Brief

Staying safe in COVID times has been a problem for over a year now.


What is Covishield

The coronavirus disease (COVID-19) pandemic has placed the clinical laboratory and testing for SARS-CoV-2 front and center in the worldwide discussion of how to end the outbreak. Clinical laboratories have responded by developing, validating, and implementing a variety of molecular and serologic assays to test for SARS-CoV-2 infection. This has played an essential role in identifying cases, informing isolation decisions, and helping to curb the spread of disease. However, as the demand for COVID-19 testing has increased, laboratory professionals have faced a growing list of challenges, uncertainties. The key problems with conventional testing is that it requires time and expertise which is technically not possible in certain scenarios.

Introducing Covishield a Rapid health tracker.

Logo

Here how it works

Our system is highly functional and sophisticated , for accurate health data analysis we sync user data either from their phone (only with sensors ) or through their smart health band, for eye data analysis user are asked to adjust their eye perfectly to the frame to take image, all these data are passed to machine learning model deployed in Google cloud.

These models are trained with bunch of data and images for better results, our data are authentic and taken from various sources.

Features.

1. Facial Recognition .

2. Audio to text.

3. Real time Heath data sync from smart devices.

4. Video emotion and sentiment analysis.

5. Eye data analysis.


Step by step

  1. start with our home page.

  1. Enable your webcam.

  1. Scan the barcode to sync accurate Health data.

4.Complete all task asked for.

5.Finally see your results.


Research

Eye data.

The standard practices for rapid risk screening of COVID-19 are the CT imaging or RT-PCR (real-time polymerase chain reaction). However, these methods demand professional efforts of the acquisition of CT images and saliva samples, a certain amount of waiting time, and most importantly prohibitive examination fee in some countries. Recently, some literatures have shown that the COVID-19 patients usually accompanied by ocular manifestations consistent with the conjunctivitis, including conjunctival hyperemia, chemosis, epiphora, or increased secretions.

Our new screening method of analyzing the eye-region images, captured by common CCD and CMOS cameras, could reliably make a rapid risk screening of COVID-19 with very high accuracy.(https://www.medrxiv.org/content)

.


Health data.

During our initial research we found some of the common symptoms includes , body temperature rises to 39.5°C (103°F), heart rate 110 beats per minute, respiratory rate 24 breaths per minute, and blood pressure 130/60 mm Hg. The oxygen saturation is 87% while the patient is breathing ambient air. The white-cell count is 7300 per microliter with lymphopenia. presence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).(https://www.nejm.org/).

2000 rows of Datasets to train ML model (these are taken from various internet sources and are not confirmed to be 100% accurate).

Spread sheet link

Using dataset (2000 rows) and Google cloud autoML, we developed a COVID-19 mortality prediction model that showed high accuracy (AUC=0.99) when applied to test datasets. This model was based on three clinical features: minimum oxygen saturation over the course of their medical encounter, body temp, pulse rate and Respiration rate.


Creation Process

UI/UX

Figma. Original design
Content Cell Content Cell
Content Cell Content Cell

System Architecture

References

We went through lot of articles, journals and Research papers to obtain required data.

  1. https://ascpt.onlinelibrary.wiley.com/doi/10.1111/cts.12874.
  2. https://www.nejm.org/doi/pdf/10.1056/NEJMcp2009575?articleTools=true.
  3. https://www.the-scientist.com/news-opinion/algorithm-spots-covid-19-cases-from-eye-images-preprint-67950

What's next for Covishield


Our system has a lot of versatility but to even start effectively in the future, we plan to implement ou


Share this project:

Updates