Automatic fundus image quality assessment on a continuous scale.

Fundus photography is commonly used for screening, diagnosis, and monitoring of various diseases affecting the eye. In addition, it has shown promise in the diagnosis of brain diseases and evaluation of cardiovascular risk factors. Good image quality is important if diagnosis is to be accurate and timely. Here, we propose a method that automatically grades image quality on a continuous scale which is more flexible than binary quality classification. The method utilizes random forest regression models trained on image features discovered automatically by combining basic image filters using simulated annealing as well as features extracted with the discrete Fourier transform. The method was developed and tested on images from two different fundus camera models. The quality of those images was rated on a continuous scale from 0.0 to 1.0 by five experts. In addition, the method was tested on DRIMDB, a publicly available dataset with binary quality ratings. On the DRIMDB dataset the method achieves an accuracy of 0.981, sensitivity of 0.993 and specificity of 0.958 which is consistent with the state of the art. When evaluating image quality on a continuous scale the method outperforms human raters.

Oximetry in glaucoma: correlation of metabolic change with structural and functional damage.

PURPOSE: To determine whether retinal vessel oxygen saturation in patients with glaucoma is associated with structural optic disc and retinal nerve fibre layer (RNFL) changes and visual field (VF) defects. METHODS: Fifty-nine patients with confirmed glaucoma were recruited at University Hospitals Leuven. Retinal oxygen saturation in patients with glaucoma was measured with a noninvasive spectrophotometric retinal oximeter (Oxymap ehf, Reykjavik, Iceland). VF and Heidelberg retinal tomographies (HRTs) were performed on the same day. Statistical analysis was performed using Student's t-test and Pearson's or Spearman correlation coefficient. RESULTS: The mean oxygen saturation in venules was higher in patients with severe VF defects compared to those patients with mild VF defects (69 ± 3% versus 65 ± 6% respectively; p = 0.0003; n = 59). Accordingly, the arteriovenous (AV) difference in oxygen saturation was lower in patients with worse VF compared to those with better VF (29 ± 3% versus 33 ± 6% respectively; p = 0.002). The oxygen saturation in venules correlated with the VF mean defects (r = -0.42; p = 0.001; n = 59) as well as with the structural HRT parameters rim area and RNFL thickness (r = -0.39; p = 0.008 and r = -0.26; p = 0.05 respectively; n = 53). The AV difference decreased significantly as the VF defect worsened (r = 0.38; p = 0.003), as the rim area diminished (r = 0.29; p = 0.03) and as the RNFL thickness decreased (r = 0.27; p = 0.05). No correlation was found between the oxygen saturation in retinal arterioles and either of these parameters. CONCLUSION: Severe glaucomatous damage is associated with increased oxygen saturation in retinal venules and decreased AV difference in oxygen saturation. These data suggest that in eyes with severe glaucomatous damage, reduced retinal oxygen consumption is consistent with tissue loss.