DataVac

• 

  datavac_logo

• 

  OCR lab report reading

• 

• 

Inspiration

The abrupt outbreak of COVID-19 leaves the world unprepared for this pandemic. Yet we don’t have much knowledge of this virus and disease progression to guide clinical practices. i.e. what factors decide transition from mild cases to severe cases; how patients respond to different treatments.

Without data, we can do nothing. However, there are many barriers to systematically conduct clinical research. Data collection is the most crucial and the most time-consuming step, especially during a pandemic. This is because the clinical records need to be cleaned and transformed into a research-ready format. Doctors and nurses are at their full capacity at the frontline. Collecting data should no longer be an additional burden for doctors! Nevertheless, we need to push the studies forward. We need to cut the time spent on collecting and processing clinical data so that the research can be done rapidly.

What it does

DataVac aims to take care of the data management process so that the doctors and researchers can spend time focusing on doing analysis. DataVac provides fast and easy data collecting data exporting platform. It aims to simplify the clinical data collection procedure, reduce the data entering error and store data in secured data warehouses, and finally share the data with pre-consent parties such as hospitals and research institutes.

Our app-based data entry interface accommodates both user-defined manual data input as well as electronic record reading. Different from other data-entering interfaces, our app helps to reduce data entering error rate by applying anomaly detection algorithms in the backend. It also implemented OCR and text mining techniques to extract information from the lab test reports. The data entered from the app are readily transformed into standardized formats such as HL7 Fast Healthcare Interoperability Resources (FHIR) and will be submitted to the secured database for further usage.

DataVac provides online and offline mode options according to the user’s needs. The users take full control over the data ownership. Under the option of offline mode, the database can be stored in a local server with security protections. On the other hand, with the consent of the data-sharing parties, the collected data can be directly streamed to a user-defined database to achieve real-time data sharing. The authorized users can preview the summary of the database content, select the data of their research interest, send a request, and download the data within 10 minutes after unlocking the downloading interface.

Confidenciality considerations

In consideration of ethical conduct in research, DataVac is committed to ensuring the privacy of patients participating in clinical research and data protection consistent with the requirements of applicable laws. The data management procedures comply with the General Data Protection Regulation (GDPR) governed by the European Union. DataVac will not collect any personal information which immediately identifies a participant’s real identity. The personal information that is written on the multi-media clinical data such as laboratory results, ECGs and CT scan images will be removed before storage. We emphasize data security by strengthened cybersecurity, limited data accessibility, data encryption, and other advanced technology.

How we built it

We generated the initial data collection idea among the teammates with research backgrounds. As the newer member joined our team, we gained more insights regarding the difficulties in clinics. We have modified our prototype to accommodate the needs from clinical settings.

We built our data entry and retrieval Application based on R shiny.

Challenges we ran into

We also found a few challenges while implementing this project:

• The challenge of welcoming cloud technology in data storage and data streaming. With the understanding of data security and data governance concerns, we found that it is difficult to build reliance on using the service outside the individual hospital system.

• The challenge of the standardized medical system. The diversity in the EU had brought pros and cons to the health care systems. Our current solution cannot accommodate these diversities. However, we bear in mind that the demand for a united and regularized system is on the rise and we strive to improve our solution that can adapt to the diversities.

• The challenge of an approval waiting period for data collection. While the deployment of the App per se is quick and easy, we are aware of the approval from ethic reviews takes time. In the urgency like COVID-19 pandemic, this can be a huge impediment to rapid research production.

Accomplishments that we're proud of

Within a short period of time, we were able to build a functioning MVP. Our user tests did show that using the features implemented in our App, the time data entering process can be drastically shortened, the extreme value data entering errors can be avoided.

---

Impact on the crisis

In this COVID-19 pandemic, we urgently need scientific knowledge to understand the behavior of this virus and disease progression. Time is precious, therefore should be smartly used. Spending time on manual data entry is not a way to go. Our experiment shows that using DataVac can drastically reduce the time spending on data entering and cleaning. This means that we can see meaningful research findings sooner. Consequentially and collectively, we can find effective treatments faster and prevent more deaths.

What we learned

The manual input UI did not reach our expectation of time-saving advantage. We realized this may due to that it is equally not easy to maneuver the buttons and checkmarks and to orient the information input sections on the screen. However, the error alert system did prove our intended concept. We will keep this drawback in mind and try to tackle this problem in the next version.

What's next for Data Vac

Necessities in order to continue the project

First, we would like to seek collaborators to test our prototype and obtain feedbacks for user experience. We would like to improve our product based on the user experience in addition to a few more advanced technical features that have been planed for the next version. We also would like to seek committed and intelligent fighters in the tech field to join our team to improve our product.

Value of the solution after the crisis

DataVac is born under the crisis of COVID 19 pandemic. However, its value does not rest in this context. It is the debut of revolution in the data collection approach in scientific research. The traditional paper and pencil approach will be superseded by digitalized systems. Using artificial intelligence technology, DataVac minimizes the human input effort and makes the researchers refocus on what is most important to them.

Built With

• r
• shiny