Table 1 Performance comparison of FarSight, NeuroGPS, HeY’s method, GFT and the proposed algorithm in different datasets

Data 1

2p-fMOST

150 × 150 × 150

788

764/0.92/

805/0.85/

767/0.78/

620/0.70/

490/0.60/

2 × 2 × 2 μm³

0.95/0.93^a

0.83/0.84

0.80/0.79

0.89/0.78

0.97/0.74

Data 2

2p-fMOST

100 × 100 × 100

288

280/0.92/

268/0.75/

265/0.84/

246/0.77/

231/0.74/

2 × 2 × 2 μm³

0.95/0.93

0.81/0.78

0.92/0.88

0.90/0.83

0.92/0.82

Data 3

2p-fMOST

100 × 100 × 100

25

24/0.96/

24/0.88/

24/0.96/

24/0.96/

24/0.96/

2 × 2 × 2 μm³

1.00/0.98

0.92/0.90

1.00/0.98

1.00/0.98

1.00/0.98

Data 4

2p-fMOST

100 × 100 × 100

164

170/0.94/

179/0.76/

172/0.88/

165/0.82/

144/0.73/

2 × 2 × 2 μm³

0.91/0.92

0.70/0.73

0.84/0.86

0.81/0.81

0.83/0.77

Data 5

2p-fMOST

100 × 100 × 100

148

147/0.96/

173/0.86/

146/0.94/

147/0.84/

135/0.86/

2 × 2 × 2 μm³

0.97/0.96

0.74/0.80

0.95/0.95

0.85/0.85

0.94/0.90

Data 6

SIM^b

100 × 100 × 100

496

486/0.95/

528/0.82/

407/0.82/

399/0.79/

452/0.89/

1 × 1 × 2 μm³

0.97/0.96

0.77/0.80

0.99/0.90

0.99/0.88

0.98/0.93

Data 7

SIM

100 × 100 × 100

93

89/0.96/

97/0.88/

84/0.90/

87/0.90/

83/0.88/

1 × 1 × 2 μm³

1.00/0.98

0.85/0.86

1.00/0.95

0.97/0.93

0.99/0.93

Data 8

Nissl staining

100 × 100 × 100

695

616/0.85/

680/0.86/

551/0.77/

502/0.66/

501/0.71/

1 × 1 × 1 μm³

0.95/0.90

0.84/0.85

0.97/0.86

0.91/0.77

0.98/0.82

Data 9

Nissl staining

100 × 100 × 50

287

266/0.90/

263/0.79/

216/0.74/

197/0.67/

187/0.64/

1 × 1 × 1 μm³

0.97/0.93

0.86/0.82

0.98/0.84

0.97/0.79

0.99/0.78

/recall

0.93 ± 0.04

0.83 ± 0.05

0.85 ± 0.08

0.79 ± 0.10

0.78 ± 0.12

/precision

0.96 ± 0.03

0.81 ± 0.07

0.94 ± 0.07

0.92 ± 0.07

0.96 ± 0.05

/F ₁

0.94 ± 0.03

0.82 ± 0.05

0.89 ± 0.06

0.85 ± 0.07

0.85 ± 0.08

/recall

0.001^**

0.021^ns

0.005^**

0.008^**

/precision

0.000^**

0.893^ns

0.209^ns

0.789^ns

/F ₁

0.000^**

0.051^ns

0.010^**

0.025^*