Assessment of fundus images
An example method for automating a quality assessment of digital fundus image can include: obtaining a digital fundus image file; analyzing a first quality of the digital fundus image file using a model to estimate an optimal time to capture a fundus image; and analyzing a second quality of the digital fundus image file using the model to estimate a disease state.
1. A method for automating a quality assessment of digital fundus image, the method comprising:
obtaining a digital fundus image file;
analyzing a first quality of the digital fundus image file using a model to estimate an optimal time to capture a fundus image, wherein the optimal time is estimated dependent on a size of a pupil;
capture a subsequent digital fundus image at the optimal time; and
analyzing a second quality of the subsequent digital fundus image file using the model to estimate a disease state.
2. The method of claim 1 , further comprising using a neural network to analyze the digital fundus image file.
3. The method of claim 2 , further comprising embedding the neural network in an imaging device used to obtain the digital fundus image file.
4. The method of claim 1 , further comprising analyzing an overall optical quality of the digital fundus image file as the first quality of the digital fundus image file.
5. The method of claim 1 , further comprising analyzing the second quality of the subsequent digital fundus image file by determining whether the digital fundus image file depicts a retina.
6. The method of claim 5 , further comprising detecting a fovea and an optic disc region in the subsequent digital fundus image file.
7. The method of claim 6 , further comprising one or more of:
determining an optical quality in the optic disc region;
determining a vessel count for a specified region of the retina;
determining a field position and classification of the subsequent digital fundus image file; and
providing feedback to allow for assessment of alternative disease states.
8. A method for automating a quality assessment of digital fundus image, the method comprising:
obtaining a digital fundus image file;
analyzing an optical quality of the digital fundus image file using a neural network to estimate an optimal time to capture a fundus image, wherein the optimal time is estimated dependent on a size of a pupil;
capture a subsequent digital fundus image at the optimal time; and
analyzing a second quality of the subsequent digital fundus image file using the model to estimate a disease state.
9. The method of claim 8 , further comprising embedding the neural network in an imaging device used to obtain the digital fundus image file.
10. The method of claim 8 , further comprising analyzing an overall optical quality of the digital fundus image file as the first quality of the digital fundus image file.
11. The method of claim 8 , further comprising analyzing the second quality of the subsequent digital fundus image file by determining whether the digital fundus image file depicts a retina.
12. The method of claim 11 , further comprising detecting a fovea and an optic disc region in the subsequent digital fundus image file.
13. The method of claim 12 , further comprising one or more of:
determining an optical quality in the optic disc region;
determining a vessel count for a specified region of the retina;
determining a field position and classification of the subsequent digital fundus image file; and
providing feedback to allow for assessment of alternative disease states.
14. A system for automating a quality assessment of digital fundus image, the system comprising:
a processor; and
memory encoding instructions which, when executed by the processor, cause the system to:
obtain a digital fundus image file;
analyze an optical quality of the digital fundus image file using a neural network to estimate an optimal time to capture a fundus image, wherein the optimal time is estimated dependent on a size of a pupil;
capture a subsequent digital fundus image at the optimal time; and
analyze a second quality of the subsequent digital fundus image file using the model to estimate a disease state.
15. The system of claim 14 , wherein the neural network is embedded in an imaging device used to obtain the digital fundus image file.
16. The system of claim 14 , wherein the memory encodes further instructions which, when executed by the processor, cause the system to analyze an overall optical quality of the digital fundus image file as the first quality of the digital fundus image file.
17. The system of claim 14 , wherein the memory encodes further instructions which, when executed by the processor, cause the system to analyze the second quality of the subsequent digital fundus image file by determining whether the digital fundus image file depicts a retina.
18. The system of claim 17 , wherein the memory encodes further instructions which, when executed by the processor, cause the system to detect a fovea and an optic disc region in the subsequent digital fundus image file.
19. The system of claim 18 , wherein the memory encodes further instructions which, when executed by the processor, cause the system to perform one or more of:
determine an optical quality in the optic disc region;
determine a vessel count for a specified region of the retina;
determine a field position and classification of the subsequent digital fundus image file; and
provide feedback to allow for assessment of alternative disease states.
20. The system of claim 14 , wherein the memory encodes further instructions which, when executed by the processor, cause the system to use a base line of one or more previous images to focus analysis on the digital fundus image file.