The MediShout app is a digital platform designed to solve various key medical problems in hospitals. It connects staff, equipment suppliers and service maintenance teams, a smart way to ensure that patients receive the most efficient care.
The MediShout app delivers the answer to the real-life problems that medical staff is facing daily. Broken IT, estate and facilities problems, faulty equipment or PPE, and stock shortages are among the main issues stopping medical staff from taking proper care of their patients.
MediShout not only manages these hospital problems but also improves efficiency and prevents delays by allowing hospital medical staff to instantly report and resolve ward issues.
Model: Team extension
4 team members
Responsive design
Project duration: Initially 6 months but now ongoing.
Not all healthcare staff have access to their mobile devices so MediShout needed to provide a browser alternative for all app functions. The MediShout web-app also hosts all the admin functions for helpdesk teams. Staff can report any non-clinical issue they face in hospitals, we call these “Shouts”, and this goes directly to the team responsible for resolving the issue. Often reporting is passed onto third party helpdesks via integrations.
MediShout prioritised fast pace development to keep up with the challenges their customers were facing during the COVID pandemic.
The MediShout mobile application was written in Flutter - a very fast growing framework with an enormous society solving almost every business problem in record time. Flutter is suitable technology for this kind of app because it combines very quick development with a beautiful user interface for Android and iOS with one codebase.
The MediShout frontend is relatively simple and intuitive and does not require lots of mobile platforms native code base changes - which is another reason why Flutter is here the best option. Of course the app is using many more advanced functionalities f.e. Barcode scanning, hiding app in background due to GDPR restrictions, biometric verification coming back from background mode.
The existing app was set up on Firebase which is an easy to start but hard to keep up with business needs solution. We decided to change the technology to Azure, Django (Python framework) and Kubernetes - a professional technologies mix that delivers unlimited development possibilities.
The most demanding issue was app adoption, integrating the app with multiple suppliers. This required direct cooperation between our development team and specific external suppliers outside the MediShout organization.
We have used our standard approach to the infrastructure, leveraging Kubernetes as the main deployment and hosting solution. This gives us the ability to scale if needed, and also automates most deployment processes.
We are focusing on making the backend secure, since it handles the core of our business logic. This means the web and mobile apps are fast and lightweight, their main purpose being an end-user interface to the system.
With this setup, we can keep control of every aspect of user management, easily integrate with 3rd parties and focus on implementing business features.
Medical staff can quickly share any operational issue with a user-friendly app. Different reporting pathways are created to solve every kind of problem, from broken computers to low stock and faulty equipment.
The system allows you to create reports of shouts sent via the app, or export data via a csv. file.
An intuitive dashboard helps navigate through shouts already sent to the system. Suppliers and service providers can get an Excel file of shout data or scan through the history.
The app makes it easy to attach photos of broken equipment, ensuring the engineers have all the information needed to resolve problems.
Whenever a problem is solved, the users are notified.
Suppliers and users can chat to make sure suppliers gather all the information about the problem, which fosters better problem solving.
MediShout’s web application is built with React, and written in 100% TypeScript. The backend is our standard solution involving the newest Python version, Django 3 and Django REST Framework. The mobile app is written in Flutter, available for both iOS and Android devices. MediShout is hosted using Microsoft Azure cloud and runs using the cloud Kubernetes engine.
Azure was selected for both business and developer reasons. From the team perspective, this solution is providing a whole set of our favorite tools, such as Kubernetes integration, load balancers, and static IP addresses with associated domains. Azure ships with its own `azure-cli` tool that makes the authentication and deployment process even easier. The documentation for building infrastructure is well written and covers most of the common scenarios. There is also a huge number of blog posts and tutorials from individual sources available, which is a huge advantage when picking a well-established solution like Azure. Last but not least, we were pleased with the clean Azure Portal interface, which is readable for both developers and staff members.
Switching from Google Cloud to Azure was surprisingly easy. Setting up a basic infrastructure for a project, including a testing environment for the dev team, database servers, file storage, handling of secret values, and automation of TLS certificates were done in one sprint.
Microsoft Azure is a leading infrastructure provider in the industry and was proven compatible with our common technology stack. Although usually considered more tied to other Microsoft technologies, like C# or .NET, it works just as well with our favorite Python tools. We are glad to expand our portfolio with Azure, and we are ready to use it in the future.
NFTs are an important driving force for crypto, already bringing in many new users to the space. CYBAVO–a blockchain and digital asset security provider headquartered in Singapore–wanted to leverage this new opportunity and add NFT support to their offering. To do so in a timely manner while ensuring security, we based it on our own OpenNFT marketplace, with further customizations.
