Skip to main content

Table 2 Performance Comparison between Artificial Neural Networks based on fused, combined and single types of data.

From: Integration and fusion of standard automated perimetry and optical coherence tomography data for improved automated glaucoma diagnostics

 

F-SAP data

(AROC:0.958)

F-OCT data

(AROC:0.978)

SAP & OCT Data

(AROC:0.968)

F-SAP & F-OCT data

(AROC:0.978)

SAP data

(AROC: 0.945)

0.502

0.047

0.147

0.047

OCT data

(AROC: 0.970)

0.431

0.576

0.879

0.562

  1. Significance (p) values of Area under Receiver Operating Characteristic (AROC) curves were calculated by DeLongs non-parametric method.
  2. SAP data: Standard Automated Perimetry data, based on Pattern Deviation (PD) probability scores
  3. F-SAP data: Fused SAP data, based on weighted transformation of PD probability scores with OCT-derived probability scores
  4. OCT data: Age - and refraction corrected Optical Coherence Tomography A-scan data, optimized by principal component analysis (PCA)
  5. F-OCT data: Fused OCT data, based on weighted transformation of A-scan measurements
  6. with PD probability scores and optimized by PCA
  7. Bold indicates statistical significance (i.e. p < 0.05)
Back to article page

BMC Ophthalmology

ISSN: 1471-2415

Contact us