Table 2 Comparison in the performance of our proposed disease probability-enhanced (DPE) follow-up chest X-ray radiology report summary generation against the state-of-the-arts.

From: Disease probability-enhanced follow-up chest X-ray radiology report summary generation

Method

Natural language processing metric\(^*\)

Clinical metrics

B1

B2

B3

B4

M

R

C

Acc5

Acc14

F1-5

F1-14

Rad-F1\(^\%\)

\(\text {MCCFormers}\)18\(^\dagger\)

21.4

19.0

17.0

15.3

31.9

34.0

0.0

–

–

–

  

\(\text {IDCPCL}\)20\(^{\dagger }\)

61.4

54.1

47.4

41.4

30.3

58.2

0.703

–

–

–

  

EKAID7

62.8

55.3

49.1

43.4

33.9

57.7

1.027

\(\text {68.2}^{\ddagger }\)

\(\text {81.0}^{\ddagger }\)

\(\underline{\text {50.1}^{\ddagger }}\)

\(\underline{\text {46.6}^{\ddagger }}\)

\(\underline{\text {31.6}^{\ddagger }}\)

DPE-all

65.6

58.8

53.1

48.0

39.4

62.9

1.680

69.7

81.9

49.9

45.9

28.1

DPE-all14

66.2

59.8

54.3

49.2

40.0

64.9

1.762

71.3

82.8

49.6

46.4

27.1

DPE-all26

69.0

62.6

57.2

52.4

42.2

69.2

2.118

76.5

85.2

58.5

55.9

36.7

DPE-light

63.9

57.4

52.0

46.8

38.2

60.6

1.565

68.0

81.1

42.9

40.9

25.6

  1. \(\dagger\) Results were retrieved from ref.7. \(\ddagger\) Results were obtained using the source codes of EKAID7. \(^*\) B, M, R and C stand for BLEU, METEOR ROUGE-L and CIDEr respectively. \(^\%\) Radgraph-F1. The best and second best performances were demonstrated in bold or underlined respectively