Ophthalmology
Algorithm detected glaucoma in older adults’ eye photos
The algorithm detected 70.6% of affected eyes while correctly classifying 95% of eyes without glaucoma in an external population-based validation.The algorithm detected 70.6% of affected eyes while correctly classifying 95% of eyes without glaucoma in an external population-based validation.
Cohort · 9,837 peopleBiomarkers
Who was studied9,837 Rotterdam Study participants
What changedThe algorithm detected 70.6% of affected eyes while correctly classifying 95% of eyes without glaucoma
The catchIncomplete optic nerve head visibility was an important source of false positives