Table 6 Comparison of ResNet-CNN model accuracy with previous models.

From: DACBT: deep learning approach for classification of brain tumors using MRI data in IoT healthcare environment

Model

Acc (%)

Ref

Space complexity

Time cmplexity

ANN and KNN

97, 98

10

\(\mathscr {O}((nm + mk) + (n*d) )\)

\(\mathscr {O}((cwh) + (nd+Kn))\)

GA-CNN

94.2

12

\(\mathscr {O}(cwh + 1)f\)

\(\mathscr {O}(f*u*m)\)

SVM and KNN

85, 88

8

\(\mathscr {O}((n) + (n*d))\)

\(\mathscr {O}((n^2) + (nd+Kn))\)

CNN-TF

94.82

13

\(\mathscr {O}(cwh + 1)f\)

\(\mathscr {O}(f*u*m)\)

Proposed method ResNet-CNN

99.90

2022

\(\mathscr {O}(cwh + 1)f\)

\(\mathscr {O}(f*u*m)\)

  1. \(c=\) the number of convolutional channels, \(h=\) height of input, \(w=\) width of input, \(f=\) the convolutional kernel size, \(n=\) the number data instances, \(k=\) the number of output neurons, \(m=\) the number of input neurons and \(d =\) the dimension or feature of the input, \(K=\) number of nearest neighbors, \(u=c*w*h\).
Back to article page