Abstract
Purpose :
To investigate the effect of an AI-assisted portable slit lamp (iSpector) and basic ophthalmology training on cataract detection, referral, and surgery rate in rural areas.
Methods :
63 village doctors were randomly assigned to the AI-assisted group (provides iSpector and training) and control group (provides training). Doctors were followed for one year before intervention as a baseline and one year after to collect the number of cataract patients detected, referred, and underwent surgery. Multivariable Poisson regression model was used to compare the difference in cataract detection, referral, and surgery rate between the two groups, adjusted for primary doctors’ baseline characteristics. We further conducted subgroup analysis to estimate the change in cataract detection, referral, and surgery rate within each group.
Results :
Compared to the control group, the detection, referral, and surgery rate of cataracts among the AI-assisted group was comparable (adjusted IRR=1.19, p=0.2), 1.7 times higher (adjusted IRR=2.7, p<0.001), and 4.9 times higher (adjusted IRR=5.85, p<0.001) separately. Providing iSpector and training increased the detection, referral, and surgery rate of cataracts by 2.0 times (IRR=3.099, p<0.001), 2.3 times (IRR=3.25, p<0.001), and 2.6 times (IRR=2.62, p=0.003) separately. Training alone elevated the detection rate of cataracts (IRR=1.99, p<0.001) but did not change the referral (IRR=0.85 p=0.5) and surgery (IRR=2.00, p=0.2) rate of cataracts.
Conclusions :
iSpector could help village doctors detect and refer cataract patients appropriately, thus increasing the probability that patients receive cataract surgery. Training helps doctors detect more cataract patients but could not help the patients receive an appropriate referral and surgery.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.