Introduction: Net Water Uptake (NWU) derived from non-contrast CT is a biomarker of early cerebral edema, calculated from density differences between infarcted and contralateral tissue. Manual NWU quantification is labor-intensive and prone to inter-operator variability, limiting clinical uptake and research scalability. We developed and internally validated a fully automated NWU evaluation pipeline. Methods: We analyzed 24-hour follow-up CT scans from the AcT (Alteplase compared to Tenecteplase) trial. Infarcts were automatically obtained by a novel segmentation framework based on a synchronous image-label diffusion probability model. The images and extracted infarcts were registered to the standard MNI152 space, allowing us to mirror the infarct onto the contralateral hemisphere symmetrically, regardless of size or tilt angle. Subsequently, the mirrored region was inverse transformed to return to its original space. Voxels outside the range of 20-80 Hounsfield Units (HU) were excluded to remove non-parenchymal tissue. Automated NWU was computed as the percentage difference in mean HU between infarct and mirrored contralateral regions. The agreement with manually determined NWU was evaluated using Pearson correlation, mean absolute error (MAE), and Bland-Altman analysis. Manual NWU was assessed in a convenience sample of 174 high-quality cases (well aligned raw CT scans in the axial plane with clear parenchymal infarct segmentations). In addition, the automated NWU algorithm was applied to 912 further cases with visible infarcts, with segmentations adjudicated as adequate or inadequate. Results: Among 174 cases, automated NWU showed excellent agreement with manual measurements (r = 0.96). Mean absolute error was 2.03% (95% CI: 1.67-2.39). Bland-Altman analysis demonstrated minimal bias (-0.42%) and satisfactory limits of agreement (-6.56% to +5.72%). Ninety percent of cases fell within ±5% of the manually determined value. Among the remaining 912 cases that were visually reviewed, NWU measurements were deemed adequate in 82.35%. Conclusions: Our automated mirrored segmentation pipeline enables accurate and reproducible NWU quantification from routine CT, matching expert manual measurement with minimal bias and needing infrequent correction. This approach supports large scale, standardized infarct edema measurement for research and potential clinical integration, removing a major bottleneck in NWU-based stroke imaging analysis.
Singh et al. (Thu,) studied this question.