Table 3.
Error rates and pairwise comparisons of “all errors” data subset, by modality; “a” indicates results presented as model-based error rates per 1000 words ± standard error
| Modality |
Error Ratea (All Participants) |
Error Ratea (Attending Subgroup) |
|---|---|---|
| CT | 19.7 ± 4.5 | 15.3 ± 2.0 |
| MRI | 29.3 ± 6.5 | 21.1 ± 2.6 |
| US | 21.9 ± 5.0 | 16.6 ± 2.4 |
| XR | 26.4 ± 6.1 | 22.2 ± 3.3 |
| Comparison |
P-Value (All Participants) |
P-Value (Attending Subgroup) |
| CT vs MRI | < 0.0001 | 0.043 |
| CT vs US | 0.28 | 0.69 |
| CT vs XR | 0.009 | 0.043 |
| MRI vs US | 0.003 | 0.14 |
| MRI vs XR | 0.28 | 0.71 |
| US vs XR | 0.13 | 0.14 |