Artificial intelligence for the diagnosis of lumbar pathologies
Low-back pain is considered the most common musculo-skeletal illness in the adult population, with a prevalence around 84%, meaning that 5 out of 6 persons will experience it at adult age. Despite its great morbidity, only the 15% of low-back pain patients are diagnosed with a specific lumbar pathology, which has lead to an increase of over a 100% of chronic low back pain (CLBPI). This pathology is the leading cause of disability worldwide.
The diagnostic evaluation of patients of low-back pain can be very challenging and requires complex clinical decision making. One of the main challenges faced by specialists is the discovery of pain-generative musculo-skeletal structures among all potential candidates. This is a key factor for handling of these patients and prevention of therapeutic errors.
Nuclear Magnetic Resonance (MR) has become the imaging method of choice for assessment of lumbar pathologies, almost replacing Computerised Axial Tomography (CAT) in routine studies of degenerative discopathy, infections, trauma, neoplasia and medular diseases.
This project aims at developing artifical intelligence tools, specifically deep learning-based tools, to process MR images for the diagnosis and grading of lumbar pathologies.
ATRYS HEALTH, a biomedical company focused on developing new tools for diagnosis and personalized treatments, relies on VICOMTECH for the development of these deep learning networks to assist the automatic diagnosis of patients.
This project has the support of CDTI for its funding through the *Ayudas Cervera para Centros Tecnológicos* programme.