Algorithm 2: Prostate cancer detection and segmentation framework

Data: DW-MRI Images Result: Segmented Prostate and Lesion Regions, IVIM Parameters, Normalized Dataset, Trained ML Model   1  Stage 1: DW-MRI Acquisition   2  Acquire DW-MRI images with varying b-values   3  Stage 2: Preprocessing   4  Motion correction, denoising, and image registration   5  Stage 3: Prostate and Lesion Segmentation   6  Utilize U-Net architecture with Attention U-Net and V-Net   7  Stage 4: Extract IVIM Parameters   8  Apply IVIM model fitting to obtain $S\left(0\right)$, f, D, and $D^{*}$   9  Stage 5: Dataset Normalization 10  Scale and center DW-MRI images and IVIM parameters 11  Stage 6: ML Learning and Optimization 12  Train ML algorithm on normalized data for prostate cancer detection 13  Stage 7: Model Evaluation 14  Evaluate the tuned model on a testing dataset 15  Use metrics like balanced accuracy, specificity, sensitivity, and Dice coefficient