Abstract
Purpose: To describe and assess the clinical impact of ‘local contrast enhancement’ on oral fluorescein angiograms.
Methods: Oral fluorescein angiograms (OFA) were performed on 12 patients with a range of diabetic retinopathy. The digital images were processed using two enhancement techniques: histogram equalisation, a widely available method; and ‘local contrast enhancement’. Twelve control images and 24 enhanced images were randomised and subjectively graded for clarity between 1 and 100 on a visual analogue scale by two masked observers. A score of 50% was chosen as the cut-off for clinically useful images.
Results: The mean score for the unenhanced images was 38.8% (SD 19.4); the histogram equalisation performed better at 54.7% (SD 10.0) (p = 0.016) and ‘local contrast enhancement at 69.4% (SD 13.6) (p < 0.001). These results equate to a 14.7% improvement in comparison with histogram equalisation and a 30.3% improvement with the ‘local contrast enhancement’ technique. All the ‘local contrast enhancement’ images were clinically useful, compared with 58.3% of the histogram equalisation images (χ2 2.08, p = 0.2), and 33.3% of the control images (χ2 = 6.75, p = 0.01).
Conclusion: This pilot study has demonstrated the power of ‘local contrast enhancement’ techniques for low-contrast fluorescein angiograms. Using ‘local contrast enhancement’ oral fluorescein angiograms can give high-quality images and may allow safer patient investigation.
Methods: Oral fluorescein angiograms (OFA) were performed on 12 patients with a range of diabetic retinopathy. The digital images were processed using two enhancement techniques: histogram equalisation, a widely available method; and ‘local contrast enhancement’. Twelve control images and 24 enhanced images were randomised and subjectively graded for clarity between 1 and 100 on a visual analogue scale by two masked observers. A score of 50% was chosen as the cut-off for clinically useful images.
Results: The mean score for the unenhanced images was 38.8% (SD 19.4); the histogram equalisation performed better at 54.7% (SD 10.0) (p = 0.016) and ‘local contrast enhancement at 69.4% (SD 13.6) (p < 0.001). These results equate to a 14.7% improvement in comparison with histogram equalisation and a 30.3% improvement with the ‘local contrast enhancement’ technique. All the ‘local contrast enhancement’ images were clinically useful, compared with 58.3% of the histogram equalisation images (χ2 2.08, p = 0.2), and 33.3% of the control images (χ2 = 6.75, p = 0.01).
Conclusion: This pilot study has demonstrated the power of ‘local contrast enhancement’ techniques for low-contrast fluorescein angiograms. Using ‘local contrast enhancement’ oral fluorescein angiograms can give high-quality images and may allow safer patient investigation.
| Original language | English |
|---|---|
| Pages (from-to) | 318-318 |
| Journal | Eye |
| Volume | 14 |
| Issue number | 3 |
| DOIs | |
| Publication status | Published - 1 May 2000 |
