[Skip to Navigation]
Invited Commentary
December 23, 2020

A Machine Learning Model With Survival Statistics to Identify Predictors of Descemet Stripping Automated Endothelial Keratoplasty Graft Failure

Author Affiliations
  • 1Department of Ophthalmology and Visual Sciences, University of Illinois at Chicago, Chicago
JAMA Ophthalmol. 2021;139(2):198-199. doi:10.1001/jamaophthalmol.2020.5741

Descemet stripping automated endothelial keratoplasty (DSAEK) is a predominant surgical method of endothelial keratoplasty for treating corneal endothelial dysfunction.1 Prediction, variable selection, and determining factors associated with graft failure may lead to improved clinical decision-making guidelines and surgical outcomes of DSAEK. Survival data allow for time-to-event analysis, which can be defined as death, onset of disease, or the success or failure of procedures. The traditional statistical methods used to analyze survival data include Kaplan-Meier analysis for estimating the survival function, the log-rank test for 2-group comparisons, and the Cox proportional hazards regression model to determine associations between risk factors and the probability of survival.2 More recently, machine learning techniques have been used to expand our ability in including high-dimensional data with multiple features for predictions and variable selection. In this issue of JAMA Ophthalmology, O’Brien et al3 use data from the Cornea Preservation Time Study to apply a random survival forest (RSF), an ensemble tree method for analysis of right-censored survival data, as a method to select important variables that predict graft failure after DSAEK.

Add or change institution
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words