Table 1.
Summary of brain imaging techniques and their application in SOZ identification for adults and children.
| Imaging technique | Invasive/non-invasive | Number of imaging sessions | Spatial resolution | Temporal resolution | Brain area | SOZ ID in adults with DRE | SOZ ID in children with DRE |
|---|---|---|---|---|---|---|---|
| PET | Invasive | 1 | 6 mm | 5 mins | Surface | Identified in 66.7% of subjects (Mayoral et al., 2019) | Only detects EN in 6.6% of patients and not SOZ (Bansal et al., 2016) |
| Ictal SPECT | Invasive | 2, needs difference between ictal and inter-ictal SPECT (Desai et al., 2013) | 6 mm | 15 mins | Surface | Detects EN and not SOZ (Kaiboriboon et al., 2002; von Oertzen, 2018) | Detects EN and not SOZ (Van Paesschen et al., 2007) |
| MEG | Non-invasive | 1 | 10–20 mm | <10 ms | Cortex | 75% accurate (Foley et al., 2021) in eight patients, 60% with ML in a large cohort (Nissen et al., 2018) | Poor concordance with SOZ. Identified onset zone is >20 mm distance from ground truth SOZ (Ntolkeras et al., 2022) |
| iEEG (gold standard) | Invasive | Long-term monitoring (14 days) | 6–8 cm | <5 ms | Cortex | 92.3% accurate (Nagahama et al., 2018) | >90% accurate (Nagahama et al., 2018) |
| fMRI | Non-invasive | 1 | 3 mm | 2–5s | Central brain | Sensitivity 83% and specificity 67% (Chen et al., 2017) | 89% accurate SOZ identification (Boerwinkle et al., 2019) |
PET, positron emission tomography; SPECT, single-photon emission computerized tomography; MEG, magnetoencephalography; iEEG, intracranial electroencephalography; fMRI, functional magnetic resonance imaging.