GMV applies artificial intelligence and "Real World Data" for the early detection of prostate cancer

Early detection is vital for the survival of men afflicted with prostate cancer. This is why, on World Prostate Cancer Day, GMV shares the progress of two projects in which it applies its technology to contribute to this goal: the Evidence Hub of the European Association of Urology (EAU) and TARTAGLIA. Both projects seek to obtain clinical evidence using large volumes of data obtained from patients who have suffered or presently suffer from these tumors with the aim of improving diagnosis, knowledge of the disease, improving prognosis and offering tools that can be applied to clinical practice to improve the quality of life of patients.

In order to set up a “UroEvidenceHub” EAU Guidelines - Uroweb evidence center for the European Association of Urology (EAU), GMV has developed an application to record anonymous data from patients with urological diseases, including prostate cancer.

Likewise, it has developed all the necessary technology to deploy a federated network that allows machine learning models to be trained safely using “Real World Data.”

Configured as a consortium, the project aims to create a federated data network to accelerate the application of artificial intelligence (AI) in healthcare systems. In the specific case of the prostate cancer use case, the data comes from three clinical partners of the project: the Foundation for the Promotion of Health and Biomedical Research of the Valencian Community (FISABIO), the Galician Health Knowledge Agency (ACIS/SERGAS) and the Vall d'Hebron University Hospital Foundation - Research Institute (VHIR). Its technical collaborators are the National Computing Center-Barcelona Supercomputer Center and the companies Veratech and Opinno.

Prostate cancer: 3rd cause of death by neoplasia in men

Although 60% of patients with prostate cancer are diagnosed at an early clinical stage, it is still necessary to make the population aware of the need for regular medical check-ups to detect the disease in its early stages. It should be noted that prostate cancer is the second most common tumor worldwide, affecting more than 1.2 million men around the world.

For this reason, it is essential to carry out annual check-ups starting at age 50, since, like female breast cancer, prostate cancer is a hormone-dependent tumor whose prognosis improves significantly when diagnosed in its early stages. In advanced cases, surgery (radical prostatectomy) and radiotherapy are treated.

With this evidence center, the European Association of Urology (EAU) aims to make the largest European "Real World Data" database in urology available to specialists, facilitating the application of evidence-based personalized medicine as well as the development of clinical guidelines for the benefit of the patient. Its deployment has been possible thanks to the previous work carried out in the PIONEER projects (IMI and the EU project for the improvement of the diagnosis and treatment of prostate cancer with big data) and OPTIMA (IMI project for the creation of a platform for the prostate, breast and lung cancer research), in which GMV is also a technology partner. In the specific case of OPTIMA, GMV is currently working on defining the technologies necessary to carry out data federation, which will allow structured and unstructured data (genetic or images) to be analyzed using statistical analysis and machine learning.

In a second phase, the federated data analysis carried out by GMV to launch the technological platform has been carried out using its aforementioned uTile PET solution, and its Antari tool to process and harmonize the data, guaranteeing its quality, observing the model standardization of European health data OMOP CDM.

As the research carried out at TARTAGLIA is demonstrating, AI techniques can improve information from digital images and create predictive tools to optimize diagnosis. On the one hand, work is being done to predict prostate cancer with clinical and genomic data from ImpMRI (multiparametric magnetic resonance imaging of the prostate, a radiological test that provides us with information about the shape and signal of the prostate in its different parts as well as the neighboring organs that surround it and their relationships). On the other, work is also being done to predict aggressiveness and improve the categorization of risk groups; to generate a federated repository of clinical, genomic, and imaging data from patients with interoperable prostate cancer, enabling regulatory-compliant data sharing for AI-related training and experimentation; avoiding invasive diagnostic procedures and advance personalized medicine; reducing the workload of radiologists through automated interpretation of images or reduce the number of unnecessary prostate biopsies; as well as over-detection of iPCa (Incidental prostate cancer or asymptomatic cancer discovered unexpectedly after microscopic examination of resected tissue) among other things.

TARTAGLIA is part of the Artificial Intelligence R&D Missions program of the Digital Spain 2025 agenda and the National Artificial Intelligence Strategy. It is financed by the European Union through Next Generation EU funds and has a budget of more than 7.5 million euros. GMV, as promoter and coordinator of the project, is a member of the consortium made up of 15 other entities that are working to create a federated network with artificial intelligence to speed up clinical and health research in our country.

*******

The TARTAGLIA project has received funding from the Ministry of Economic Affairs and Digital Transformation (File No. MIA.2021.M02.0005), corresponding to the funds of the Recovery, Resilience and Transformation Plan, The Entities that make up the project are: GMV; Accexible Impacto SL; Barcelona Supercomputing Center - Centro Nacional de Supercomputación; Dasel SL; Fundació ACE, Institut Català de Neurociències Aplicáis; Fundación para el Fomento de la Investigación Sanitaria y Biomédica de la Comunitat Valenciana; Fundación Canaria Instituto de Investigación Sanitaria de Canarias; Fundació ICT Salut Social; Fundación para la investigación del Hospital Universitario y Politécnico La Fe-Comunidad Valenciana; Fundación Rioja Salud; Entrepreneur Capital SL; Pixelabs SL; Agencia Gallega para la Gestión del Conocimiento en Salud (ACIS);Universidad Complutense de Madrid; PricewaterhouseCoopers Asesores de Negocios SL; Fundació Hospital Universitari Vall D' Hebron Institut de Recerca; Veratech for *Health, SL