Skip to main content
. 2022 Jul 4;10:892658. doi: 10.3389/fpubh.2022.892658

Table 4.

Results obtained performing MIL and the barcode analysis on single-scale and multi-scale inputs.

MIL Barcode
5x 10x 20x 5x + 10x 5x + 10x + 20x 5x 10x 20x 5x + 10x 5x + 10x + 20x
Weighted F1 0.88 0.88 0.87 0.90 0.90 0.78 0.70 0.74 0.80 0.73
Macro F1 0.85 0.84 0.82 0.87 0.86 0.68 0.61 0.64 0.71 0.62
MCC 0.87 0.87 0.86 0.90 0.89 0.77 0.69 0.74 0.80 0.73
ACC 0.86 0.64 0.63 0.86 0.78 0.42 0.47 0.47 0.56 0.43
BLCA 0.83 0.83 0.79 0.86 0.83 0.70 0.48 0.69 0.68 0.65
BRCA 0.93 0.94 0.91 0.93 0.93 0.90 0.82 0.84 0.89 0.82
CESC 0.79 0.77 0.73 0.81 0.80 0.53 0.40 0.53 0.57 0.52
CHOL 0.60 0.80 0.55 0.67 0.83 0.22 0.20 0.00 0.25 0.00
COAD 0.93 0.90 0.85 0.96 0.90 0.82 0.75 0.81 0.86 0.77
DLBC 0.80 0.50 0.00 0.50 0.00 0.00 0.00 0.00 0.00 0.00
ESCA 0.68 0.69 0.73 0.74 0.74 0.50 0.35 0.51 0.55 0.53
GBM 0.85 0.90 0.89 0.90 0.91 0.80 0.76 0.80 0.77 0.80
HNSC 0.87 0.89 0.87 0.88 0.89 0.85 0.51 0.61 0.82 0.62
KICH 0.86 0.87 0.98 0.93 0.92 0.89 0.83 0.80 0.89 0.78
KIRC 0.94 0.95 0.93 0.96 0.95 0.94 0.89 0.89 0.93 0.90
KIRP 0.87 0.82 0.81 0.89 0.90 0.83 0.68 0.68 0.90 0.68
LGG 0.92 0.92 0.94 0.94 0.94 0.89 0.86 0.88 0.84 0.87
LIHC 0.93 0.93 0.91 0.92 0.94 0.82 0.81 0.87 0.87 0.86
LUAD 0.81 0.81 0.85 0.87 0.86 0.69 0.56 0.62 0.72 0.59
LUSC 0.77 0.77 0.79 0.82 0.82 0.60 0.54 0.61 0.66 0.60
MESO 0.67 0.67 0.57 0.73 0.83 0.11 0.11 0.29 0.09 0.20
OV 0.98 0.93 0.93 0.98 1.00 0.67 0.67 0.56 0.70 0.56
PAAD 0.90 0.86 0.89 0.93 0.89 0.72 0.55 0.59 0.75 0.42
PCPG 0.97 0.94 0.96 0.96 0.96 0.90 0.84 0.91 0.94 0.92
PRAD 0.94 0.96 0.96 0.97 0.97 0.96 0.97 0.91 0.97 0.88
SARC 0.84 0.80 0.86 0.85 0.85 0.73 0.63 0.70 0.84 0.70
SKCM 0.85 0.90 0.88 0.93 0.92 0.59 0.53 0.59 0.72 0.60
STAD 0.85 0.88 0.86 0.89 0.86 0.66 0.50 0.66 0.73 0.60
TGCT 0.95 0.93 0.95 0.97 0.97 0.87 0.79 0.75 0.89 0.79
THCA 0.97 0.98 0.97 0.96 0.98 0.93 0.95 0.94 0.94 0.94
THYM 0.93 0.96 0.92 0.96 0.96 0.78 0.78 0.78 0.77 0.79
UCEC 0.87 0.87 0.87 0.91 0.89 0.85 0.77 0.77 0.84 0.77
UCS 0.43 0.52 0.70 0.58 0.64 0.22 0.32 0.12 0.45 0.13

The results are obtained by training the model on the entire set of slides, after reserving a 20% split for testing. The value of the best performing magnification (or combination thereof) is highlighted in green. The top 3 rows show the aggregated classification results: weighted F1, macro F1 and MCC scores for the overall classification across all different tumor types on the test set.