Vicomtech develops a tool that allows predicting, through artificial intelligence, future diseases based on the current pathology


A group of Vicomtech researchers has developed, in collaboration with the Asuncion Klinika hospital in Toulouse, a tool that, through the use of artificial intelligence, is capable of predicting the future diseases that a person will suffer based on their current pathology.

To do this, the Asuncion Klinika electronic records database was used between January 2000 and August 2021, which has allowed researchers to access 526,597 diagnostic records coded with ICD9 and ICD10, covering a large part of the population of Tolosaldea. Specifically, the medical records of 71,846 patients have been used, completely anonymously.

Different studies have been able to show that the estimation of the progression of the disease from the current state of the patient is key for precision medicine. However, the number of tools for this analysis using artificial intelligence are limited or non-existent. At DISTRA, a dynamic and interactive tool based on artificial intelligence has been developed to be able to analyze the temporal patterns of disease progression.

The result of the project is an interactive and visual tool so that clinicians can explore possible diseases that can derive from a disease and be able to mitigate its possible complications or future diseases. Thus, in clinical practice, when a patient is diagnosed with a pathology, eg. atrial fibrillation, the clinician can review what other diseases he may have and thus be able to establish the guidelines that he considers appropriate. The app will be presented this week at the IX International Congress on Digital Health to be held between September 12 and 16, within the framework of the Miramar Palace Summer Courses


Parque Científico y Tecnológico de Gipuzkoa,
Paseo Mikeletegi 57,
20009 Donostia / San Sebastián (Spain)

+(34) 943 309 230

Zorrotzaurreko Erribera 2, Deusto,
48014 Bilbao (Spain)

close overlay

Behavioral advertising cookies are necessary to load this content

Accept behavioral advertising cookies