MRI-Based Surgical Planning for Lumbar Spinal Stenosis
- 9k Downloads
The most common reason for spinal surgery in elderly patients is lumbar spinal stenosis (LSS). For LSS, treatment decisions based on clinical and radiological information as well as personal experience of the surgeon show large variance. Thus a standardized support system is of high value for a more objective and reproducible decision. In this work, we develop an automated algorithm to localize the stenosis causing the symptoms of the patient in magnetic resonance imaging (MRI). With 22 MRI features of each of five spinal levels of 321 patients, we show it is possible to predict the location of lesion triggering the symptoms. To support this hypothesis, we conduct an automated analysis of labeled and unlabeled MRI scans extracted from 788 patients. We confirm quantitatively the importance of radiological information and provide an algorithmic pipeline for working with raw MRI scans. Both code and data are provided for further research at www.spinalstenosis.ethz.ch.
KeywordsMachine learning Deep learning Lumbar spinal stenosis
This research was partially supported by the Max Planck ETH Center for Learning Systems, the SystemsX.ch project SignalX, the Baugarten Foundation, the Helmut Horten Foundation, the Pfizer-Foundation for geriatrics & research in geriatrics, the Symphasis Charitable Foundation, the OPO Foundation, NIH/NCI Cancer Center Support Grant P30 CA008748 and an Oxford - Google DeepMind scholarship.
- 2.Kreiner, S., Shaffer, W.O., Baisden, J., Gilbert, T., et al.: Evidence-based clinical guidelines for multidisciplinary spine care diagnosis and treatment of degenerative lumbar spinal stenosis. North Am. Spine Soc. (2014)Google Scholar
- 6.Steurer, J., Roner, S., Gnannt, R., Hodler, J.: Quantitative radiologic criteria for the diagnosis of lumbar spinal stenosis: a systematic literature review. BMC Musculoskelet Disord. 12, 175 (2011)Google Scholar
- 16.Abadi, M., Agarwal, A., Barham, P., Brevdo, E., et al.: Tensorflow: large-scale machine learning on heterogeneous distributed systems. arXiv preprint (2016)Google Scholar
- 17.Zeiler, M.D., Krishnan, D., Taylor, G.W., Fergus, R.: Deconvolutional networks. In: IEEE Conference on CVPR, pp. 2528–2535 (2010)Google Scholar
- 18.Zemel, R.S.: Autoencoders, minimum description length and Helmholtz free energy. NIPS (1994)Google Scholar