DocToRead is an innovative app based on AI technology that translates medical documents into easy-to-understand language. The app uses advanced technologies to decode so-called “doctor’s Latin” and complicated medical wording and translate it into clear, understandable text.
Our client approached us with their idea for an app. This should make it easier for non-medical professionals to better understand medical documents such as doctor’s letters. In addition to the need for the app to be able to read the necessary documents (texts and images) automatically (OCR: Optical Character Recognition), it also needed to reproduce the recognized content in a simplified manner. It was therefore necessary to connect the app to an AI in order to achieve the desired results.
In addition to extensive research into the AIs required for the app’s backend, our designers began to create a user-friendly and appealing frontend for DocToRead. Our recommendation was to implement the application based on the Flutter framework, as this allows mobile apps for both iOS and Android to be implemented quickly and easily.
In addition to developing the app and connecting the AI, a major challenge was evaluating the optimal prompts, i.e. the specific inputs to the AI. It quickly became clear that the app could not simply be built around the AI. We then quickly decided to consult our software testing team.
Our idea was that our team of testers, experienced in both manual and automated testing, would be best placed to formulate a result expectation and figure out the best way to get to that result. While the initial test data was being evaluated, it quickly became clear that prompts could be greatly optimized. Prompts are multi-layered and the results that the AI delivers are very dependent on the way a prompt is formulated. There are a variety of functional settings, such as the type of language (technical or colloquial), the context (professional or casual), but also cultural influences. For example, the same prompt with different language variants in the AI (German, English, Spanish, etc.) also led to different results. The biggest challenge was the multilingualism of the app. It was more complex than expected to find a prompt that would lead to the desired and, above all, reliable result in the languages used.
Another major challenge was compliance with data protection guidelines. In close cooperation with various specialist lawyers and organizations, we succeeded in implementing the app securely and in compliance with the GDPR.
Through constant testing, translating, comparing and optimizing, an app was created that serves as a kind of medical interpreter in conjunction with AI. This offers enormous added value, particularly in the medical environment, for patients who are mostly laypeople in this respect.
However, the project is far from over. Depending on AI releases, the prompt will have to be adapted again. We are also considering implementing language-dependent prompts in the future – a really exciting topic.