DICOM-Based Quantification of MRI Artifact from Auditory Implants: a Technical Note Comparing Bone-Conduction- and Cochlear Implants
摘要
MRI in patients with auditory implants is limited by implant-related signal void, geometric distortion, and intravoxel dephasing. Artifact extent is often described qualitatively. We aimed to quantify MRI artifact burden from a novel transcutaneous bone-conduction implant (SentioTM) and to compare it with cochlear implant controls under routine, label-compliant 1.5 T clinical conditions.
MethodsWe retrospectively analyzed MRI examinations of patients with a SentioTM implant (n = 7) and cochlear implant controls (n = 5). All scans were performed at 1.5 T, the manufacturer-approved field strength for the SentioTM Ti implant, on the same scanner using routine transverse turbo spin-echo sequences (T2-weighted and T1-weighted; 5‑mm slice thickness). Artifacts were quantified slice-by-slice with a custom Python-based DICOM pipeline. The artifact-to-brain ratio (ABR, %) and maximum void area (cm2) were computed; ABRmax (peak slice) served as patient-level summary. Groups were compared using the Mann-Whitney U test and Cliff’s delta.
ResultsSentioTM implants showed smaller and more confined artifacts. Median ABRmax was 1.1% [IQR 0.2] with a median maximum void area of 1.7 cm2 [IQR 0.4]. Cochlear implant controls showed larger artifacts (ABRmax 2.6% [IQR 1.0]; median maximum void area 3.8 cm2 [IQR 2.0]; p = 0.007–0.01; δ = 0.81–0.89). Supratentorial diagnostic image quality outside the immediate implant region was preserved in all T2w/T1w SentioTM examinations and partially preserved in 3/5 cochlear implant examinations.
ConclusionQuantitative artifact assessment using ABRmax is feasible in routine 1.5 T MRI. Under identical protocol conditions, the SentioTM implant produced a smaller and more localized MRI artifact burden than the cochlear implant controls examined here. Routine turbo spin-echo imaging remained diagnostically useful outside the implant region.