Table 1 Classification results for Fibroblasts and PC3 Cell Collection.

From: Estimation of three-dimensional chromatin morphology for nuclear classification and characterisation

Features

Fibroblast cell line

PC3 cell line

AUC

F1 score

Precision

Recall

AUC

F1 score

Precision

Recall

3D SRP

\({\textbf{0.997}} \pm {\textbf{0.003}}\)

\({\textbf{0.99}} \pm {\textbf{0.02}}\)

\({\textbf{0.98}} \pm {\textbf{0.03}}\)

\({\textbf{0.998}} \pm {\textbf{0.002}}\)

\({\textbf{0.999}} \pm {\textbf{0.002}}\)

\({\textbf{0.98}} \pm {\textbf{0.02}}\)

\({\textbf{0.97}} \pm {\textbf{0.003}}\)

\({\textbf{0.98}} \pm {\textbf{0.03}}\)

Pseudo 3D SRP

\(0.94 \pm 0.04\)

\(0.94 \pm 0.04\)

\(0.97 \pm 0.04\)

\(0.91 \pm 0.06\)

\(0.98 \pm 0.01\)

\(0.85 \pm 0.20\)

\(0.75 \pm 0.32\)

\(0.94 \pm 0.08\)

3D LBP

\({\textbf{0.997}} \pm {\textbf{0.003}}\)

\({\textbf{0.99}} \pm {\textbf{0.02}}\)

\({\textbf{0.98}} \pm {\textbf{0.03}}\)

\({\textbf{0.996}} \pm {\textbf{0.001}}\)

\(0.994 \pm 0.004\)

\(0.97 \pm 0.03\)

\(0.97 \pm 0.04\)

\(0.97 \pm 0.04\)

Pseudo 3D LBP

\(0.88 \pm 0.13\)

\(0.79 \pm 0.15\)

\(0.77 \pm 0.18\)

\(0.82 \pm 0.10\)

\(0.97 \pm 0.02\)

\(0.90 \pm 0.06\)

\(0.84 \pm 0.11\)

\(0.97 \pm 0.04\)

3D SIFT

\(0.96 \pm 0.02\)

\(0.90 \pm 0.05\)

\(0.90 \pm 0.06\)

\(0.91 \pm 0.13\)

\(0.98 \pm 0.03\)

\(0.96 \pm 0.06\)

\(0.97 \pm 0.04\)

\(0.94 \pm 0.07\)

Pseudo 3D SIFT

\(0.85 \pm 0.04\)

\(0.74 \pm 0.08\)

\(0.69 \pm 0.10\)

\(0.79 \pm 0.08\)

\(0.93 \pm 0.02\)

\(0.80 \pm 0.13\)

\(0.84 \pm 0.21\)

\(0.80 \pm 0.05\)

3D RSurf

\(0.98 \pm 0.02\)

\(0.92 \pm 0.03\)

\(0.93 \pm 0.04\)

\(0.92 \pm 0.02\)

\(0.97 \pm 0.03\)

\(0.88 \pm 0.09\)

\(0.85 \pm 0.12\)

\(0.93 \pm 0.09\)

Pseudo 3D RSurf

\(0.94 \pm 0.02\)

\(0.86 \pm 0.02\)

\(0.87 \pm 0.05\)

\(0.85 \pm 0.05\)

\(0.97 \pm 0.03\)

\(0.89 \pm 0.05\)

\(0.89 \pm 0.01\)

\(0.91 \pm 0.08\)

FV-CNN

\(0.996 \pm 0.004\)

\(0.93 \pm 0.16\)

\(0.90 \pm 0.25\)

\(0.989 \pm 0.01\)

\(0.995 \pm 0.01\)

\(0.96 \pm 0.06\)

\(0.95 \pm 0.03\)

\(0.96 \pm 0.08\)

C-FC

\(0.72 \pm 0.08\)

\(0.62 \pm 0.08\)

\(0.61 \pm 0.11\)

\(0.65 \pm 0.08\)

\(0.76 \pm 0.05\)

\(0.72 \pm 0.04\)

\(0.70 \pm 0.06\)

\(0.76 \pm 0.08\)

  1. Metric values of pseudo 3D feature descriptors and features obtained from the penultimate fully-connected layer (C-FC) of the CNN model are also included for comparison.