Dataset Information

The assessment of fundus image quality labeling reliability among graders with different backgrounds.

ABSTRACT:

Purpose

For the training of machine learning (ML) algorithms, correctly labeled ground truth data are inevitable. In this pilot study, we assessed the performance of graders with different backgrounds in the labeling of retinal fundus image quality.

Methods

Color fundus photographs were labeled using a Python-based tool using four image categories: excellent (E), good (G), adequate (A) and insufficient for grading (I). We enrolled 8 subjects (4 with and 4 without medical background, groups M and NM, respectively) to whom a tutorial was presented on image quality requirements. We randomly selected 200 images from a pool of 18,145 expert-labeled images (50/E, 50/G, 50/A, 50/I). The performance of the grading was timed and the agreement was assessed. An additional grading round was performed with 14 labels for a more objective analysis.

Results

The median time (interquartile range) for the labeling task with 4 categories was 987.8 sec (418.6) for all graders and 872.9 sec (621.0) vs. 1019.8 sec (479.5) in the M vs. NM groups, respectively. Cohen's weighted kappa showed moderate agreement (0.564) when using four categories that increased to substantial (0.637) when using only three by merging the E and G groups. By the use of 14 labels, the weighted kappa values were 0.594 and 0.667 when assigning four or three categories, respectively.

Conclusion

Image grading with a Python-based tool seems to be a simple yet possibly efficient solution for the labeling of fundus images according to image quality that does not necessarily require medical background. Such grading can be subject to variability but could still effectively serve the robust identification of images with insufficient quality. This emphasizes the opportunity for the democratization of ML-applications among persons with both medical and non-medical background. However, simplicity of the grading system is key to successful categorization.

SUBMITTER: Laurik-Feuerstein KL

PROVIDER: S-EPMC9321443 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:PurposeThe assessment of retinal image (RI) quality holds significant importance in both clinical trials and large datasets, because suboptimal images can potentially conceal early signs of diseases, thereby resulting in inaccurate medical diagnoses. This study aims to develop an automatic method for Retinal Image Quality Assessment (RIQA) that incorporates visual explanations, aiming to comprehensively evaluate the quality of retinal fundus images (RIs).MethodsWe developed an automatic RIQA system, named Swin-MCSFNet, utilizing 28,792 RIs from the EyeQ dataset, as well as 2000 images from the EyePACS dataset and an additional 1,000 images from the OIA-ODIR dataset. After preprocessing, including cropping black regions, data augmentation, and normalization, a Swin-MCSFNet classifier based on the Swin-Transformer for multiple color-space fusion was proposed to grade the quality of RIs. The generalizability of Swin-MCSFNet was validated across multiple data centers. Additionally, for enhanced interpretability, a Score-CAM-generated heatmap was applied to provide visual explanations.ResultsExperimental results reveal that the proposed Swin-MCSFNet achieves promising performance, yielding a micro-receiver operating characteristic (ROC) of 0.93 and ROC scores of 0.96, 0.81, and 0.96 for the "Good," "Usable," and "Reject" categories, respectively. These scores underscore the accuracy of RIQA based on Swin-MCSF in distinguishing among the three categories. Furthermore, heatmaps generated across different RIQA classification scores and various color spaces suggest that regions in the retinal images from multiple color spaces contribute significantly to the decision-making process of the Swin-MCSFNet classifier.ConclusionsOur study demonstrates that the proposed Swin-MCSFNet outperforms other methods in experiments conducted on multiple datasets, as evidenced by the superior performance metrics and insightful Score-CAM heatmaps.Translational relevanceThis study constructs a new retinal image quality evaluation system, which will contribute to the subsequent research of retinal images.

Project description:Data quality is increasingly recognized as one of the most important confounding factors in brain imaging research. It is particularly important for studies of brain development, where age is systematically related to in-scanner motion and data quality. Prior work has demonstrated that in-scanner head motion biases estimates of structural neuroimaging measures. However, objective measures of data quality are not available for most structural brain images. Here we sought to identify quantitative measures of data quality for T1-weighted volumes, describe how these measures relate to cortical thickness, and delineate how this in turn may bias inference regarding associations with age in youth. Three highly-trained raters provided manual ratings of 1840 raw T1-weighted volumes. These images included a training set of 1065 images from Philadelphia Neurodevelopmental Cohort (PNC), a test set of 533 images from the PNC, as well as an external test set of 242 adults acquired on a different scanner. Manual ratings were compared to automated quality measures provided by the Preprocessed Connectomes Project's Quality Assurance Protocol (QAP), as well as FreeSurfer's Euler number, which summarizes the topological complexity of the reconstructed cortical surface. Results revealed that the Euler number was consistently correlated with manual ratings across samples. Furthermore, the Euler number could be used to identify images scored "unusable" by human raters with a high degree of accuracy (AUC: 0.98-0.99), and out-performed proxy measures from functional timeseries acquired in the same scanning session. The Euler number also was significantly related to cortical thickness in a regionally heterogeneous pattern that was consistent across datasets and replicated prior results. Finally, data quality both inflated and obscured associations with age during adolescence. Taken together, these results indicate that reliable measures of data quality can be automatically derived from T1-weighted volumes, and that failing to control for data quality can systematically bias the results of studies of brain maturation.