Medi Scanner

Logo for the application
login screen
Sample Prescription used
Medicine Ordering Screen
AWS architecture flow digaram

💡Inspiration

The trouble to enter medicine names manually in online medicine booking applications and also adding prescriptions to justify inspired us to build the solution. This is a pervasive problem for most of us. So we propose to save your precious time by providing you with an application which can list and book all your medicines prescribed by just uploading the prescription into the application.

⚙️What it does

This solution is a add-on feature of already existing Medicine provider apps. It is an application where user can upload his/her prescription from camera or gallery and get the list of medicines prescribed to him. User can also book the medicines through the application without having to explicitly search for the medicines.

🛠️ How we built it

• This solution uses react native for mobile.
• For backend it uses AWS services like Textract, AWS Comprehend Medical, SQS, SNS, Lambda, Dynamo DB, API Gateway. Please refer Mural link for design flow.

⚠️ Challenges we ran into

Learning Curve: For us Medical comprehend and Textract were new services. AWS documentation and engineering blogs helped us to reduce the complexity for use. We faced difficulty in uploading image to S3 directly from mobile app.

Problem Space:

1- Medicine Detection: Medical comprehend is not able to detect all the medicines. Ex: It detected Tab Pruslatin-M as two separate medicines Tab and Pruslatin-M. We could have used SageMaker , Web Scrapping and build additional ML engine to refine the outcome. But due to time limitation and late joining we were unable to do that.

2- Quantity calculation: Once we receive doses from Medical comprehend, to place order system is supposed to calculate the quantity for medicine. As its addon in existing app, we assume that they have micro service which will give us number of unit in one piece.

👩‍💼🧑‍💻 Accomplishments that we're proud of

• We are proud to make this project functional and working to help users in a certain manner.
• We are immensely proud of achieving a good team contribution with each of the members has contributed to the project in some way or other.

⚓ What we learned

• We learned about how to integrate or work with Mural.
• We learned many technologies and versatility of AWS services like Medical Comprehend and Textract.
• We learned uploading image to S3 directly from Mobile APP using IAM role and React Native.

🤝 What's next for Medi Scanner

• Refinement of Medicine prescription validation logic.
• Refinement of Quantity calculation logic.
• Add new feature of medicine reminder.
• Add new feature of showing diagnosis, treatment, etc.

Built With

Submitted to

Created by

I worked as backend developer (python + AWS stack). Developed backend for Medi Scanner app using lambda and other AWS services.

Rishabh Saxena
I worked as backend AWS developer in the application development and wrote lambda functions in AWS for the e2e flow.

Sakshi Agarwal
I have helped to designed the end to end Flow and Low level system design

Vikas Singh
I worked as a frontend Mobile App Developer and created image upload feature from camera / gallery, processing it from Frontend and send it to AWS S3 Storage.

Manas Vijaywargiya
I worked as a React Native Mobile developer created Sign up and login screen. Supported for API call and Navigation. Created a logo and video for application to publish on multiple platforms.

Madhavi Vishwas