Jasper van der Graaf, M.Sc.
Jasper van der Graaf completed a Bachelor’s degree in Human Movement Sciences at the University of Groningen in 2015. Following his graduation he finished the pre-master for Technical Medicine at the University of Twente. On September 2016, he enrolled in the Master of Technical Medicine, specializing in Medical Imaging and Interventions after which he received his Master’s Degree in December 2019. In May 2020 he started as a PhD candidate at both the Orthopaedic Research Lab (ORL) and the Diagnostic Image Analysis Group under supervision of Miranda van Hooff. His research is focused on AI-based MRI analysis for treatment decision support in patients with chronic degenerative low back pain.
Low back pain (LBP) is worldwide responsible for more years lived with disability than any other health condition. In the Netherlands, approximately 44% of the population experiences at least one episode of LBP in their lifetime, with one in five reporting persistent back pain lasting longer than three months (chronic low back pain, CLBP). CLBP often results in substantial limitations in functional activities and is responsible for high healthcare and socioeconomic costs. In the vast majority of patients with CLBP (85-90%), the etiology is unknown and for medical specialists it is challenging to identify patients who would benefit from surgical or non-surgical interventions.
Diagnosis and treatment decisions are often based on magnetic resonance imaging (MRI) of the lumbar spine. The use of this modality in patients with LBP has drastically increased in the last decades. Automatic image analysis has the potential to counterbalance the associated increase in workload for radiologists and spinal surgeons, and to improve the diagnostic value of MRI by enabling more objective and quantitative image interpretation. However, to be of value for evaluating complex multifactorial disorders like LBP, this analysis needs to comprehend multiple anatomical elements of the spine, such as the vertebrae and the intervertebral discs (IVDs).
A robust automatic algorithm for segmentation of these structures is therefore an essential component. In our first project we aim to create a convolutional neural network that automatically segments vertebrae, inter-vertebral discs and the spinal canal on MRI scans. Anonymized clinical MRI scan from the RadboudUMC are manually annotated which will be used to train the network. Currently the dataset consists of MR scans from RadboudUMC, Rijnstate, Jeroen Bosch Hospital and the Sint Maartenskliniek. Eventually we also aim to organize a segmentation challenge using this dataset (which will be made publicly available).
To identify which MR image features are related to CLBP a narrative review was written. In this review an overview of all possible image features is presented in five different categories: discogenic, neurogenic, osseous, facetogenic and paraspinal. Per category an overview of all relevant literature is given. The result of this review will be a list of image features that have a high predictive value for CLBP which eventually will be used in the decision support tool. The goal is to automatically extract all these features from the MR images.