Algorithm 1 TN Classification Algorithm
1: function LoadDatasets
2:       Dataset ← DDTI	▹ 134 images: 14 benign, 62 malignant
3:       return Dataset
4: end function
5: function Preprocess(Dataset)
6:       Benign ← GetBenignImages(Dataset)	▹ 14 images
7:       Malignant ← GetMalignantImages(Dataset)	▹ 62 images
8:       New_Benign ← SMOTE (Benign, target = 28)	▹ Oversample benign
9:       Balanced_Dataset ← Combine(New_Benign, Malignant[0:28])
10:      Augmented_Dataset ← Augment(Balanced_Dataset, Techniques = {Brightness, Flip, Rotate, Resise_to_512 × 512})
11:      return Augmented_Dataset	▹ Target: 2048 images
12: end function
13: function ExtractFeatures(Augmented_Dataset)
14:      Features ← []
15:      for each image in Augmented_Dataset do
16:            Bandelet_Features ← ApplyBandeletTransform(image)
17:            Features ← Add(Bandelet_Features)
18:      end for
19:      return Features
20: end function
21: function Classify(Augmented_Dataset, Features)
22:      Train_Data ← Take80Percent(Augmented_Dataset)	▹ 1638 images
23:      Val_Data ← Take20Percent(Augmented_Dataset)	▹ 410 images
24:      Models ← {VGG16}	▹ Simplified list
25:      for each model in Models do
26:            LoadPretrained(model)
27:            FineTune(model, Train_Data, Features)
28:            Predictions ← Test(model, Val_Data)
29:            Save(Predictions)
30:      end for
31:      return Predictions
32: end function
33: function Evaluate(Predictions, Val_Data)
34:      for each model in Predictions do
35:            Accuracy ← CalculateAccuracy(Predictions, Val_Data)
36:            Sensitivity ← CalculateSensitivity(Predictions, Val_Data)
37:            Display(model, Accuracy, Sensitivity)
38:      end for
39: end function
40: function Main
41:      Dataset ← LoadDatasets()
42:      Augmented_Dataset ← Preprocess(Dataset)
43:      Features ← ExtractFeatures(Augmented_Dataset)
44:      Predictions ← Classify(Augmented_Dataset, Features)
45:      Evaluate(Predictions, Augmented_Dataset)
46: end function