Skip to main content

Table 3 Performance of the text mining algorithm to automate the extraction of the Gleason score from narrative prostate biopsy narrative reports

From: Using text mining techniques to extract prostate cancer predictive information (Gleason score) from semi-structured narrative laboratory reports in the Gauteng province, South Africa

 

Manual coding

Exact Match: Yes

Exact Match: No

First algorithm output

Predicted

Exact Match: Yes

984

0

Exact Match: No

16

0

Precision = 1.00

  

Recall = 0.98

  

F-score = 0.99

  
 

Manual coding

Exact Match: Yes

Exact Match: No

Updated algorithm output

Predicted

Exact Match: Yes

1000

0

Exact Match: No

0

0

Precision = 1.00

  

Recall = 1.00

  

F-score = 1.00

  
 

Manual coding

Exact Match: Yes

Exact Match: No

Validation dataset output

Predicted

Exact Match: Yes

988

0

Exact Match: No

12

0

Precision = 1.00

  

Recall = 0.988

  

F-score = 0.99

  
  1. A contingency table was used to compare the manually coded and algorithm predicted values. We reported the precision, recall and F-score reported for the first and updated text mining algorithm output as well as for the validation dataset.
Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us