SolDevelo developed a high-frequency image capture app for QED.ai, enhancing AI-driven data collection in healthcare and agriculture. With georeferencing and ODK integration, our solution streamlines data gathering, improving AI model training.

The Client

Founded in 2008, QED.ai is a pioneering technology company dedicated to leveraging artificial intelligence to solve critical data challenges in healthcare and agriculture – two sectors that are essential for global well-being. Recognizing that a lack of reliable, structured data often hinders progress in these fields, QED.ai has developed innovative solutions that streamline data collection, analysis, and utilization.

QED.ai specializes in technologies designed to increase efficiency, reduce costs, and minimize human error – key factors in environments where resource constraints can severely impact outcomes. One of their flagship solutions enables rapid digitization of paper-based patient forms, eliminating manual data entry errors and accelerating access to medical information. Additionally, they have developed advanced sensor-based devices that measure critical agricultural parameters with a single click or even through fully automated processes.

With operations primarily focused in African and Asian markets, QED.ai strives to empower local communities by providing AI-driven tools that enhance decision-making, optimize resource allocation, and ultimately improve both economic and social conditions.

The Challenge

To further expand their AI-driven data solutions, QED.ai needed a reliable and efficient mobile application that could facilitate high-frequency image capture – a crucial component for training artificial intelligence models. Their goal was to collect vast amounts of visual data, which could then be processed to improve AI algorithms.

Key challenges they faced included:

• High-frequency image capture – The application needed to support continuous, rapid photo-taking (e.g., every 3 seconds) without compromising performance or usability.
• Georeferencing capabilities – Since many AI models in healthcare and agriculture rely on location-specific insights, the app had to automatically record GPS coordinates for each image.
• Seamless data collection – The images had to be captured in a structured manner to enhance the quality of training datasets.

By addressing these challenges, QED.ai aimed to enhance the accuracy and efficiency of AI training, making their solutions even more impactful in the communities they serve. They turned to SolDevelo to bring this vision to life – leveraging our expertise in Android development, and open-source technologies, especially ODK.

The Solution

To meet QED.ai’s need for an efficient, high-frequency image capture tool, we developed a custom Android application that not only fulfilled the core technical requirements but also provided additional functionalities to enhance usability, optimize data collection, and improve AI training workflows.

Our solution was designed with the following key capabilities:

Flexible photography modes for various use cases

• Manual mode – This mode allows users to capture images by pressing a button, offering full control over the photography process. This is particularly useful when capturing specific subjects that require human judgment.
• Automatic mode – In this mode, the app continuously captures images at predefined intervals, allowing the user to focus on framing shots rather than repeatedly pressing a button. This ensures consistent data collection, and enables hands-free operation, which is valuable for fieldwork in agriculture and healthcare settings.

Configurable time intervals for customization

Given that different AI training datasets require different levels of detail, we incorporated a customizable time interval feature. Users can set the time gap between consecutive photos. This flexibility ensures that data collection aligns with QED.ai’s specific AI model training needs.

Georeferencing for enhanced AI insights

To further enrich the captured data, every image is tagged with precise latitude and longitude coordinates, ensuring that AI models can, for example, correlate environmental and geographical factors with agricultural data or track healthcare trends by mapping patient records across different regions.

Integration with ODK Collect

One of the most important aspects of our solution was its integration with ODK Collect, a widely used open-source tool for mobile data collection. It is a trusted platform that enables field teams to collect structured, offline-compatible data via mobile devices, which is later uploaded and analyzed in centralized systems.

Integrating our solution with ODK Collect ensured that:

• Users could seamlessly transfer captured images and metadata into existing ODK workflows.
• The application worked smoothly within established data collection pipelines, eliminating extra steps for field researchers.
• The solution remained compatible with widely adopted open-source tools, ensuring scalability and future adaptability.

SolDevelo has extensive experience in developing and enhancing ODK-based systems. By leveraging our deep knowledge of ODK’s architecture, we ensured that QED.ai’s solution was not only functional but also aligned with best practices in digital data collection.

The Results: Delivering a powerful user-centric solution

Our efforts resulted in the application tentatively named QED Camera.

Beyond meeting the fundamental technical requirements, we also incorporated additional features to improve the overall user experience and AI data quality:

• Customizable image compression – Allows users to adjust the balance between image quality and storage efficiency, ensuring optimized data transfer speeds in low-bandwidth environments.
• Silent mode – Disabling the shutter sound ensures that image capture remains discreet, which is especially useful in sensitive healthcare settings.
• Live indicators – Users can monitor:
  • The number of images captured,
  • The time elapsed during a session,
  • The available storage space.

These enhancements simplify data collection, reduce errors, and empower field workers with real-time control over their data-gathering process.

A solution built for impact

We delivered a product that not only met all client expectations but also offered extra functionalities that facilitate further development of AI-based solutions. Thanks to this application, QED.ai can continue to expand its initiatives, which in the long run will contribute to improving living conditions in developing countries, both in terms of finances and time management, as well as support the growth of local communities.

For us at SolDevelo, the collaboration with QED.ai was not only an opportunity to apply our expert knowledge – especially in integrating the ODK system, but also a chance to contribute to global initiatives.

Technologies used