Advanced Muscle Segmentation

Screenshot 2021-02-15 at 17.59.19.png

Magnetic resonance imaging (MRI) is the best imaging

modality to quantitatively assess muscles since it provides

better soft-tissue contrast than other imaging techniques and

additionally it does expose the patient to ionising radiations.

Automated segmentation of skeletal muscle from MR

images is a challenging task since individual muscles share

similar intensity/contrast values.

Our team has developed, starting from a multi-atlas segmentation through a machine learning method, a solution that provides accurate and automated segmentation of gluteal muscles from pelvic MRI to extract biomarkers for muscle health, fat content and volume, in the general population, in athletes and in patients.

Our Publications

Gender similarities and differences in skeletal muscle and body composition: an MRI study of recreational cyclists

Gender similarities and differences in skeletal muscle and body composition: an MRI study of recreational cyclists

Mid-life cyclists preserve muscle mass and composition: a 3D MRI study

Mid-life cyclists preserve muscle mass and composition: a 3D MRI study

Automated multi-atlas segmentation of gluteus maximus from Dixon and T1-weighted magnetic resonance images

Automated multi-atlas segmentation of gluteus maximus from Dixon and T1-weighted magnetic resonance images

Automated measurement of fat infiltration in the hip abductors from Dixon magnetic resonance imaging

Automated measurement of fat infiltration in the hip abductors from Dixon magnetic resonance imaging

Reference values for volume, fat content and shape of the hip abductor muscles in healthy individuals from Dixon MRI

Reference values for volume, fat content and shape of the hip abductor muscles in healthy individuals from Dixon MRI

Intramuscular fat in gluteus maximus for different levels of physical activity

Intramuscular fat in gluteus maximus for different levels of physical activity