Dataset Information

Automatic Determination of Endothelial Cell Density From Donor Cornea Endothelial Cell Images.

ABSTRACT:

Purpose

To determine endothelial cell density (ECD) from real-world donor cornea endothelial cell (EC) images using a self-supervised deep learning segmentation model.

Methods

Two eye banks (Eversight, VisionGift) provided 15,138 single, unique EC images from 8169 donors along with their demographics, tissue characteristics, and ECD. This dataset was utilized for self-supervised training and deep learning inference. The Cornea Image Analysis Reading Center (CIARC) provided a second dataset of 174 donor EC images based on image and tissue quality. These images were used to train a supervised deep learning cell border segmentation model. Evaluation between manual and automated determination of ECD was restricted to the 1939 test EC images with at least 100 cells counted by both methods.

Results

The ECD measurements from both methods were in excellent agreement with rc of 0.77 (95% confidence interval [CI], 0.75-0.79; P < 0.001) and bias of 123 cells/mm2 (95% CI, 114-131; P < 0.001); 81% of the automated ECD values were within 10% of the manual ECD values. When the analysis was further restricted to the cropped image, the rc was 0.88 (95% CI, 0.87-0.89; P < 0.001), bias was 46 cells/mm2 (95% CI, 39-53; P < 0.001), and 93% of the automated ECD values were within 10% of the manual ECD values.

Conclusions

Deep learning analysis provides accurate ECDs of donor images, potentially reducing analysis time and training requirements.

Translational relevance

The approach of this study, a robust methodology for automatically evaluating donor cornea EC images, could expand the quantitative determination of endothelial health beyond ECD.

SUBMITTER: Benetz BAM

PROVIDER: S-EPMC11346145 | biostudies-literature | 2024 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Automatic Determination of Endothelial Cell Density From Donor Cornea Endothelial Cell Images.

Benetz Beth Ann M BAM Shivade Ved S VS Joseph Naomi M NM Romig Nathan J NJ McCormick John C JC Chen Jiawei J Titus Michael S MS Sawant Onkar B OB Clover Jameson M JM Yoganathan Nathan N Menegay Harry J HJ O'Brien Robert C RC Wilson David L DL Lass Jonathan H JH

Translational vision science & technology 20240801 8

<h4>Purpose</h4>To determine endothelial cell density (ECD) from real-world donor cornea endothelial cell (EC) images using a self-supervised deep learning segmentation model.<h4>Methods</h4>Two eye banks (Eversight, VisionGift) provided 15,138 single, unique EC images from 8169 donors along with their demographics, tissue characteristics, and ECD. This dataset was utilized for self-supervised training and deep learning inference. The Cornea Image Analysis Reading Center (CIARC) provided a secon ...[more]

PMID: 39177992

Similar Datasets

Project description:ImportanceDetermining factors associated with endothelial cell loss after Descemet stripping automated endothelial keratoplasty (DSAEK) could improve long-term graft survival.ObjectiveTo evaluate the associations of donor, recipient, and operative factors with endothelial cell density (ECD) 3 years after DSAEK in the Cornea Preservation Time Study.Design, setting, and participantsThis cohort study was a secondary analysis of data collected in a multicenter, double-masked, randomized clinical trial. Forty US clinical sites with 70 surgeons participated, with donor corneas provided by 23 US eye banks. Individuals undergoing DSAEK for Fuchs dystrophy or pseudophakic/aphakic corneal edema were included.InterventionsThe DSAEK procedure, with random assignment of a donor cornea with a preservation time of 0 to 7 days or 8 to 14 days.Main outcomes and measuresEndothelial cell density at 3 years as determined by a reading center from eye bank and clinical specular or confocal central endothelial images.ResultsThe study included 1090 participants (median age, 70 years) with 1330 affected eyes (240 bilateral cases [22.0%]), who underwent DSAEK for Fuchs dystrophy (1255 eyes [94.4%]) or pseudophakic/aphakic corneal edema (PACE) (75 eyes [5.6%]). Of these, 801 eyes (60.2%) belonged to women and 1207 (90.8%) to white individuals. A total of 749 participants (913 eyes; 164 [21.9%] bilateral cases) had functioning grafts with acceptable endothelial images preoperatively and at 3 years postoperatively and were included in this analysis. Factors associated with a lower ECD at 3 years (estimated effect with 99% CI) in the final multivariable model included donors with diabetes (-103 [-196 to -9] cells/mm2), lower screening ECD (-234 [-331 to -137] per 500 cells/mm2), recipient diagnosis of PACE (-257 [-483 to -31] in cells/mm2), and operative complications (-324 [-516 to -133] in cells/mm2). Endothelial cell loss (ECL) from a preoperative measurement to a 3-year postoperative measurement was 47% (99% CI, 42%-52%) for participants receiving tissue from donors with diabetes vs 43% (99% CI, 39%-48%) without diabetes; it was 53% (99% CI, 44%-62%) for participants diagnosed with PACE vs 44% (99% CI, 39%-49%) for those diagnosed with Fuchs dystrophy, and 55% (99% CI, 48%-63%) in participants who experienced operative complications vs 44% (99% CI, 39%-48%) in those who did not. No other donor, recipient, or operative factors were significantly associated with 3-year ECD.Conclusions and relevanceDonor diabetes, lower screening ECD, a PACE diagnosis in the recipient, and operative complications were associated with lower ECD at 3 years after DSAEK surgery and may be associated with long-term graft success. While causation cannot be inferred, further studies on the association of donor diabetes and PACE in recipients with lower 3-year ECD warrant further study.

Project description:PurposeCorneal endothelial cell density (ECD) gradually decreases after corneal transplantation by unknown biologic, biophysical, or immunologic mechanism. Our purpose was to assess the association between donor corneal endothelial cell (CEC) maturity in culture and postoperative endothelial cell loss (ECL) after successful corneal transplantation.DesignProspective cohort study.ParticipantsThis cohort study was conducted at Baptist Eye Institute, Kyoto, Japan, between October 2014 and October 2016. It included 68 patients with a 36-month follow-up period who had undergone successful Descemet stripping automated endothelial keratoplasty (DSAEK) or penetrating keratoplasty.MethodsHuman CECs (HCECs) from remaining peripheral donor corneas were cultured and evaluated for maturity by surface markers (CD166+, CD44-/dull, CD24-, and CD105-) using fluorescence-activated cell sorting. Postoperative ECD was assessed according to the mature-differentiated HCEC contents: high-maturity group: > 70%, middle-maturity group: 10% to 70%, low-maturity group: < 10%. The successful rate of ECD maintained at 1500 cells/mm2 at 36 months postoperative was analyzed using the log-rank test.Main outcome measuresEndothelial cell density and ECL at 36 months postoperative.ResultsThe 68 included patients (mean [standard deviation] age 68.1 [13.6] years, 47.1% women, 52.9% DSAEK). The high, middle, and low-maturity groups included 17, 32, and 19 eyes, respectively. At 36 months postoperative, the mean (standard deviation) ECD significantly decreased to 911 (388) cells/mm2 by 66% in the low-maturity group, compared with 1604 (436) by 40% and 1424 (613) cells/mm2 by 50% in the high and middle-maturity groups (P < 0.001 and P = 0.007, respectively) and the low-maturity group significantly failed to maintain ECD at 1500 cells/mm2 at 36 months postoperative (P < 0.001). Additional ECD analysis for patients who underwent DSAEK alone displayed a significant failure to maintain ECD at 1500 cells/mm2 at 36 months postoperative (P < 0.001).ConclusionsThe high content of mature-differentiated HCECs expressed in culture by the donor peripheral cornea was coincident with low ECL, suggesting that a high-maturity CEC content predicts long-term graft survival. Understanding the molecular mechanism for maintaining HCEC maturity could elucidate the mechanism of ECL after corneal transplantation and aid in developing effective interventions.Financial disclosuresProprietary or commercial disclosure may be found after the references.

Project description:ObjectiveTo determine whether the 10-year success rate of penetrating keratoplasty for corneal endothelial disorders is associated with donor age.DesignMulticenter, prospective, double-masked clinical trial.ParticipantsA total of 1090 participants undergoing penetrating keratoplasty at 80 sites for Fuchs' dystrophy (62%), pseudophakic/aphakic corneal edema (34%), or another corneal endothelial disorder (4%) and followed for up to 12 years.MethodsForty-three eye banks provided corneas from donors aged 12 to 75 years, using a randomized approach to assign donor corneas to study participants without respect to recipient factors. Surgery and postoperative care were performed according to the surgeons' usual routines.Main outcome measuresGraft failure defined as a regraft or, in the absence of a regraft, a cloudy cornea that was sufficiently opaque to compromise vision for 3 consecutive months.ResultsIn the primary analysis, the 10-year success rate was 77% for 707 corneas from donors aged 12 to 65 years compared with 71% for 383 donors aged 66 to 75 years (difference, +6%; 95% confidence interval, -1 to +12; P = 0.11). When analyzed as a continuous variable, higher donor age was associated with lower graft success beyond the first 5 years (P<0.001). Exploring this association further, we observed that the 10-year success rate was relatively constant for donors aged 34 to 71 years (75%). The success rate was higher for 80 donors aged 12 to 33 years (96%) and lower for 130 donors aged 72 to 75 years (62%). The relative decrease in the success rate with donor ages 72 to 75 years was not observed until after year 6.ConclusionsAlthough the primary analysis did not show a significant difference in 10-year success rates comparing donor ages 12 to 65 years and 66 to 75 years, there was evidence of a donor age effect at the extremes of the age range. Because we observed a fairly constant 10-year success rate for donors aged 34 to 71 years, which account for approximately 75% of corneas in the United States available for transplant, the Cornea Donor Study results indicate that donor age is not an important factor in most penetrating keratoplasties for endothelial disease.

Project description:ObjectiveTo examine the long-term effect of donor diabetes history on graft failure and endothelial cell density (ECD) after penetrating keratoplasty (PK) in the Cornea Donor Study.DesignMulticenter, prospective, double-masked, controlled clinical trial.ParticipantsOne thousand ninety subjects undergoing PK for a moderate risk condition, principally Fuchs' dystrophy or pseudophakic or aphakic corneal edema, were enrolled by 105 surgeons from 80 clinical sites in the United States.MethodsCorneas from donors 12 to 75 years of age were assigned by 43 eye banks to participants without respect to recipient factors. Donor and recipient diabetes status was determined from existing medical records. Images of the central endothelium were obtained before surgery (baseline) and at intervals for 10 years after surgery and were analyzed by a central image analysis reading center to determine ECD.Main outcome measuresTime to graft failure (regraft or cloudy cornea for 3 consecutive months) and ECD.ResultsThere was no statistically significant association of donor diabetes history with 10-year graft failure, baseline ECD, 10-year ECD, or ECD values longitudinally over time in unadjusted analyses, nor after adjusting for donor age and other significant covariates. The 10-year graft failure rate was 23% in the 199 patients receiving a cornea from a donor with diabetes versus 26% in the 891 patients receiving a cornea from a donor without diabetes (95% confidence interval for the difference, -10% to 6%; unadjusted P=0.60). Baseline ECD (P=0.71), 10-year ECD (P>0.99), and changes in ECD over 10 years (P=0.86) were similar comparing donor groups with and without diabetes.ConclusionsThe study results do not suggest an association between donor diabetes and PK outcome. However, the assessment of donor diabetes was imprecise and based on historical data only. The increasing frequency of diabetes in the aging population in the United States affects the donor pool. Thus, the impact of donor diabetes on long-term endothelial health after PK or endothelial keratoplasty, or both, warrants further study with more precise measures of diabetes and its complications.

Dataset Information

Automatic Determination of Endothelial Cell Density From Donor Cornea Endothelial Cell Images.

Purpose

Methods

Results

Conclusions

Translational relevance

Publications

Automatic Determination of Endothelial Cell Density From Donor Cornea Endothelial Cell Images.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets