SYNSYS: Syndromic Surveillance System

SYNSYS uses public domain data from search engines and applies machine learning at the elementary level to identify when search trends are abnormal based on past precedent. This data can be used to predict outbreaks and anticipate all-time highs once an outbreak is already.

The Team Lead and sole founder is Esha Datanwala.

The biggest problem with the handling of the COVID-19 pandemic in most countries worldwide was the lack of ability to predict how and when outbreaks would happen. This lead to an underestimation of the disease and an overwhelming load on healthcare facilities. All of this is because governments did not have time to even begin preparing before things got bad. SYNSYS addresses this problem at its core by being a predictive technology and using key behavioral indicators of the people to predict the outbreak weeks in advance so that governments and organizations have time to consolidate and prevent the outbreak from spreading in an uncontrollable way.

The target audience for this solution would be governmental agencies or organizations with international significance, as only in their hands would this solution be able to help. Ultimately, the people it hopes to help is every country's population, as pandemics and epidemics can severely affect all income groups and contribute to further socio-economic problems. 

The primary reason this solution relies only on search trends (which are public information) is that depending on smartphones or devices to track these things is not viable for lower-income countries. Even though COVID was mostly a disease spread by the privileged, there can be outbreaks in the future that are not spread in the same way and an effective solution must be cognizant of that. 

The public good from this solution is the knowledge it provides – it is completely transparent. It uses non-intrusive data and its findings don't need to be hidden behind some sort of paywall. The utility of this solution depends on the people wanting to prevent an outbreak in the future, which is in everybody's best interests. SYNSYS provides all the proof you need to act and makes changes to prevent an outbreak.

Considering the nature of this solution, it can have a tangible impact only when utilized by organizations or agencies that have influence over policy or healthcare. The analysis conducted by SYNSYS needs to be used to influence pre-emptive actions. I expect this solution to be very impactful considering its data-backed warnings; it targets aforementioned organizations and healthcare bodies who can prepare appropriate and draft policy or take on-ground action to contain outbreaks. Prevention is always going to be more effective than cure, therefore the tangible impact SYNSYS will create is a better and stronger response to disease outbreaks, especially now as epidemics and pandemics are going to become much more frequent and deadly, according to the United Nations' biodiversity panel.

A predictive solution can save lives, rescue economies, and overall make the standard of living in nearly every single country much better by reducing the risk of a widespread disease outbreak.

1. Development of SYNSYS: 4-9 months. This includes hiring developers, developing the system, alpha, and beta testing, UI designing and development, and ML testing.
2. Adoption by 1 government agency/international body: 1-2 years. Organizations or bodies interested in building their pandemic detection and response teams would benefit the most. This also takes into account negotiations and tweaking the system to be most accurate for the country/region (different search engines). This is an estimate that takes into account general bureaucracy.
3. Deployment: 2-3 months. As soon as SYNSYS is adopted, it can be put to use immediately. Within 1-2 months, it would have attuned to the country's or region's search trend's sensitivity and nuances and have adjusted its expectations for abnormalities accordingly. As mentioned previously, if put in use by state governments for localized data, it would take a month extra to be tuned accordingly.
4. Additional data sources: 1-2 years. This would take place at the same time as the adoption timeline. To make SYNSYS truly sensitive, having information specific to a country's healthcare system, providers, and hospitals would help SYNSYS adapt better and faster to the country's landscape. 

At present, the biggest performance indicator this proof of concept has had has been its ability to find the correlation between search trends for "COVID-19 Testing" and daily COVID-19 cases in the United States. While the original hypothesis of symptomatic behavioral analysis has been proved, this was the next step to ascertain whether behavioral analysis was going to be truly effective. 

Moving forward, once SYNSYS is being actively deployed and used, performance indicators will rise up as detecting future anomalies and using present correlations to predict more waves of infections. When being used by an actionable organization, SYNSYS' information will be able to prepare a better response and healthcare infrastructure. The biggest performance indicator in this would be a slight, slow rise in disease cases (whether COVID-19 or otherwise, as this system would be tracking trends for all common symptoms of illness), especially if compared to regions or countries not aware of the impending outbreak. The lack of a steep rise in cases shows the readiness of the infrastructure to tackle the outbreak; by making sure it never gets uncontrollable. 

The Trinity Challenge is very uniquely designed to place an emphasis on using tech and data to combat real-life outbreaks and infections. It actualizes the reality that we all live in where data can be an extremely powerful tool when wielded for the right reasons. SYNSYS is a proof of concept solution that is a bit unconventional in its set up and structure; it does not pry or seek to gain from people's personal information. It uses data anybody can find to help everybody. The hope is that The Trinity Challenge is able to see the sheer altruism in this solution and help make it a reality, as opposed to other avenues that would require some sort of moral compromise.

Organizations that would be able to help with the build and deployment of SYNSYS:

• Organizations that run their own search engines 
• Organizations specializing in machine learning and data science
• Organizations associated with governmental healthcare bodies or agencies
• Organizations associated with international healthcare bodies or agencies

Ultimately, more than anything, what SYNSYS needs to succeed is people and organizations willing to trust data. Having any organization in the aforementioned categories partnering with SYNSYS either to assist in the development stage or else in the deployment stage would be perfect.