Post

New research · Ophthalmology
Translational vision science & technology · 16h
AI / informaticsTranslational vision science & technology · 2025

RB-Care: An Artificial Intelligence System for Automatic Quantitative Assessment and Surveillance of Retinoblastoma.

Yarou Hu, Xinyu Zhao, Zhenquan Wu … Guoming Zhang
Read paper
OphthalmologyAI / informatics

Artificial intelligence system RB-Care classified retinoblastoma fundus images with 97.3% accuracy.

RB-Care: An Artificial Intelligence System for Automatic Quantitative Assessment and Surveillance of Retinoblastoma.

Yarou Hu … Guoming Zhang
Translational vision science & technology · 2025
Background

Accurate assessment and surveillance of retinoblastoma (retinoblastoma) require more efficient and objective measurements.

Purpose

This study aims to develop an artificial intelligence (artificial intelligence) system, named retinoblastoma-Care, for automatic classification and quantitative assessment of retinoblastoma.

Methods

A total of 3730 wide-field fundus images were included for the development and validation of 2 models in RB-Care.

n = 3730 wide-field fundus images
0.9734
Results
0.9734
correctly sorted images into normal or tumor types almost every time
n = 3730 wide-field fundus images
More results

The segmentation model can make precise boundary detection and quantitative measurement on tumors and optic discs, achieving mean Intersection over Union (mIoU) of 0.9670 and Dice similarity coefficient (dice similarity coefficient) of 0.9780 for tumor segmentation, and mIoU of 0.9999 and dice similarity coefficient of 0.9999 for optic disc segmentation, which reaches a comparable level with ophthalmologists.

“
Conclusion · 1 of 2

The retinoblastoma-Care achieved excellent performance in both retinoblastoma classification and segmentation.

Conclusion · 2 of 2

Consequently, the tumor size and the distance between tumor and optic disc can be quantified, which provides an objective measurement tool for quantitative assessment and surveillance of retinoblastoma in clinical settings.

Read paper
0 comments

No comments yet. Be the first.

Related papers

LatestFoundational
AI / Informatics
0·962 for OCT
MerMED-FM was very accurate at diagnosing diseases using eye scans
AI / Informatics
0.98
Artificial intelligence's eyelid-height measurements matched doctors' manual measurements almost perfectly
AI / Informatics
92.1%
combining eye scans and photos correctly told benign from cancerous lesions apart nearly every time
Cohort Study
28.6%
recurred locally in more than 1 in 4 patients over years of follow-up
Cohort Study
-0.395
excision group's post-op eyelid fullness score was lower - greater improvement
Cohort Study
86.7%
of infants probed after 12 months still had unresolved tear duct blockage
Observational
16%
eyelid tissue in rosacea patients showed less of this key repair-signaling protein inside cell nuclei
Cohort Study
25%
about 1 in 4 treated cases had symptoms return after improving