作者: Gabriele Abbati , Stefan Bauer , Sebastian Winklhofer , Peter J. Schüffler , Ulrike Held
DOI: 10.1007/978-3-319-66179-7_14
关键词: Magnetic resonance imaging 、 Radiology 、 Surgical planning 、 Lesion 、 Spinal surgery 、 Computer science 、 Lumbar spinal stenosis 、 Stenosis
摘要: The most common reason for spinal surgery in elderly patients is lumbar stenosis (LSS). For LSS, treatment decisions based on clinical and radiological information as well personal experience of the surgeon show large variance. Thus a standardized support system high value more objective reproducible decision. In this work, we develop an automated algorithm to localize causing symptoms patient magnetic resonance imaging (MRI). With 22 MRI features each five levels 321 patients, it possible predict location lesion triggering symptoms. To hypothesis, conduct analysis labeled unlabeled scans extracted from 788 patients. We confirm quantitatively importance provide algorithmic pipeline working with raw scans. Both code data are provided further research at www.spinalstenosis.ethz.ch.