Breast Mass Detection in Mammography Based on Image Template Matching and CNN

. 2021 Apr 18;21(8):2855. doi: 10.3390/s21082855

Algorithm 1. PSO for breast mass bounding box refinement.

Input:
Number of particles, maximum iteration number, classification model of breast mass based
on CNN: Prediction_Model, rough bounding box of breast mass

X (x, y, w, h)

Output:

b e s t_{X}

for

i = 1

to number of particles do

X_{i} = X

V_{i} = [0, 0, 0, 0]

p b e s t_{i} = P r e d i c t i o n_M o d e l (X_{i})

g b e s t < p b e s t_{i}

then

g b e s t = p b e s t_{i}

b e s t_{x} = X

for

p = 1

to Maximum iteration number do
for

i = 1

to Number of particles do
1. Update

V_{i}

V_{i} = V_{i}^{(- 1)} + c_{p} \times r a n d () \times p b e s t_{i} - X_{i}^{(- 1)}) + c_{g} \times r a n d () \times (g b e s t - X_{i}^{(- 1)})

2. Update

X_{i}

X_{i} = X_{i}^{(- 1)} + V_{i}

3. Get fitness from

X_{i}

P r e d i c t i o n_M o d e l

f i t n e s s = P r e d i c t i o n_M o d e l (X_{i})

4. Update

p b e s t_{i}

p b e s t_{i} < f i t n e s s

then

p b e s t_{i} = f i t n e s s

5. Update

t b e s t

and

b e s t_{x}

for

j = 1

to number of particles do
if

g b e s t < p b e s t_{j}

then

g b e s t = p b e s t_{j}

b e s t_{x} = X_{i}