Skip to main content
. 2020 Feb 24;6(2):6. doi: 10.3390/jimaging6020006

Table 2.

Performance evaluation of the proposed algorithm and the compared methods. Size represents the number of scans in the dataset. FPI: False Positive per Image; FPE: False Positive per Exam.

Method Year Database Size Sensitivity FPI FPE
Dolejsi [36] 2009 TIME-LIDC-ANODE 38 89.60 12.03 -
Golosio [31] 2009 LIDC 484 71.00 - 4
Messay [29] 2010 LIDC 84 82.66 - 3
Tan [30] 2011 LIDC 399 87.50 - 4
Stelmo [21] 2012 LIDC 29 85.93 0.001 0.14
Teramoto [54] 2013 LIDC 84 80.00 - 4.2
Bergtholdt [56] 2016 LIDC-IDRI 243 85.90 - 2.5
Wu [55] 2017 LIDC-IDRI 60 79.23 - -
Froz [53] 2017 LIDC-IDRI 833 91.86 - -
Saien [11] 2018 LIDC/LIDC-IDRI 70 83.98 0.02 -
Ours 2019 LIDC 75 93.75 0.13 0.22