Interpretable CNN-Multilevel Attention Transformer for Rapid Recognition of Pneumonia From Chest X-Ray Images

Deep Learning-Based Identification of Intraocular Pressure-Associated Genes Influencing Trabecular Meshwork Cell Morphology

Purpose

Genome-wide association studies have recently uncovered many loci associated with variation in intraocular pressure (IOP). Artificial intelligence (AI) can be used to interrogate the effect of specific genetic knockouts on the morphology of trabecular meshwork cells (TMCs) and thus, IOP regulation.

Design

Experimental study.

Subjects

Primary TMCs collected from human donors.

Methods

Sixty-two genes at 55 loci associated with IOP variation were knocked out in primary TMC lines. All cells underwent high-throughput microscopy imaging after being stained with a 5-channel fluorescent cell staining protocol. A convolutional neural network was trained to distinguish between gene knockout and normal control cell images. The area under the receiver operator curve (AUC) metric was used to quantify morphological variation in gene knockouts to identify potential pathological perturbations.

Main Outcome Measures

Degree of morphological variation as measured by deep learning algorithm accuracy of differentiation from normal controls.

Results

Cells where LTBP2 or BCAS3 had been perturbed demonstrated the greatest morphological variation from normal TMCs (AUC 0.851, standard deviation [SD] 0.030; and AUC 0.845, SD 0.020, respectively). Of 7 multigene loci, 5 had statistically significant differences in AUC (P < 0.05) between genes, allowing for pathological gene prioritization. The mitochondrial channel most frequently showed the greatest degree of morphological variation (33.9% of cell lines).

Conclusions

We demonstrate a robust method for functionally interrogating genome-wide association signals using high-throughput microscopy and AI. Genetic variations inducing marked morphological variation can be readily identified, allowing for the gene-based dissection of loci associated with complex traits.

Financial Disclosures

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Ensemble learning for retinal disease recognition under limited resources

Retinal optical coherence tomography (OCT) images provide crucial insights into the health of the posterior ocular segment. Therefore, the advancement of automated image analysis methods is imperative to equip clinicians and researchers with quantitative data, thereby facilitating informed decision-making. The application of deep learning (DL)-based approaches has gained extensive traction for executing these analysis tasks, demonstrating remarkable performance compared to labor-intensive manual analyses. However, the acquisition of retinal OCT images often presents challenges stemming from privacy concerns and the resource-intensive labeling procedures, which contradicts the prevailing notion that DL models necessitate substantial data volumes for achieving superior performance. Moreover, limitations in available computational resources constrain the progress of high-performance medical artificial intelligence, particularly in less developed regions and countries. This paper introduces a novel ensemble learning mechanism designed for recognizing retinal diseases under limited resources (e.g., data, computation). The mechanism leverages insights from multiple pre-trained models, facilitating the transfer and adaptation of their knowledge to retinal OCT images. This approach establishes a robust model even when confronted with limited labeled data, eliminating the need for an extensive array of parameters, as required in learning from scratch. Comprehensive experimentation on real-world datasets demonstrates that the ensemble models constructed by the proposed ensemble method show superior performance over the baseline models under sparse labeled data, especially the triple ensemble model, which achieves the accuracy of 92.06%, which is 8.27%, 7.99%, and 11.14% better than the three baseline models, respectively. In addition, compared with the three baseline models learned from scratch, the triple ensemble model has fewer trainable parameters, only 3.677M, which is lower than the three baseline models of 8.013M, 4.302M, and 20.158M, respectively.

Editorial Topical Collection: "Explainable and Augmented Machine Learning for Biosignals and Biomedical Images"

Machine learning (ML) is a well-known subfield of artificial intelligence (AI) that aims at developing algorithms and statistical models able to empower computer systems to automatically adapt to a specific task through experience or learning from data [...].