Table 4 FFPE and Frozen tissue sample results with statistical analysis on the TCGA testing sets in (a) classification of aggressive and non-aggressive EC, (b) TMB prediction of the aggressive EC, and (c) TMB prediction of the non-aggressive EC

From: Deep learning for endometrial cancer subtyping and predicting tumor mutational burden from histopathological slides

Task

Tissue

Method

Acc.

Sens.

Spec.

MeanSS

AUROC (μ ± σ)

Fisher’s test p-value

(a) Classification of aggressive and non-aggressive EC

FFPE

TR-MAMIL

0.88

0.96

0.71

0.84

0.86 ± 0.01

<0.001***

 

Frozen

 

0.89

0.93

0.84

0.88

0.87 ± 0.04

<0.001***

(b) TMB prediction for the aggressive subtype

FFPE

TR-MAMIL

0.68

0.83

0.54

0.69

0.61 ± 0.04

<0.001***

 

Frozen

 

0.72

0.71

0.73

0.72

0.69 ± 0.06

<0.001***

(c) TMB prediction for the non-aggressive subtype

FFPE

TR-MAMIL

0.75

1.00

0.57

0.79

0.65 ± 0.15

0.81

 

Frozen

 

0.66

0.72

0.63

0.68

0.57 ± 0.08

0.022*

  1. MeanSS: Mean of sensitivity and specificity; *p < 0.05; ***p < 0.001.