Skip to main content

Table 5 The table reported the frequency for the top five reported Gleason scores with the remaining values grouped and reported as “Others”

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

No

Study dataset

Validation dataset

Gleason score

n = 

%

Gleason score

n = 

%

1

5 + 4 = 9

176

17.6

3 + 3 = 6

377

37.7

2

3 + 3 = 6

175

17.5

3 + 4 = 7

194

19.4

3

4 + 3 = 7

164

16.4

4 + 3 = 7

149

14.9

4

3 + 4 = 7

147

14.7

4 + 4 = 8

100

10.0

5

4 + 4 = 8

142

14.2

4 + 5 = 9

74

7.4

6

Others

196

19.6

Others

106

10.6

Total

 

1000

100

Total

1000

100

High-Risk GS ≥ 8

 

318

31.8

High-Risk GS ≥ 8

174

17.4

  1. Data is reported for this study as well as for the separate dataset
  2. GS: Gleason score