Table 2 Overall results with statistical analysis on an independent TMA testing set and a TCGA testing set in (a) classification of abnormal p53 (mutation-type) and normal p53 (wild type), and (b) TP53 mutation prediction in EC

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

Task

Method

Acc.

Sens.

Spec.

MeanSS

AUROC (μ ± σ)

Fisher’s Test p-value

(a) Classification of abnormal and normal p53 (TMA)

TR-MAMIL

0.78

0.70

0.86

0.78

0.78±0.04

<0.001***

 

ClassicMIL31

0.71

0.73

0.69

0.71

0.73 ± 0.01

<0.001***

 

CLAM26

0.75

0.68

0.81

0.74

0.77 ± 0.02

<0.001***

 

Wang et al.32

0.66

0.97

0.38

0.68

0.75 ± 0.02

<0.001***

 

Improved_InceptionV3_MS33

0.70

0.84

0.58

0.71

0.79 ± 0.03

0.496

 

TOAD27

0.68

0.89

0.50

0.70

0.76 ± 0.04

<0.001***

 

TransMIL28

0.72

0.84

0.62

0.73

0.77 ± 0.04

<0.001***

 

MRAN29

0.73

0.61

0.83

0.72

0.74 ± 0.08

<0.001***

(b) TP53 mutation prediction

TR-MAMIL

0.70

0.80

0.65

0.73

0.68 ± 0.08

<0.001***

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