Health Information Ready for Analysis

The project focuses on the application of artificial intelligence techniques, such as machine learning, in the healthcare sector, where their impact is increasingly significant, as reflected by over 1,000 publications on PubMed in 2018 alone.
However, the accuracy of these techniques is strongly influenced by three main factors: the quantity of available data, the quality of that data, and data privacy.
The availability of data requires integrating AI techniques into Big Data architectures capable of storing and processing large volumes, as well as generating synthetic datasets when real-world data is insufficient.
Data quality is critical for analysis, as errors or incomplete information can lead to significant economic costs and directly affect patient health.
Studies indicate that a high percentage of the U.S. population has experienced medical errors, with misdiagnosis being the most frequently reported cause.
Improving data quality is therefore a critical, repetitive, and costly task, often representing the majority of a data analyst’s effort.
Recent surveys confirm that this remains the main challenge in healthcare data analytics projects.
The project addresses these issues by developing methodologies and tools to optimize data integration, cleaning, and exploitation, ensuring accuracy and reliability in predictive models.
The expected outcomes include improving healthcare system efficiency, reducing medical errors, and providing robust support for AI research applied to healthcare.