Deep Learning for Analyzing Retina Changes in Gene Therapy Patients
Author Information
Author(s): German Pinedo-Diaz, Birgit Lorenz, Sandrine H. Künzel, Sarah Thiele, Susana Ortega-Cisneros, Eduardo Bayro Corrochano, Frank G. Holz, Alexander Effland
Primary Institution: Center for Research and Advanced Studies, Cinvestav, Zapopan, Mexico
Hypothesis
Can deep learning-based segmentation quantify outer retina changes in patients with RPE65 mutations undergoing gene therapy?
Conclusion
The study shows that automated analysis can reveal significant structural differences in the outer retina of patients undergoing gene therapy.
Supporting Evidence
- Significant differences in EZ biomarkers were found between RPE65-IRD patients and healthy controls.
- EZ thickness and granularity changes were significant in adults but not in pediatric patients.
- Automated segmentation allows for efficient analysis of retinal changes over time.
Takeaway
Researchers used computers to look at pictures of the eye to see how a special treatment helps kids and adults with a certain eye problem.
Methodology
The study used deep learning to analyze SD-OCT scans and quantify five biomarkers of the ellipsoid zone in patients with RPE65-IRD.
Potential Biases
Manual segmentation variability could introduce bias.
Limitations
The study had a small dataset and potential effects on OCT scans from high myopia.
Participant Demographics
Average age was 22.87 years, with 68 males and 48 females.
Statistical Information
P-Value
<0.0013
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website