Contribution of optical character recognition (OCR) error correction and negated assertions on the performance of the system
| Yes | Other | |||||
| PPV | Sensitivity | F-measure | PPV | Sensitivity | F-measure | |
| Leiomyomas | ||||||
| Baseline | 95.86% | 98.25% | 97.04% | 97.97% | 95.24% | 96.59% |
| +Negated assertions | 98.10%* | 97.95% | 98.03%* | 97.71% | 97.87%* | 97.79%* |
| +OCR correction | 98.11%* | 98.39% | 98.25%* | 98.19% | 97.87%* | 98.03%* |
| Endometriosis | ||||||
| Baseline | 88.98% | 100.00% | 94.17% | 100.00% | 98.91% | 99.45% |
| +Negated assertions | 91.15% | 98.10% | 94.50% | 99.83% | 99.16% | 99.49% |
| +OCR correction | 91.15% | 98.10% | 94.50% | 99.83% | 99.16% | 99.49% |
| Adenomyosis | ||||||
| Baseline | 93.59% | 95.87% | 94.72% | 97.06% | 95.40% | 96.22% |
| +Negated assertions | 98.27%* | 95.87% | 97.06%* | 97.15%* | 98.82%* | 97.98%* |
| +OCR correction | 98.28%* | 96.62% | 97.45%* | 97.66%* | 98.82%* | 98.23%* |
Baseline configuration refers to the exact match of search terms.
*Performance difference against baseline is very significant at alpha = 0.01.
PPV, positive predictive value.