Table 1 Statistics of the cohorts before processing - Stanford US dataset and EUH MRI dataset
From: Transfer language space with similar domain adaptation: a case study with hepatocellular carcinoma
Stanford US dataset | EUH MRI dataset | |
|---|---|---|
Cohort level | ||
Number of unique words | 17194 | 19828 |
Common words in two domains | 1790 | |
Report level | ||
Average number of words (+/- std) | 167 (+/- 39) | 197 (+/-47) |
Average number of sentences (+/- std) | 27 (+/-7) | 32 ((+/-8) |
Number of unique words in templated reports | 2774 | |
Number of unique words in reports without template | 7930 | |
Average number of words (+/- std) describing | ||
liver related finding in templated reports | 36 (+/- 25) | 109 (+/- 63) |
Average number of words (+/- std) describing | ||
liver related finding in reports without template | 47 (+/- 27) | 104 (+/- 52) |