Efficient diagnosis of pathological language patterns in health research
CHALLENGE #1
Efficient integration: How can complex analytics be seamlessly integrated into the digital platform ‘Mili’?
CHALLENGE #2
Scalability and efficiency: What measures optimise the performance and scalability of voice data analysis?
CHALLENGE #3
User-friendliness for all: How to create an intuitive solution that also allows non-experts to carry out complex language data analysis?
About the project
Project objectives
The project aims to develop an intuitive digital platform—Component, which greatly facilitates access to language data analysis, while at the same time improving health efficiency and user-friendliness;research improved. The focus is on expert support*inside and researchers when assessing medical conditions such as Alzheimer’s and schizophrenia through automated and AI-based language data analysis. A key aspect of the project is the automation of analytical definitions, which reduces the necessary manual effort and speeds up the analysis process. By simplifying their use, users without deeper technical knowledge should be able to make effective use of the enhanced functionalities of the platform. The aim is to make the platform more widely available Make applications accessible. The project therefore makes a significant contribution to the further development of the medical sector. Research it allows for a more precise and quicker assessment of pathological language patterns.
Starting position
AI:element develops AI-enabled language biomarkers for healthresearch, assess pathological conditions such as Alzheimer’s and schizophrenia. The digital platform ‘Mili’ allows voice datasamplification in clinical and domestic environments and the integration of language biomarkers. Challenges were related to the integration of complex analytics such as the language biomarker and optimising efficiency and scalability. A new engine allows researchers with no coding skills to have direct access to advanced languagefeatures, which simplifies and speeds up studies.
DESCRIPTION OF COMPANY
EDIH enabled us to develop directly implementable ideas that drive us forward in a sustainable way.
Janna Herrmann
Chief Technology Officer, KI:elements GmbH (ki:)
Approach
After an initial consultation and a detailed analysis of the company’s processes, the project was divided into three phases. In the first phase (January to April 2024), a prototype was developed to automate the creation of analytical rules for voice data. This prototype was implemented with the JavaScript and Flutter technologies and was finally made available as running software. The second phase (April-July 2024) focused on the development of a new feature called Sigma Feature Engine, which facilitates the analysis of language data. Care was taken to ensure that the software with an updated version of:r ki:elements’s proprietary language library Sigma remains compatible. The third phase (September 2024) looked at how the system works under high load and where there are vulnerabilities. The results were used to focus on phase three. The close cooperation with the Saarland EDIH allowed the team to carefully evaluate the business processes, assess feasibility and successfully develop the prototype.
Result of the project
The project result includes several software solutions that significantly improve the user-friendliness and efficiency of voice data analysis in Mili. Automation significantly reduces manual effort, minimises errors and saves costs in the long term. Instead of manually writing complex YAML code, users may:*en now just create analytical definitions via an intuitive surface. The system proposes automatically matching parameters and improves accuracy through type validation and code generation. The solutions increase productivity, support flexible combinations of biomarkers and engineers, facilitating data processing. The advantages for ki:elements are obvious: Time savings, error avoidance and user-friendly operation that it does nottechnicsxpert*en makes it possible to carry out analyses. In addition, a Sigma-Feature engine has been developed to efficiently process voice recordings and extract relevant features from a variety of tasks.
Main lessons learned from the project
This says ki:elements on cooperation with the Saarland EDIH
The support provided by the digitalisation project has been instrumental in bringing us forward with our digitalisation project.
Other digitalisation projects were subsequently addressed.
New innovative ideas for follow-up have emerged from the digitalisation project.
The project has helped to make our company fit for the future.
Our expectations for the project have been met.
We are very satisfied with the outcome of the project.