Pixyl.Neuro.MS
AI solution to support diagnosis and monitoring of multiple sclerosis (MS)
Pixyl.Neuro.MS is a deep learning tool assisting radiologists in MRI diagnosis and monitoring of multiple sclerosis (MS) patients. For each region of the brain, Pixyl.Neuro.MS counts the lesions present and calculates their overall volume, integrating the results fully automatically in the radiologist’s usual reading environment. The algorithm studies T2 FLAIR or T1 GADO sequences to identify hypersignals caused by white matter inflammation in MS.

Automatic detection, quantification and characterization of lesions in anatomical areas of interest
Pixyl.Neuro.MS measures the volume and lesion load for the entire brain and for every brain regions of interest (periventricular, juxtacortical, infratentorial, deep white matter) with regard to McDonald’s criteria (2017 revision)).
Longitudinal analysis of disease evolution if prior exam is present in the PACS
Pixyl.Neuro.MS automatically measures disease activity by segmenting and characterizing white matter evolution of lesions based on prior exams.
Automated production of a structured report and a segmented series (lesion outline) directly in the PACS
Pixyl.neuro provides the radiologist with a complete quantitative report on the lesions, their characteristics and evolution, as well as their visualization on the annotated brain images.
- multiple sclerosis
- T2 FLAIR
- T1 GADO
- lesion load
- MRI
- detection
- longitudinal analysis
Save time, reassure your diagnosis and streamline your workflow with Incepto
Testimonial
At the MAIL clinical group in Grenoble, we use Pixyl.Neuro.MS on FLAIR sequences for our multiple sclerosis patients.
The sensitivity of the algorithm is excellent, highlighting even the smallest lesions that would have been missed by the naked eye. Pixyl’s analysis is also valuable for high-lesion loads, enabling the detection of tiny variations of confluent lesions, thus improving their follow up.
The lesion-volume quantification offers a particularly useful rationalization for therapeutic decisions, thus improving patient care.
Dr Stéphane Cantin,
Neuroradiologist and president of the Mail clinical group
CHU Grenoble, France
Publications
- Benjamin Lambert and Maxime Louis and Senan Doyle and Florence Forbes and Michel Dojat and Alan Tucholka. “Leveraging 3D Information in Unsupervised Brain MRI Segmentation.” 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI) (pp. 187-190). 2021
http://arxiv.org/abs/2101.10674 - Pauline Roca, Arnaud Attyé, Lucie Colas, Alan Tucholka, Pascal Rubini, Stenzel Cackowski, Juliette Ding, Jean-François Budzik, Felix Renard, Senan Doyle, Emmanuel L. Barbier, Imad Bousaid, Romain Casey, Sandra Vukusic, Nathalie Lassau, Sébastien Verclytte, and François Cotton. “Artificial intelligence to predict clinical disability status scale score in patients with multiple sclerosis using FLAIR MRI” Diagnostic and Interventional Imaging 2020.
https://doi.org/10.1016/j.diii.2020.05.009 - Alexander Robert W. “Use of Software Analytics of Brain MRI (with & without contrast) As Objective Metric in Neurological Disorders and Degenerative Diseases” 2017, International Physical Medicine & Rehabilitation Journal
https://doi.org/10.15406/ipmrj.2017.02.00046