Retinal fundus photographs capture hemoglobin loss after blood donation
Mitani, A.; Traynis, I.; Singh, P.; Corrado, G. S.; Webster, D. R.; Peng, L. H.; Varadarajan, A. V.; Liu, Y.; Hammel, N.
Show abstract
Recently it was shown that blood hemoglobin concentration could be predicted from retinal fundus photographs by deep learning models. However, it is unclear whether the models were quantifying current blood hemoglobin level, or estimating based on subjects pretest probability of having anemia. Here, we conducted an observational study with 14 volunteers who donated blood at an on site blood drive held by the local blood center (ie, at which time approximately 10% of their blood was removed). When the deep learning model was applied to retinal fundus photographs taken before and after blood donation, it detected a decrease in blood hemoglobin concentration within each subject at 2-3 days after donation, suggesting that the model was quantifying subacute hemoglobin changes instead of predicting subjects risk. Additional randomized or controlled studies can further validate this finding.
Matching journals
The top 10 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Comparison of haemoglobin concentration measurements using HemoCue-301 and Sysmex XN-Series 1500: a survey among anaemic Gambian infants aged 6-12 months 92%
- SpO 2 /FiO 2 Ratio as a Better Metric for Assessment of RBC Transfusion Effectiveness in Non-traumatic Critically Ill Patients with Physiologic Derangements 92%
- Seroprevalence of Anti-SARS-CoV-2 Antibodies in a Cohort of New York City Metro Blood Donors using Multiple SARS-CoV-2 Serological Assays: Implications for Controlling the Epidemic and Reopening. 92%
Similar papers in this journal
- Randomized controlled trial of convalescent plasma therapy against standard therapy in patients with severe COVID-19 disease 92%
- Estimation of the Hemoglobin Glycation Rate Constant 92%
- Hilab system, a new point-of-care hematology analyzer supported by the Internet of Things and Artificial Intelligence 91%
Similar papers in this journal
- Population Performance and Individual Agreement of Coronary Artery Disease Polygenic Risk Scores 86%
- Effect of Vitamin D 3 Supplementation vs Placebo on Hospital Length of Stay in Patients with Severe COVID-19: A Multicenter, Double-blind, Randomized Controlled Trial 85%
- Prevalence of SARS-CoV-2 Antibodies from a one-year National Serosurveillance of Kenyan Blood Transfusion Donors 85%
Similar papers in this journal
- Blood Unit Segments Accurately Represent the Biophysical Properties of Red Blood Cells in Blood Bags but Not Hemolysis 93%
- The Local Initiative For Emergency Blood (LIFE-Blood) Study: A mixed-methods, single center exploration of civilian walking blood bank need, feasibility, and safety in a low-resource blood desert 92%
- The risk of transfusion transmissible infection in a civilian walking blood bank using rapid diagnostic tests: A modeling study 92%
Similar papers in this journal
- Application of machine learning to discriminate photoreceptor cell species in xenotransplanted chimeric retinas 89%
- Contextualising Single-Arm Ophthalmology Trials With Real-World Data: An Emulated Target Trial Comparing The Efficacy Of Interventions For Neovascular Age-Related Macular Degeneration 87%
- Managing Endothelial Dysfunction in COVID-19: A Pilot, Double-Blind, Placebo-Controlled, Randomized Clinical Trial at the Lebanese American University Medical Center - Rizk Hospital (MEDIC-LAUMCRH) 87%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.