Table 2 Quantitative evaluation of DBAHNet segmentation performance across various medical treatments for the main datasets. The results include the Dice similarity coefficient (DSC) and 95th percentile of the Hausdorff distance (HD95) for both the cortical (C) and the trabecular (T) compartments. \(N_{\text {sub}}\) represents the total number of subsets, with each scan divided into 10 subsets along the z-axis.

From: A robust deep learning approach for segmenting cortical and trabecular bone from 3D high resolution µCT scans of mouse bone

Doses

Avg DSC

DSC C

DSC T

Avg HD95

HD95 C

HD95 T

Dataset 114

 20 (µg/kg/day) (\(N_{\text {sub}}\) = 60)

0.9036

0.9460

0.8612

0.0737

0.0430

0.1045

 20 (µg/kg/day) + ML (\(N_{\text {sub}}\) = 60)

0.9137

0.9478

0.8795

0.0748

0.0469

0.1027

 40 (µg/kg/day) (\(N_{\text {sub}}\) = 80)

0.9440

0.9799

0.9082

0.0531

0.0251

0.0810

 80 (µg/kg/day) (\(N_{\text {sub}}\) = 100)

0.9498

0.9907

0.9088

0.0424

0.0107

0.0742

Mean performance

0.9278

0.9661

0.8894

0.0610

0.0314

0.0906

Dataset 215

 0.15 (µg/kg/day) (\(N_{\text {sub}}\) = 50)

0.9737

0.9854

0.9619

0.0561

0.0229

0.0893

 0.15 (µg/kg/day) + ML (\(N_{\text {sub}}\) = 50)

0.9723

0.9843

0.9603

0.0600

0.0227

0.0975

 1.5 (µg/kg/day) (\(N_{\text {sub}}\) = 50)

0.9770

0.9879

0.9661

0.0458

0.0222

0.0694

 1.5 (µg/kg/day) + ML (\(N_{\text {sub}}\) = 50)

0.9712

0.9851

0.9573

0.0718

0.0241

0.1196

 15 (µg/kg/day) (\(N_{\text {sub}}\) = 50)

0.9786

0.9884

0.9687

0.0369

0.0195

0.0543

 15 (µg/kg/day) + ML (\(N_{\text {sub}}\) = 50)

0.9697

0.9833

0.9561

0.0718

0.0325

0.1113

 150 (µg/kg/day) (\(N_{\text {sub}}\) = 10)

0.9797

0.9898

0.9696

0.0153

0.0133

0.0172

 150 (µg/kg/day) + ML (\(N_{\text {sub}}\) = 10)

0.9682

0.9821

0.9543

0.0735

0.0219

0.1250

Mean performance

0.9738

0.9860

0.9616

0.0540

0.0224

0.0856

Dataset 316

 4 N (\(N_{\text {sub}}\) = 40)

0.9733

0.9845

0.9622

0.0408

0.0176

0.0639

 8 N (\(N_{\text {sub}}\) = 46)

0.9723

0.9840

0.9607

0.0501

0.0221

0.0781

Mean performance

0.9728

0.9842

0.9614

0.0455

0.0199

0.0710

All datasets

Mean performance

0.9631

0.9811

0.9345

0.0548

0.0248

0.0846