ABSTRACT
Purpose: To gain an understanding of data labeling requirements to train deep learning models for measurement of geographic atrophy (GA) with fundus autofluorescence (FAF) images. Design: Evaluation of artificial intelligence (AI) algorithms. Subjects: The Age-Related Eye Disease Study 2 (AREDS2) images were used for training and cross-validation, and GA clinical trial images were used for testing. Methods: Training data consisted of 2 sets of FAF images; 1 with area measurements only and no indication of GA location (Weakly labeled) and the second with GA segmentation masks (Strongly labeled). Main Outcome Measures: Bland-Altman plots and scatter plots were used to compare GA area measurement between ground truth and AI measurements. The Dice coefficient was used to compare accuracy of segmentation of the Strong model. Results: In the cross-validation AREDS2 data set (n = 601), the mean (standard deviation [SD]) area of GA measured by human grader, Weakly labeled AI model, and Strongly labeled AI model was 6.65 (6.3) mm2, 6.83 (6.29) mm2, and 6.58 (6.24) mm2, respectively. The mean difference between ground truth and AI was 0.18 mm2 (95% confidence interval, [CI], -7.57 to 7.92) for the Weakly labeled model and -0.07 mm2 (95% CI, -1.61 to 1.47) for the Strongly labeled model. With GlaxoSmithKline testing data (n = 156), the mean (SD) GA area was 9.79 (5.6) mm2, 8.82 (4.61) mm2, and 9.55 (5.66) mm2 for human grader, Strongly labeled AI model, and Weakly labeled AI model, respectively. The mean difference between ground truth and AI for the 2 models was -0.97 mm2 (95% CI, -4.36 to 2.41) and -0.24 mm2 (95% CI, -4.98 to 4.49), respectively. The Dice coefficient was 0.99 for intergrader agreement, 0.89 for the cross-validation data, and 0.92 for the testing data. Conclusions: Deep learning models can achieve reasonable accuracy even with Weakly labeled data. Training methods that integrate large volumes of Weakly labeled images with small number of Strongly labeled images offer a promising solution to overcome the burden of cost and time for data labeling. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
ABSTRACT
BACKGROUND: Neuroinflammation plays an important role in ischemic brain injury and recovery, however the interplay between brain development and the neuroinflammatory response is poorly understood. We previously described age-dependent differences in the microglial response and the effect of microglial inhibition. Here we investigate whether age-dependent microglial responses may be related to pre-injury developmental differences in microglial phenotype. METHODS: Measures of microglia morphology were quantified using semi-automated software analysis of immunostained sections from postnatal day 2 (P2), P9, P30 and P60 mice using IMARIS. Microglia were isolated from P2, P9, P30 and P60 mice, and expression of markers of classical and alternative microglial activation was assessed, as well as transforming growth factor beta (TGF-ß) receptor, Serpine1, Mer Tyrosine Kinase (MerTK), and the suppressor of cytokine signaling (SOCS3). Hypoxia-ischemia (HI) was induced in P9 and P30 mice using unilateral carotid artery ligation and exposure to 10% oxygen for 50â¯min. Microglia morphology and microglial expression of genes in the TGF-ß and MerTK pathways were determined in ipsilateral and contralateral hippocampus. RESULTS: A progressive and significant increase in microglia branching morphology was seen in all brain regions from P2 to P30. No consistent classical or alternative activation profile was seen in isolated microglia. A clear transition to increased expression of TGF-ß and its downstream effector serpine1 was seen between P9 and P30. A similar increase in expression was seen in MerTK and its downstream effector SOCS3. HI resulted in a significant decrease in branching morphology only in the P9 mice, and expression of TGF-ß receptor, Serpine1, MerTK, and SOCS3 were elevated in P30 mice compared to P9 post-HI. CONCLUSION: Microglia maturation is associated with changes in morphology and gene expression, and microglial responses to ischemia in the developing brain differ based on the age at which injury occurs.
