for more functionality and scaleability Challenges we ran into
Distinguishing between regular/COVID pricing, as many services were in process of updating their policies
We lost some information (later retrieved) on providing edit access to visitors, therefore later restricted it to our team
Due to overwhelming community response, we needed to device ways to keep up with real-time document updating
Finding a place for B2B services due to many requests, so we decided to make a separate tab for them
Accomplishments that we're proud of
Released the document within about 36 hours of starting; with about 65 service provider entries on March 22, 2020, ~10:35p US ET. Within 36 hours, the LinkedIn post had 4.5k views, and we had received requests from all over the world for edits, additions, redistribution, and sharing. What we learned
We learned that there is a dire need for:
Awareness among public about telehealth services, particularly its uptake during such times
Widespread distribution efforts at individual and governmental level for a single Go-To resource directory for telehealth
All telehealth service providers to be listed in a global aggregate directory
Certification of such a global directory by a central healthcare body such as WHO
What's next for Global Telehealth Resource Aggregator
Continue to build out an exhaustive, complete database of existing and emerging telehealth services worldwide
Build our team
Scale globally for mass distribution
Potentially request certification from WHO
Folding@Together is an alternative way to contribute computing power towards protein folding simulations like Folding@Home, a vital tool for researchers in the search for potential cures to Covid-19.
https://devpost.com/software/folding-together
https://www.foldtogether.org/
https://github.com/MeanPug/folding-together/
https://youtu.be/Re99Sk1K0Yk
What is Folding@Together
Folding@Together is an effort to lower the barrier to entry for contributing to the Folding@Home network of protein folding simulation solvers. These solvers are an important tool in the arsenal of researchers searching for cures to diseases like Cancer, Ebola, and Covid-19. Contributing directly to the project is simple and free, requiring downloading of software which makes use of unused compute cycles to run folding simulations. However, for many the prospect of downloading and running software from the internet is a scary one. Compound this with the idea of running technical software in the background that one doesn’t understand and it becomes clear why adoption is tricky.
Folding@Together allows donors to contribute to the project without downloading software by accepting cash microdonations which are used to spin up virtual machines on the cloud to run Folding@Home software. These donations are pinned back to the individual donor allowing us to send a transparency report detailing how the cash donation was applied to simulation solving.
How it works How Folding@Together is built
Folding@Together is built on the AWS cloud and makes use of technologies like Python, Django, Kubernetes, and Lambda State Functions. We isolate individual donations via one-to-one mapping against Cloudformation Stacks configured to run Folding@Home solvers. The diagram below details a birds-eye view of this architecture.
Architecture
The project is completely open source (MIT License) and we welcome any and all contributions Folding@Together In The Future
While we accomplished more than I thought possible for the short period we had, there’s still a ton we can improve/refurbish, including:
Creating a backing 501c(3) organization for F@T to further incentivize donations by making them tax deductible.
Making node scheduling against the current credit balance a more intelligent process
Providing more information to potential donors on the project and its goals.
Expanding beyond just Folding@Home to other solver networks like Rosetta@Home
Optimizing node scheduling by adopting a cross-cloud approach
COVID-19 Global Hackathon 1.0
Created by
I worked on the back-end functions that track donations and schedule compute clusters to process folding@home work units using donations for funding.
Mike Atkinson
Mike Atkinson
I launched the Folding@Together project and implemented the frontend and payments processing Bobby Steinbach Bobby Steinbach
Founding partner @ MeanPug Digital, proud dog-dad Describe your contribution Cancel RISHABH CHAKRABARTY RISHABH CHAKRABARTY
UNDERGRADUATE RESEARCHER |DEEP LEARNING RESEARCH| COMPETITIVE PROGRAMMING| FPGA RESEARCH|HCI | GLOBAL VOLUNTEER @planetary-society
https://devpost.com/software/80_corona-contact-tracker
www.coronow.app
CoroNow app is the smartest way of social distancing. While it helps you to keep distance from other individuals it also informs you if you had contact with positive tested individuals in the past.
Social distancing is the most effective way to slow down the pandemic. Putting the whole population on quarantine might work for a few weeks before the worldwide economie collapses. Therefore we need to move away from quarantine towards smart social distancing and corona tracing apps like Coronow. Inspiration
We are currently at the start of an pandemic crisis. If no measures are taken, the virus will expand at exponential speed, our healthcare system will collapse, causing deaths in the thousands or even millions worldwide. We want people to stay safe and healthy and help them to act proactively in order to keep other people safe as well. Tech is the answer. What it does
CoroNow exchanges short-distance Bluetooth signals between phones to detect other participating CoroNow users in close proximity of up to 2 meters. Records of such encounters are stored locally on each user’s phone. If a user has been positively tested for COVID-19, his/her encounters within the past 14 days get notified (through a push notification) on what next steps to take (social distancing, observing whether they develop symptoms, contacting a near-by hospital etc). Additionally, the infected user can consent to send his/her CoroNow data to the health authorities.
This enables users to take the necessary action sooner, such as keeping isolated for the required time until they might show symptoms and are tested for COVID-19.
Additionally features are a chat bot/ questionnaire to ask the user for symptoms and general health conditions and to provide information. Also, there are regular reminders to take the necessary measures when you are on the move (social distancing, cleaning hands, keeping you hands off your face etc.) and social rewards if you adhere to them offering an external incentive for social distancing. As a view to external verified institutions such as health authorities, CoroNow offers a dashboard to share instructions and alerts. Challenges we ran into
We pivoted from tracking people’s movements and matching motion profiles to using bluetooth technology to identify close proximity to infected people due to 3 main reasons:
less data-security concerns. The data stays with the user.
less computationally expensive than matching entire motion profiles over days more accuracy
more accuracy than in motion profiles (users can have same GPS coordinates but not be physically close, e.g. in high buildings)
planning
communication
Accomplishments that we are proud of
One of the features is that health systems can be prepared by mapping a new infection wave. If someone is infected a push notification will be sent based on the movement profile. This tracking option can be used by the government which impose a curfew or by the authority which decide to place an infected person in quarantine.
Other interesting and helpful features include:
social rewards (gamification) to playfully increase awareness
innovative user interaction to make it easy for everyone to understand
What we learned
Coordinating a large group remotely can be challenge but you find solutions.
It is important and fun to appreciate each other's work.
A crisis like this makes people focus on the things that matter and help them become creative.
We are all in the same boat. This crisis leaves (almost) nobody unaffected. Working together towards a common goal helps to overcome this crisis that poses a challenge to all of us.
What's next for 08_Corona Contact Tracker_CoroNow
Get the prototype with core features up and running.
Test the prototype
Build collaborations with sponsors, governments, health authorities etc.
Clarify data security concerns, as (also the legal) situation is dynamic.
Iterate on the core functionality of the prototype.
Add additional features to app, such as notification of (negative) test results and measurement of blood oxygen test via smartphone camera according to Sp02 standard.
The information and logic can be adjusted as CoroNow is adapted by other countries.
Distrbuted contact tracing MeetSecure is a smartphone application that trains users to observe social distancing rules. It reminds them if they forget to observe the rules, so easily forgotten against an invisible danger. It allows them to prove their observation of distanciation rules. Users are motivated to develop reputations as people who protect others from infection. Their reputation is based on certificates validated by trusted sources and stored on immutable media. Cryptographic mechanisms ensure personal privacy and the integrity of certificates. Contact-tracing data and documentation are combined to produce risk estimates. Eventually, when a vaccine is available, a user could present a digitally signed vaccination certificate showing they present no risk at all. The risk estimates assist users in engaging in safe and rewarding social interactions, in a way that warrants privacy while recommending each citizen the maximum amount of freedom without becoming a risk.
Proposal submitted to the Frence Defecnce Innovation Agency.
Members:
David Stodolsky, PhD Institute for Social Informatics
Andrew Trask, @Openmined
...
...
...
MeetSecure for Risk Avoidance MeetSecure is a contagion containment application (app) allowing risk avoidance, as opposed to just the diagnostic confirmation offered by contact-tracing apps. That is, users can preempt an encounter based upon the risk of infection estimated by the app. MeetSecure provides warnings of social distancing violations that may expose the user to health threats.
A Bluetooth radio negotiation channel is automatically established when another user’s smartphone comes within ten meters of the user’s phone. The other person receives a “Do not disturb” message through their phone when they come within eight meter’s of the user. An alert is triggered, if they come closer without the user’s permission. Others without phones are detected by variation in the local electromagnetic and acoustic fields caused by their presence. A user would first receive a vibration from their phone warning of a social distancing violation. This allows them to cancel public acoustic or electromagnetic responses in case they are unnecessary. The phone could speak a warning message, if necessary to stop a social distance violation. A loud alarm signal and light flashes could follow, if this failed to resolve the issue. Contact-tracing apps are unable to accurately determine the risk of an encounter.
A combination of time and distance is typically used to indicate a risky contact. For example, co-presence within two meters for more than 20 minutes indicates possible transmission of the SARS-CoV-2 virus. However, two people might meet, hug, and kiss within one minute, thereby providing good conditions for transmission. This encounter would not satisfy the criteria for a risky contact. MeetSecure divides personal space into four zones of interaction: none, verbal, bodily, and cellular. While a discussion at a distance less than a couple of meters provides some risk of transmission, a kiss makes the transfer of infectious particles very likely.
The user initializes the app by entering interaction preferences and by collecting data already online that can be used in negotiation. One interaction preference could be bodily contact with a member of the opposite sex. A vaccination certificate or an immunity “passport” could be downloaded to the phone. Negotiation would end with transmission of a “Do not disturb” message, unless the approaching person was of the opposite sex, interested in bodily contact, and had presented either a vaccination certificate or an immunity passport, assuming this was the only interaction preference entered and it was specified that the interaction must be safe. Otherwise, negotiation would proceed first to discussion and then to bodily contact, assuming mutual consent was obtained. Consent could be signalled by confidential indication of a decision, such as rotation of the phone until two vibrations were felt. At the conclusion of the interaction, there would be an accurate record of transmission risk. Users would also be able to show that they had conformed to social distancing rules and had obtained consent for the interaction.
Lumiata Hackathon.
TEAM:
Yen Low.
Juan Banda
Rishabh C.
Solve underdiagnosis/undertesting of COVID-19 Crowdsource symptoms from social media to estimate COVID-19 burden and extent of underreporting due to undertesting/underdiagnosis of COVID-19 cases. Predict prevalence from symptomatic keywords and tweet text. Pro: tried-and-tested method (CDC already does this for flu surveillance) Con: our tweet set may undercount symptoms cos we’re limited to those with COVID keywords Ref: review, Northeastern, Google flu trends paper Data needed: tweets, case data, census data (maybe), symptoms list, symptoms data (Andy), more symptoms dictionaries: https://arxiv.org/abs/2003.09865 Seed tweet then scrape that user’s b4 and after tweets Misc: Google Flu Trends
Address public concerns by early detection of emerging topics. Spot hot topics emerging from twitter conversations so authorities and private enterprise can anticipate and meet the public needs in terms of communications, new products and services. Pro: familiarity with topic modeling; only needs 1 dataset Con: only requires one dataset so lower barriers to entry (tweets already hydrated and annoted!) Ref: early prototype, WaPo’s fact checker on HCQ Data needed: tweets Skills: LDA (daily/real-time), display topic trends, interpret trends
Geographical differences in attitudes towards COVID-19 Who were more likely to treat COVID-19 seriously vs being skeptical? Was this divide split along political/class lines and their disease profiles? How did their conversation topics differ (e.g. religion, government, science) Pro: uses several data sets and skills which we may be uniquely positioned for to have a study good enough to publish or at least a news article Con: Ref:https://diabetes.jmir.org/2020/1/e14431/ Data needed: tweets, case data, line list, census data, known risk factors Skills needed: topic modeling, sentiment analysis, geo, demography, econs, sociology, interpret text/geo, display maps
How are we flattening the curve?
Not all counties are the same in terms of underlying risk factors, density, etc. Are our current measures sufficient for flattening the curve. Do we need to do more, e.g. reduce mobility, close 20% of the shops? Create a simulation tool that tells us the number of lives saved per action.
Pro: a cool tool? Highly actionable to encourage good behaviors
Con: tons of modeling assumptions
Ref: Seattle, dashboard, geo spread, unacast, Kelvin Systrom’s notebook
Data needed: case data, tweets (to infer action and maybe mobility), mobility data, connectivity data
Skills needed: time series, mechanistic modeling, interactive/simulation dashboard
HARVARD UNIVERSITY.
Who We Are We're a team of students visualizing the Coronavirus pandemic. Specifically, we want to show objective statistics from reliable sources on what has been and what will be most impacted, provide help to organizations fighting on the frontlines, and develop data-driven policy proposals.
Published Work: Check out some of the work we’ve currently published: Resource spreadsheet and presentation for people to find and give help for the fight (April 17th) Self-updating visualizations of basic COVID-19 and hospital stats in the United States (April 15th) Visualization of unemployment by state in the last four weeks (April 10th ) Dashboard and presentation on COVID-19 in the state of Georgia (April 1st)
Current Partnerships
We at CVT are not in this fight alone. We’ve partnered with numerous organizations aiming to deploy visualizations and research to help fight COVID-19. Check out our current partners!
Center for Geographic Analysis
Opportunity Insights
Harvard Political Review
COVID Alliance
Harvard Innovation Labs
MIT Rapid Innovation Dashboard
Harvard Opportunity Insights (OI) Research Project Lead(s) Rishabh and Lucas
Project Description
This project is assisting Opportunity Insights by collecting county-level economic data, particularly unemployment insurance claims and bankruptcy claims. The focal point of this project would be a collection of real-time public data to create well-processed, high-quality datasets from unstructured data which will be analyzed by researchers and be used to translate these research findings into policy change. Our main deliverable is a github and airtable of links. For real-life problems and insights into the same, we need data that originated from the problem domains.
Some of the important tasks in this research project include Data selection: the selection of the right features for a particular dataset, Data preprocessing and sampling: organizing and formatting the data, cleaning the data, and Feature Engineering and extraction: optimizing the number of features and Data Conversion: Scaling and Composition-combining different features into a single feature; to create datasets with balanced taxonomy, sufficient data, high-quality labels/data, minimal errors and diversification of samples.
https://coronalert.me/
https://drive.google.com/file/d/15-RlFinWUVGSipqmJbswEC4p7qfmG0a0/view
Coronalert is a pandemic zone alarm that helps keep people private and safe, using the latest blockchain and encryption technology. Coronalert is a passive monitor: no personal data leaves your device
LUXETICS
https://www.youtube.com/watch?v=CzJvC1qnRrE&feature=youtu.be
https://docs.google.com/presentation/d/1p1ulpvU6R9mHxri6-H7Z3cuQZRLM72lWNeI5UiC1Avs/edit?usp=sharing
WINNER 2020 PHYSICAL HEALTH TRACK!!!
Inspiration Just like the stock index, the Twitter Pandemic Index reflects global sentiment about the COVID-19 pandemic.
What it does
Slide along the Twitter Pandemic Index and see the most frequent words, hashtags, emojis and symptoms mentioned on those up and down days. For example, the Twitter Pandemic Index follows the ups and downs of several milestone events (ref: CNN timeline): 2020-02-02: first death outside China (Philippines) 2020-02-07: the world mourns the death of Dr Li Wenliang whose early warning about the coronavirus was silenced by China 2020-02-29: first death in the US (Washington state) 2020-03-03: Federal Reserve cuts interest rate by 0.5% 2020-03-11: WHO declares COVID-19 a pandemic 2020-03-13: US announces relief package
One interesting observation is that since the middle of March, the Twitter Pandemic Index rose to a new high (but still not in the positive zone) when people worked from home and started using more ❤️ 🙏in their emojis, suggesting that working from home improved sentiment and unified people, bringing out the best in humanity.
https://docs.google.com/document/d/1o4C0nPYW-71K5oc9bybB1ml2Hxdx6_XsxdaPT5q7JXQ/edit?usp=sharing
https://devpost.com/software/tweet-pandemic-index
https://github.com/EXYNOS-999/decentralized-pandemic-reserve
https://www.youtube.com/watch?v=nG3cCnCNum8&feature=youtu.be
https://e2us1r.axshare.com/#id=e51bqy&p=choose_flow
Problems
Companies are repurposing their production lines to join the fight against COVID-19. However, pivoting manufacturing capabilities is no easy task and companies must overcome different levels of complexity in order to make this shift.
Material Example - car maker Shanghai General Motors Wuling (SGMW) was able to receive medical-grade textiles from a supplier that previously provided interior textile for cars.
Manufacturing Example - The idea here is to use automotive companies' idle capacity to serve as contract manufacturers (CMOs) for medical device companies, which are already producing at their maximum capacity.
Existing Supplies Example - Existing supplies in reserves. Tapping into community supply, combining resources into a general supply to move stock piles to areas of need.
Pandemics are a Global problem yet our responses are localized. Beating a pandemic requires international cooperation and resources. When Countries compete with each other for imported supplies, or local communities compete against world trade, in the end, this inefficiency costs lives. We propose a Decentralized Autonomous Supply Chain Database that can match resource need based on manufacturing equipment, materials, and product availability.
Create a global pandemic supply train of resources, supplies, and manufacturing capabilities.
A proactive response to outbreaks by using existing models to contain hotspots.
Match needs with resources
Solution Short-Term Impact:
The magnitude of the coronavirus crisis and the scale of resource coordination required has necessitated an international response. Connecting manufacturing with those who are in need and those who can help. Fulfilling immediate global supply needs through a global supply chain. Our platform would allow for a factory in Paris to pivot from perfume to hand sanitizer and matched with a grocery store in rural Arkansas in need of hand sanitizer for their work force.
Repurposing helps companies to protect their own workforce and serve the greater good. Repurposing also allows companies to keep production lines up and running in times of low demand, generate moderate revenues, and positively impact their reputation. Long-Term Impact
To create a global resource to respond to a pandemic. A Decentralized Autonomous Pandemic Reserve that offers every country:
Access to global, standardized data
Access to a global supply chain
Contribution to international reserves
Futures trading in international markets *
A Global response to a global crisis through an international cooperative DAO using AI to model, inform and manage global resources.
Technical Requirements (MVP in bold)
Supply Chain
Existing supplies in reserves. Tapping into community supply, combining resources into a general supply
Facilitate distribution of supplies
Manufacturing - Creating an aggregate list of needs (masks, ventilators) for all locations - Creating a list of manufacturers. - Incentives and payments for fulfilling orders
Emergency Response
A volunteer network where businesses and individuals can enroll.
Business: Commit to a reserve supply (5,000 masks, 10 ventilators, pivot as a local manufacturer
Individual: Commit to serve (Doctors, nurses, admin, transportation…). A standing Medical Reserve
Locations: Convention Centers, Sporting Arenas, warehouses. Space for beds, procedures. - Allocating and distributing supplies, resources, and locations through a DAO.
Ex: NYC needs 200,000 masks for the next week. Paris needs 500,000. San Francisco needs 600,000. Global supply produces 200,000/week. Based on outbreak curves/models, allocate masks to all 3 locations based on curve %