AI-Generated Summaries for Eye Disease

Overseen By Prashant Tailor, MD

Age: 18+

Sex: Any

Trial Phase: Academic

Sponsor: University of California, Los Angeles

Disqualifiers: Cognitive impairments, Prisoners, others

No Placebo Group All trial participants will receive the active study treatment (no placebo)

Compensation: Varies

Number of Visits: 2

460 Participants Needed

Trial Summary

Will I have to stop taking my current medications?

The trial information does not specify whether you need to stop taking your current medications.

What data supports the effectiveness of the treatment AI-Generated Plain Language Summaries for eye disease?

The study on AI-assisted diabetic retinopathy screening in Australia shows that AI systems can be accurate and well-accepted by patients and clinicians, suggesting that AI-generated summaries could also be effective in helping patients understand eye disease information. ^{1 2 3 4 5}

How is the AI-Generated Plain Language Summaries treatment different from other treatments for eye disease?

The AI-Generated Plain Language Summaries treatment is unique because it uses artificial intelligence to create easy-to-understand summaries of complex medical information, helping patients better understand their eye conditions and treatments. This approach differs from traditional treatments that focus on medical interventions, as it aims to improve patient comprehension and engagement. ^{6 7 8 9 10}

What is the purpose of this trial?

This clinical trial is testing whether plain language summaries made by artificial intelligence help people understand their eye doctor's notes better. Adults receiving eye care at the Jules Stein Eye Institute will get either the usual medical notes or a note with the addition of an AI-generated summary that explains the information in simple, everyday words. Participants will then answer a short survey and receive a follow-up call to share how clear the information was, how well they understood their diagnosis and treatment, and whether they feel more confident about their care. The goal is to find out if these plain language summaries can make it easier for people to understand their eye care and improve communication between patients and health care providers.

Research Team

Prashant Tailor, MD

Principal Investigator

University of California, Los Angeles

Eligibility Criteria

This trial is for English-speaking adults who are receiving eye care at the Jules Stein Eye Institute. It's designed to help those who may struggle with understanding medical jargon, possibly including patients with aphasia or other communication difficulties.

Inclusion Criteria

Receiving ophthalmology care at the Jules Stein Eye Institute

Able to provide informed consent

Exclusion Criteria

Prisoners or wards of the state

I am unable to understand and agree to the study details on my own.

I do not have conditions like dementia that affect my understanding.

Timeline

Screening

Participants are screened for eligibility to participate in the trial

1-2 weeks

Intervention

Participants receive either standard ophthalmology notes or notes with an AI-generated plain language summary

1 day

1 visit (in-person)

Immediate Post-Visit Assessment

Participants complete a survey to assess comprehension and satisfaction immediately after their clinic visit

1 day

Follow-up

Participants receive a follow-up telephone interview to assess retention of information and provide additional feedback

1 week

1 call (telephone)

Treatment Details

Interventions

• AI-Generated Plain Language Summaries

Trial Overview The study is testing AI-generated summaries that translate ophthalmology notes into plain language. Participants will be randomly given either standard medical notes or ones with an AI summary, and their comprehension and confidence in care will be assessed through surveys and follow-up calls.

Participant Groups

Group I: SON + AI-Generated Plain Language Summaries Experimental Treatment 1 Intervention

Participants in this arm receive the standard ophthalmology notes plus an AI-generated plain language summary, reviewed for accuracy before distribution. They will complete the same surveys to assess whether the additional summary improves their understanding and satisfaction compared to the control group.

Group II: Standard Ophthalmology Notes (SON) Only Active Control 1 Intervention

Participants in this arm receive the standard ophthalmology notes typically provided after their clinic visit, with no additional plain language summary. They will complete surveys that measure their comprehension and satisfaction with the visit notes.

Who Is Running the Clinical Trial?

University of California, Los Angeles

Lead Sponsor

Trials

1,594

Recruited

10,430,000+

Findings from Research

The Total Visual Acuity Extraction Algorithm (TOVA) successfully automates the extraction of visual acuity data from unstructured clinical notes, achieving a 95% concordance rate with manual extraction in a validation study of 644 notes.

TOVA demonstrated high reliability and accuracy, with interrater reliability statistics indicating strong agreement and a Pearson correlation coefficient of 0.983, making it a valuable tool for efficiently processing large volumes of ophthalmology data.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Validation of the Total Visual Acuity Extraction Algorithm (TOVA) for Automated Extraction of Visual Acuity Data From Free Text, Unstructured Clinical Records. Baughman, DM., Su, GL., Tsui, I., et al.[2022]

The study analyzed 14,537 cataract surgeries over 32 months using electronic health record (EHR) data, demonstrating that EHR can effectively evaluate surgical outcomes for both resident and attending surgeons.

Results showed that 74% of resident surgeries had better postoperative visual acuity, with no significant difference in outcomes between residents and attending surgeons, indicating that residents can achieve comparable surgical results despite starting with poorer baseline visual acuity.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Assessing Resident Cataract Surgical Outcomes Using Electronic Health Record Data. Xiao, G., Srikumaran, D., Sikder, S., et al.[2023]

A survey of 466 patients with wet age-related macular degeneration revealed that the most important factors influencing their treatment preferences were clarity of vision, treatment effects on symptoms, and quality of vision, highlighting the significance of visual outcomes in their care.

The study demonstrated that patients actively engage in their treatment management, as indicated by their high levels of knowledge and confidence, which can guide future clinical development to better align with patient needs.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Patient Preferences in the Management of Wet Age-Related Macular Degeneration: A Conjoint Analysis. Skelly, A., Taylor, N., Fasser, C., et al.[2023]

The AI-assisted diabetic retinopathy screening model demonstrated high diagnostic accuracy with a sensitivity of 96.9% and specificity of 87.7%, based on an analysis of 203 participants from various clinics.

Patient and clinician satisfaction was very high, with 93.7% of participants expressing satisfaction with the service, indicating that the AI system is not only effective but also well-accepted in real-world healthcare settings.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Real-world artificial intelligence-based opportunistic screening for diabetic retinopathy in endocrinology and indigenous healthcare settings in Australia. Scheetz, J., Koca, D., McGuinness, M., et al.[2021]

In a study of 1052 adult patients over 18 months, AI-based image analysis for diabetic retinopathy screening achieved 100% sensitivity and 89.2% specificity, indicating it is highly effective in identifying patients needing further evaluation.

The study found that 91.7% of patients had AI-gradable images, with the ability to improve image quality through reflex dilation for those with initially ungradable images, although older patients had lower image gradeability.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Diabetic Retinopathy Telemedicine Outcomes With Artificial Intelligence-Based Image Analysis, Reflex Dilation, and Image Overread. Mehra, AA., Softing, A., Guner, MK., et al.[2023]

A review of 26 websites on age-related macular degeneration (AMD) found that while many sites provide comprehensive information about AMD and emerging treatments, the overall quality of the information was disappointing, with average quality scores around 62%.

Most websites had high readability scores (SMOG ratings above 10), indicating that the information is difficult for the average person to understand, highlighting the need for physicians to guide patients to better resources.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Age-related macular degeneration: what do patients find on the internet? Rennie, CA., Hannan, S., Maycock, N., et al.[2021]

ChatGPT-4.0 demonstrated an 85% accuracy rate in diagnosing corneal diseases from case reports, significantly outperforming ChatGPT-3.5, which had a 60% accuracy rate.

When compared to human experts, ChatGPT-4.0 showed a high interobserver agreement of 85% with one specialist, indicating its potential for clinical use in diagnosing corneal conditions.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Performance of ChatGPT in Diagnosis of Corneal Eye Diseases. Delsoz, M., Madadi, Y., Munir, WM., et al.[2023]

The RetCAD AI software demonstrated high accuracy in identifying and grading age-related macular degeneration (AMD) and diabetic retinopathy (DR) from fundus photos, achieving a sensitivity of 97.3% and specificity of 73.3% for AMD, and 80.0% sensitivity and 90.1% specificity for DR at the patient level.

With a large sample size of 1245 eyes from 630 patients, the study shows that RetCAD can provide reliable automated grading, making it a cost-effective tool for screening in settings where expert evaluation is limited.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Simultaneous screening and classification of diabetic retinopathy and age-related macular degeneration based on fundus photos-a prospective analysis of the RetCAD system. Skevas, C., Weindler, H., Levering, M., et al.[2022]

Artificial intelligence (AI) has the potential to significantly enhance the diagnosis and treatment of retinal diseases, which are a major cause of blindness, by providing decision support for conditions like diabetic retinopathy and age-related macular degeneration.

Integrating AI into clinical practice could improve healthcare efficiency, facilitate quicker access to vision-saving treatments, and allow clinicians to focus more on patient interactions and personalized care plans.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Artificial intelligence in retinal disease: clinical application, challenges, and future directions. Daich Varela, M., Sen, S., De Guimaraes, TAC., et al.[2023]

The rise of big data in healthcare, driven by electronic health records and AI, has greatly enhanced the ability to analyze and understand diseases, particularly in ophthalmology, allowing for better diagnosis and treatment outcomes.

In ophthalmology, large datasets from clinical registries and studies are being utilized to improve understanding of corneal diseases and cataracts, although there are still barriers to fully realizing the potential of these resources in clinical practice.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Big data in corneal diseases and cataract: Current applications and future directions. Ting, DSJ., Deshmukh, R., Ting, DSW., et al.[2023]

References

1. United States pubmed.ncbi.nlm.nih.gov

Validation of the Total Visual Acuity Extraction Algorithm (TOVA) for Automated Extraction of Visual Acuity Data From Free Text, Unstructured Clinical Records. [2022]

2. Netherlands pubmed.ncbi.nlm.nih.gov

Assessing Resident Cataract Surgical Outcomes Using Electronic Health Record Data. [2023]

3. United States pubmed.ncbi.nlm.nih.gov

Patient Preferences in the Management of Wet Age-Related Macular Degeneration: A Conjoint Analysis. [2023]

4. England pubmed.ncbi.nlm.nih.gov

Real-world artificial intelligence-based opportunistic screening for diabetic retinopathy in endocrinology and indigenous healthcare settings in Australia. [2021]

5. United States pubmed.ncbi.nlm.nih.gov

Diabetic Retinopathy Telemedicine Outcomes With Artificial Intelligence-Based Image Analysis, Reflex Dilation, and Image Overread. [2023]

6. England pubmed.ncbi.nlm.nih.gov

Age-related macular degeneration: what do patients find on the internet? [2021]

7. United States pubmed.ncbi.nlm.nih.gov

Performance of ChatGPT in Diagnosis of Corneal Eye Diseases. [2023]

8. China pubmed.ncbi.nlm.nih.gov

Simultaneous screening and classification of diabetic retinopathy and age-related macular degeneration based on fundus photos-a prospective analysis of the RetCAD system. [2022]

9. Germany pubmed.ncbi.nlm.nih.gov

Artificial intelligence in retinal disease: clinical application, challenges, and future directions. [2023]

10. Switzerland pubmed.ncbi.nlm.nih.gov

Big data in corneal diseases and cataract: Current applications and future directions. [2023]