PEARL: Performance Evaluation of Artificial Intelligence Assisted Diabetic Retinopathy Grading in the Leuven University Hospital: Can Technology Improve the Resident?
Study Details
Study Description
Brief Summary
To evaluate the added value of the use of artificial intelligence in the diagnosis of referable diabetic retinopathy in a teaching hospital setting
| Condition or Disease | Intervention/Treatment | Phase |
|---|---|---|
|
Detailed Description
Diabetes mellitus is one of the major health challenges of our era. It is estimated that 642 million people will be diagnosed with this disease worldwide by 2040. Diabetes is a disease effecting the entire body and comes with many possible complications due to its' effect on the microvasculature. The most prevalent of these complications is diabetic retinopathy which is caused by both microvascular and neural damage.
According to studies in the United States by the Eye Diseases Prevalence Research group, about 40% of patients present with some degree of retinopathy. 8% of patients even have vision-threatening diabetic retinopathy.
Diabetic retinopathy is one of the main causes of blindness in our current society. However annual screening and timely referral for treatment can prevent this from occurring. The best illustration is the fact that since the implementation of a nationwide screening program, diabetes mellitus is no longer the leading cause of blindness in the UK.
Therefore, many countries have organized some sort of screening program. However, there are big organizational differences between countries. This can range from an annual dilated fundoscopy by an ophthalmologist (as is the case in Belgium) to non-mydriatic fundus photographs evaluated by a trained grader who is not a (para)medic.
Even with the most efficient screening pathway possible, the increase of patient numbers will become a problem since the human factor in the screening pathway (doctor, optometrist, trained grader,…) cannot increase its' capacity with the same speed. The current system will reach its limits at one point or another. Furthermore, it is known that a significant proportion of diabetes patients do not comply with the recommended annual screening. These problems will result in longer waiting lists, underdiagnosis because of overworked doctors, long waiting lists and possibly lack of high quality care.
Simply replacing the ophthalmologist by a trained grader probably won't solve all these problems. It will merely postpone them and will still remain costly and labor-intensive. The situation in countries which already use trained graders confirms these suspicions. Furthermore there is also room for improvement in the quality of care and the accuracy of diagnosis in these set ups.
In recent years, artificial intelligence, more specifically deep learning, has been postulated as a means to solve these problems. Even in the first studies, deep learning algorithms have already been shown to reach high sensitivity and specificity in detecting referable diabetic retinopathy. Further development of these algorithms and more thorough research have confirmed these findings. The use of AI has been studied in many medical fields, however diabetic retinopathy screening remains the pioneer, as is confirmed by the first-ever FDA authorization for an AI medical application being the diabetic retinopathy screening system IDx.
Current research mostly focusses of the performance of an artificial intelligence algorithm as an autonomous diagnostic tool without interaction with a human besides the acquisition of the images. Fear exists among medical professionals that artificial intelligence will start replacing them partially in the near future and make them obsolete on the long term. However, these novel technologies could also be used to aid the health professional in making the diagnosis in a more accurate way rather than replacing them.
Therefore, in the PEARL project, we wish to evaluate the use of an artificial intelligence algorithm as a diagnostic aid to improve the diagnostic accuracy of the physician rather than replacing the physician, certainly in a training context.
Study Design
Outcome Measures
Primary Outcome Measures
- does AI augment diagnostic performance of resident [4 months]
sensitivity and specificity
Secondary Outcome Measures
- does physical consultation augment diagnostic performance of AI [6 months]
sensitivity and specificity
Eligibility Criteria
Criteria
Inclusion Criteria:
-
- Diagnosis of diabetes mellitus
-
Age > 18 years old
-
Patient is capable of giving informed consent
-
Fluent in written and oral Dutch, or interpreter present
Exclusion Criteria:
-
- History of treatment for diabetic retinopathy or diabetic macular edema (laser or intravitreal injections)
-
Participant is contraindicated for imaging by fundus imaging systems used in the study
Contacts and Locations
Locations
No locations specified.Sponsors and Collaborators
- Universitaire Ziekenhuizen Leuven
Investigators
None specified.Study Documents (Full-Text)
None provided.More Information
Publications
- Abràmoff MD, Folk JC, Han DP, Walker JD, Williams DF, Russell SR, Massin P, Cochener B, Gain P, Tang L, Lamard M, Moga DC, Quellec G, Niemeijer M. Automated analysis of retinal images for detection of referable diabetic retinopathy. JAMA Ophthalmol. 2013 Mar;131(3):351-7. doi: 10.1001/jamaophthalmol.2013.1743.
- Farley TF, Mandava N, Prall FR, Carsky C. Accuracy of primary care clinicians in screening for diabetic retinopathy using single-image retinal photography. Ann Fam Med. 2008 Sep-Oct;6(5):428-34. doi: 10.1370/afm.857.
- Harding SP, Broadbent DM, Neoh C, White MC, Vora J. Sensitivity and specificity of photography and direct ophthalmoscopy in screening for sight threatening eye disease: the Liverpool Diabetic Eye Study. BMJ. 1995 Oct 28;311(7013):1131-5.
- Kempen JH, O'Colmain BJ, Leske MC, Haffner SM, Klein R, Moss SE, Taylor HR, Hamman RF; Eye Diseases Prevalence Research Group. The prevalence of diabetic retinopathy among adults in the United States. Arch Ophthalmol. 2004 Apr;122(4):552-63.
- Lee R, Wong TY, Sabanayagam C. Epidemiology of diabetic retinopathy, diabetic macular edema and related vision loss. Eye Vis (Lond). 2015 Sep 30;2:17. doi: 10.1186/s40662-015-0026-2. eCollection 2015. Review.
- Liew G, Michaelides M, Bunce C. A comparison of the causes of blindness certifications in England and Wales in working age adults (16-64 years), 1999-2000 with 2009-2010. BMJ Open. 2014 Feb 12;4(2):e004015. doi: 10.1136/bmjopen-2013-004015.
- Lin DY, Blumenkranz MS, Brothers RJ, Grosvenor DM. The sensitivity and specificity of single-field nonmydriatic monochromatic digital fundus photography with remote image interpretation for diabetic retinopathy screening: a comparison with ophthalmoscopy and standardized mydriatic color photography. Am J Ophthalmol. 2002 Aug;134(2):204-13.
- Ogurtsova K, da Rocha Fernandes JD, Huang Y, Linnenkamp U, Guariguata L, Cho NH, Cavan D, Shaw JE, Makaroff LE. IDF Diabetes Atlas: Global estimates for the prevalence of diabetes for 2015 and 2040. Diabetes Res Clin Pract. 2017 Jun;128:40-50. doi: 10.1016/j.diabres.2017.03.024. Epub 2017 Mar 31.
- Sayres R, Taly A, Rahimy E, Blumer K, Coz D, Hammel N, Krause J, Narayanaswamy A, Rastegar Z, Wu D, Xu S, Barb S, Joseph A, Shumski M, Smith J, Sood AB, Corrado GS, Peng L, Webster DR. Using a Deep Learning Algorithm and Integrated Gradients Explanation to Assist Grading for Diabetic Retinopathy. Ophthalmology. 2019 Apr;126(4):552-564. doi: 10.1016/j.ophtha.2018.11.016. Epub 2018 Dec 13.
- Sussman EJ, Tsiaras WG, Soper KA. Diagnosis of diabetic eye disease. JAMA. 1982 Jun 18;247(23):3231-4.
- S65943