In the framework of the CANCERLESS and CO-CAPTAIN projects, technology plays a crucial role in the collection and analysis of data to improve healthcare. Developed by the Universitat Politècnica de València, CIDMA (CANCERLESS Intervention Data Management Application) is a platform created to centralise, digitise and optimise the management of pilot data, facilitating the work of researchers and healthcare professionals. In this interview, Vicent Blanes and Ascensión Doñate, both UPV Senior Researchers, explain how it works, its main objectives and the impact it is having on health research and decision-making.
When and how was the CIDMA platform conceptualised?
CIDMA was created during the CANCERLESS project, in fact, the name of the platform comes from CANCERLESS Intervention Data Management Application. At the time, the aim of CIDMA was to be able to aggregate information from different pilots carrying out a similar intervention and a common evaluation into a single homogeneous information system. For CO-CAPTAIN, an adaptation and improvement of existing technology has been proposed, since the interventions and multi-centre studies suggested have a similar methodology at the level of data collection and evaluation.
What are the main objectives of the platform?
CIDMA has three main objectives.
The first is to provide a centralised aggregation of data from different pilots into a single repository, allowing the parts of the project that rely on the use of data and derived scientific studies to be carried out more easily.
The second objective is the digitisation and pre-processing of the information. During the interventions this information is collected on paper by means of an interview in order to better serve the participants and create rapport. This information is then uploaded to CIDMA, where, apart from being digitised, it acquires a common format. Moreover, as the server remains active 24/7, the data is accessible to the members of each pilot at any time.
Finally, and perhaps the most important objective, is to provide navigational aids. CIDMA allows information about the participants to be stored, such as their contact details. Also, once information is entered, the most relevant health outcomes (such as self-perceived quality of life) are displayed in tables and graphs on the participant’s interface, allowing navigators to get an idea of the participant’s status at any time and without having to manually review the questionnaires.
How does CIDMA contribute to improving data analysis and the implementation of micro-simulations in the project?
The digitalisation, homogenisation and pre-processing of the information that is carried out through CIDMA removes part of the data cleaning tasks when processing the data. In addition, CIDMA allows data to be downloaded directly in CSV format from its interface, which also eliminates the need to consult a database and tabulate the results.
What technologies or methodologies have been used in the development of the platform?
In order to make CIDMA accessible from any device, a web application was developed. Given that the field of data analysis and development of predictive models is strongly linked to the Python programming language, we decided to use it as the basis of the platform, which is why we chose the Django framework as CIDMA’s main technology. In addition, we have used another technology in the user interface such as Bootstrap, to allow the elements on the screen to adapt to any resolution and HTMX, which allows operations to be performed on the page without the need for a total refresh of the page.
What are the main functionalities offered for data analysis and decision making?
Apart from the automatic pre-processing of data during input, CIDMA has the possibility of real-time calculation of metrics and scores derived from user responses, so that once the data is entered, browsers can access this information which can help them in making decisions related to the intervention.
How is the accuracy and reliability of the processed data ensured?
For all the features included in the application, a series of tests have been developed. CIDMA has a battery of unit tests that are automatically triggered when a new version needs to be deployed. This ensures that none of the main features are ‘broken’ by human error. Of course, data input, extraction and analysis have their own set of dedicated tests to ensure reliability and accuracy.
How is data confidentiality guaranteed?
CIDMA is deployed on UPV servers, whose physical access is restricted to authorised personnel. Regarding digital security, widely known and tested technologies such as Django or the NGINX application server have been used, as well as good practices in the creation and deployment of the software to guarantee security. Furthermore, the UPV has an active mechanism that allows it to search for vulnerabilities in the security of its applications, thus allowing it to take measures such as updating to a software version in which the vulnerability has been solved.
In addition to security to prevent unauthorised access, we have a multilevel backup system that saves the status of the database on a daily basis, allowing it to be restored in the event of a breakdown or human error.
What notable results have been achieved so far thanks to the platform? How many or which projects have so far used the platform for data analysis of their research?
So far, CIDMA has been used in the CANCERLESS and CO-CAPTAIN projects, both funded by the European Commission. In addition, it received recognition from the Innovation Radar programme of the same institution, which fills us with joy, as it shows that the work and effort have been positively valued by the European Commission. For all the above reasons, CIDMA has played a key role in the elaboration of deliverables and publications generated from data collected during the interventions.
How is CIDMA expected to evolve or be used beyond the COCAPTAIN project?
We hope to include CIDMA, and all the technology we have created and developed in other projects, since with each adaptation there is a process of improvement that directly helps the usability and quality of the software. Our aim is for CIDMA to be a good tool for browsers and that this software created for research can be put into practice and have a positive influence on people’s lives.
