BSC researchers develop a prototype of automatic classification of tumor mutations with artificial intelligence

06 November 2019

This use case has been made with Natural Language Processing and is part of the Plan for the Advancement of Language Technology.

The Natural Language Processing group of the CASE department at the Barcelona Supercomputing Center (BSC) has developed a prototype of application for automatic classification of genetic mutations for tumor processes.

The classification is carried out after training automatically the prototype with clinical articles obtained from the Memorial Sloan Kettering Cancer Center (MSKCC). The function of this prototype is to help pathologists to distinguish which of the many genetic mutations that their patients have can contribute to the growth of tumors, a task that is now carried out manually and takes a long time, as usually the genetic sequencing of a cancer tumor presents thousands of mutations and the vast majority are neutral for the disease.

The prototype is trained by applying Natural Language Processing techniques, a field of artificial intelligence that is responsible for programming computers to process and analyze large amounts of information in natural language.

This use case has been carried out within the framework of the Plan for the Advancement of Language Technology (Plan TL) of the Secretary of State for Digital Advancement (SEAD), and has served as the starting point to define basic natural language processing flows, which will be implemented in the TL Plan platform.

This prototype demonstrates that a user who is not familiar with language technologies can easily integrate natural language processing components to perform complex tasks.