Related Experiment Video
Updated: Jul 19, 2025
Reconstruct Human Retinoblastoma In Vitro
Published on: October 11, 2022
Semi-supervised segmentation of retinoblastoma tumors in fundus images
Amir Rahdar1, Mohamad Javad Ahmadi1, Masood Naseripour2
1Chashmyar Company, Tehran, Iran.
Abstract:
Retinoblastoma is a rare form of cancer that predominantly affects young children as the primary intraocular malignancy. Studies conducted in developed and some developing countries have revealed that early detection can successfully cure over 90% of children with retinoblastoma. An unusual white reflection in the pupil is the most common presenting symptom. Depending on the tumor size, shape, and location, medical experts may opt for different approaches and treatments, with the results varying significantly due to the high reliance on prior knowledge and experience. This study aims to present a model based on semi-supervised machine learning that will yield segmentation results comparable to those achieved by medical experts. First, the Gaussian mixture model is utilized to detect abnormalities in approximately 4200 fundus images. Due to the high computational cost of this process, the results of this approach are then used to train a cost-effective model for the same purpose. The proposed model demonstrated promising results in extracting highly detailed boundaries in fundus images. Using the Sørensen-Dice coefficient as the comparison metric for segmentation tasks, an average accuracy of 93% on evaluation data was achieved.
Insights
This study introduces a semi-supervised machine learning model for early retinoblastoma detection. The model achieves 93% accuracy in segmenting ocular abnormalities from fundus images, aiding in early cancer diagnosis.
Area of Science:
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background:
- Retinoblastoma is the primary intraocular malignancy in young children.
- Early detection of retinoblastoma offers a >90% cure rate.
- Accurate tumor segmentation is crucial for effective treatment planning.
Purpose of the Study:
- To develop a semi-supervised machine learning model for retinoblastoma segmentation in fundus images.
- To achieve segmentation accuracy comparable to expert ophthalmologists.
- To create a cost-effective automated detection system.
Main Methods:
- Utilized Gaussian mixture models for initial abnormality detection in ~4200 fundus images.
- Trained a cost-effective model using initial detection results for improved efficiency.
- Employed semi-supervised machine learning for image segmentation.
Main Results:
- The proposed model achieved an average accuracy of 93% using the Sørensen-Dice coefficient.
- Demonstrated high precision in extracting detailed tumor boundaries from fundus images.
- The model provides results comparable to those of medical experts.
Conclusions:
- Semi-supervised machine learning offers a viable approach for accurate retinoblastoma segmentation.
- The developed model can assist in early diagnosis and treatment planning for retinoblastoma.
- Automated segmentation can improve the efficiency and consistency of retinoblastoma detection.
More Related Videos
11:03Subretinal Transplantation of Human Embryonic Stem Cell Derived-retinal Pigment Epithelial Cells into a Large-eyed Model of Geographic Atrophy
Published on: January 22, 2018
07:55Establishment and Propagation of Human Retinoblastoma Tumors in Immune Deficient Mice
Published on: August 4, 2011
