Search papers, labs, and topics across Lattice.
This study investigates the impact of super-resolution (SR) techniques on the segmentation of white matter hyperintensities (WMH) in FLAIR MRI scans, specifically assessing whether SR erases existing small lesions or hallucinates non-existent ones. Using high-resolution FLAIR scans from 29 individuals, the authors simulated lower-resolution scans and applied various SR methods, including multi-contrast implicit neural representation (INR) and a single-contrast self-supervised model (ECLARE). The findings reveal that SR predominantly erases small real lesions rather than hallucinating absent ones, with ECLARE demonstrating the best performance in recovering small lesion signals across different slice thicknesses.
Super-resolution techniques can significantly erase small white matter lesions in MRI scans, challenging assumptions about their reliability in clinical settings.
White matter hyperintensities (WMH), bright regions on Fluid-attenuated Inversion Recovery (FLAIR) scans are associated with cerebrovascular pathology and neurodegeneration. FLAIR is usually acquired with thick slices in clinical settings, giving it poor through-plane resolution. Super-resolution (SR) is a widely used method for recovering an isotropic volume from an anisotropic scan. Yet whether applying it prior to WMH segmentation preserves lesion content remains unknown: a model may erase small real lesions or hallucinate absent ones. We used 1-mm isotropic high-resolution (HR) FLAIR scans from 29 individuals in the ADNI cohort, each manually segmented for WMH by an expert. Then, we degraded each to simulated 3 and 5 mm through-plane acquisitions. Multi-contrast implicit neural representation (INR), a single-contrast self-supervised model (ECLARE), and cubic interpolation were used to upsample them onto the HR grid. WMH segmentation from a simulated thick slice and the original HR FLAIR set the floor and ceiling, respectively, for the per-lesion analysis. Of four WMH segmentation methods (WMH-SynthSeg, segcsvd, MARS-WMH, TrUE-Net), we ran the analysis under the most sensitive one to small lesions on HR (MARS-WMH) with the evaluation metrics of detection sensitivity, erasure rate (HR-detected lesions lost after reconstruction), and hallucination rate (predicted components absent from both the manual and HR segmentation). The dominant effect of SR was erasure of small real lesions, not hallucination, and it increased with slice thickness, though every reconstruction still improved lesion detection over the raw thick slice. ECLARE recovered small lesion signal best at both thicknesses, while the INR was no better than cubic interpolation.
