Background Amyotrophic lateral sclerosis (ALS) is the most common adult-onset motor neuron disease. ALS is currently regarded as a multisystem disease, where pathological processes clearly extend beyond the motor system. Several disease mechanisms are involved, including protein aggregation, increases in reactive oxygen species levels, calcium disturbances, mitochondrial degeneration and neuroinflammatory processes. Similar to other neurodegenerative disorders magnetic resonance imaging markers, indicative of excess iron levels, have been reported in the brains of patients with ALS and ALS animal models. Here we employed a statistical analysis of susceptibility-weighted imaging (SWI) to assess patients with ALS, to answer the question whether SWI is able to detect typical ALS white matter disturbances. Materials & Methods We analysed the SWI of white matter in a cohort of 30 patients with ALS and 30 healthy age-matched controls. Scans were obtained on a 1.5 Tesla Siemens Sonata scanner. SWI images were calculated based on magnitude and phase data. Homodyne filtering of the complex-valued GRE data was performed to correct raw phase images for phase wraps and for background phase contributions. To this end, the complex-valued GRE data are divided in image domain with a complex-valued low-pass filtered copy of the GRE data and the arc tangent is taken to yield corrected phase images. The low-pass filtering was carried out by slice-wise multiplication of the GRE data with a 2D Hanning window in the Fourier domain using a width of 25% of the data matrix size. Since a left-handed MRI system was used during the study, the corrected phase images, , were converted according to Haacke et al. (2004) to create weighted phase images, , with enhanced paramagnetic structures such as iron-laden tissue and venous vessels. The sensitivity toward paramagnetic structures of the magnitude images was then amplified by multiplying the magnitude with the weighted phase images yielding the so called susceptibility weighted images. Preprocessing of SWI was then done using SPM8 (Wellcome Trust Centre for Neuroimaging, UCL, London, UK) and the VBM8 toolbox with MATLAB R2009b (TheMathworks, Natick, USA) as mathematical framework. After coregistration of the SWI images on their corresponding T1 images, these SWI maps were normalized into MNI space by using the warps from DARTEL normalization of the T1 images. Images were finally smoothed with a 4 mm FWHM Gaussian kernel. Group comparison of ALS patients and healthy controls for the whole brain was performed in SPM8 using the model ’compare populations: one scan/subject (ANCOVA)’ with age and sex as nuisance variable. Results Signal alterations were found in the corpus callosum; along the corticospinal tract (subcortical motor cortex, posterior limb of the internal capsule and brainstem levels) and in the subgyral regions of frontal, parietal, temporal, occipital and limbic lobes. Conclusion SWI of the white matter is capable of assessing iron and myelin disturbances. The SWI patterns observed in this study indicate that widespread iron disturbances occur in patients with ALS and correlate with disease severity. Moreover SWI detects the hallmark white matter changes in ALS, suggesting that SWI could serve as biomarker.