Table 5 Classification Accuracy of the Multivariate Algorithm for Subgroups of Patients in the Combined Set (Training and Validation Set).

From: Highly Sensitive Marker Panel for Guidance in Lung Cancer Rapid Diagnostic Units

  

Cases correctly classifieda/Total cases

% Sn at 43% Sp

Lung Cancer

 

88/92

95.6

NSCLC

 

76/80

95

 

I

18/19

94.7

 

II

3/3

100

 

III

20/21

95.2

 

IV

35/37

94.6

 

ADC

41/44

93.2

 

SqCC

20/20

100

 

LCC

13/13

100

 

BAC

1/2

50

 

ND

1/1

100

SCLC

 

12/12

100

  

Cases correctly classified a /Total cases

% Sp at 95% Sn

Control

 

41/94

43.6

Healthy

 

18/44

40.9

Benign

 

23/50

46

 

RI

19/41

46.3

 

ILD

4/9

44.4

Accuracy in the Classification of Cancer and Controls regarding CT imaging

Absence of Nodules in CT scan

 

Cases correctly classified a /Total cases

% Sn at 43% Sp

Lung Cancer

 

7/7

100

NSCLC

 

6/6

100

 

I

1/1

100

 

II

2/2

100

 

III

1/1

100

 

IV

2/2

100

SCLC

Extended

1/1

100

Presence of Nodules in CT scan

 

Cases correctly classified a /Total cases

% Sp at 95% Sn

Control

 

1/3

33.3

Healthy

 

0/1

0

Benign

 

1/2

50

  1. Abbreviations: Sn = Sensitivity, Sp = Specificity, NSCLC = Non Small Cell Lung Cancer, ADC = Adenocarcinoma, SqCC = Squamous Cell Carcinoma, LCC = Large Cell Carcinoma, BAC = Bronchioloalveolar Carcinoma, ND = Not Differentiated Carcinoma, SCLC = Small Cell Lung Cancer, RI = Respiratory Infection, ILD = Interstitial Lung Disease.
  2. aCut-off p = 0.266 for a sensitivity of 97% and specificity of 43% in the training set.