. Author manuscript; available in PMC: 2019 Sep 20.

Published in final edited form as: Deep Learn Med Image Anal Multimodal Learn Clin Decis Support (2018). 2018 Sep 20;11045:83–91. doi: 10.1007/978-3-030-00889-5_10

Table 1.

F1 scores of the models obtained from querying iterations of different AL algorithms. The scores of intermediate querying iterations are based on grid samples, whereas the initial and final scores are reported based on full segmentation.

	Adolescents			Newborns
Initial	85.73 ± 3.91			79.93 ± 2.92
# Queries	Fisher (%)	entropy (%)	random (%)	Fisher (%)	entropy (%)	random (%)
100	87.11 ± 3.04	86.85 ± 3.29	82.61 ± 5.05	84.26 ± 2.86	83.33 ± 2.84	76.4 ± 6.22
500	90.9 ± 2.07	90.62 ± 2.16	85.28 ± 3.48	86.92 ± 2.37	86.47 ± 2.29	80.75 ± 2.96
1000	92.42 ± 1.76	92.57 ± 1.64	86.71 ± 2.88	88.11 ± 2.23	87.89 ± 2.12	82.12 ± 2.84
1500	93.57 ± 1.37	93.5 ± 1.39	87.78 ± 2.44	89.07 ± 2.02	88.82 ± 2	83.11 ± 2.85
Final	95.21 ± 0.94	95.15 ± 0.9	91 ± 1.48	90.24 ± 1.84	89.88 ± 1.72	86.92 ± 2.2