AI Article Synopsis
- - The study aims to quantify uncertainty in deep learning segmentation of geographic atrophy (GA) using retrospective OCT image analysis from the SWAGGER cohort.
- - Two Bayesian deep learning techniques, Monte Carlo dropout and ensemble methods, were developed and compared against a traditional model, focusing on segmenting GA lesions and evaluating their performance with Dice scores and Shannon Entropy.
- - Results showed that the Bayesian models had higher Dice scores (0.90 and 0.88) and greater pixel-level uncertainty compared to the traditional model (0.82), suggesting that these techniques may enhance model reliability and assist clinicians in their decision-making for GA.
Article Abstract
Purpose: To apply methods for quantifying uncertainty of deep learning segmentation of geographic atrophy (GA).
Design: Retrospective analysis of OCT images and model comparison.
Participants: One hundred twenty-six eyes from 87 participants with GA in the SWAGGER cohort of the Nonexudative Age-Related Macular Degeneration Imaged with Swept-Source OCT (SS-OCT) study.
Methods: The manual segmentations of GA lesions were conducted on structural subretinal pigment epithelium en face images from the SS-OCT images. Models were developed for 2 approximate Bayesian deep learning techniques, Monte Carlo dropout and ensemble, to assess the uncertainty of GA semantic segmentation and compared to a traditional deep learning model.
Main Outcome Measures: Model performance (Dice score) was compared. Uncertainty was calculated using the formula for Shannon Entropy.
Results: The output of both Bayesian technique models showed a greater number of pixels with high entropy than the standard model. Dice scores for the Monte Carlo dropout method (0.90, 95% confidence interval 0.87-0.93) and the ensemble method (0.88, 95% confidence interval 0.85-0.91) were significantly higher ( < 0.001) than for the traditional model (0.82, 95% confidence interval 0.78-0.86).
Conclusions: Quantifying the uncertainty in a prediction of GA may improve trustworthiness of the models and aid clinicians in decision-making. The Bayesian deep learning techniques generated pixel-wise estimates of model uncertainty for segmentation, while also improving model performance compared with traditionally trained deep learning models.
Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11459066 | PMC |
| http://dx.doi.org/10.1016/j.xops.2024.100587 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Accurate and Efficient Algorithm for Detection of Alzheimer Disability Based on Deep Learning.
Cell Physiol Biochem
December 2024
Joint Institute for Nuclear Research, 141980 Dubna, Russiac.
Background/aims: Alzheimer's Disease (AD) is a progressive neurodegenerative disorder that severely affects cognitive functions and memory. Early detection is crucial for timely intervention and improved patient outcomes. However, traditional diagnostic tools, such as MRI and PET scans, are costly and less accessible.
View Article and Find Full Text PDFExamining objective and subjective sleep measures and neurocognition in older adults with HIV: A cross-sectional study in the deep South.
Appl Neuropsychol Adult
December 2024
School of Nursing, University of Alabama at Birmingham, Birmingham, AL, USA.
As people live longer with HIV, reports of poor sleep and neurocognitive impairments are expected to increase. Poor sleep and neurocognitive impairments commonly occur in people living with HIV (PLWH) and some medications (e.g.
View Article and Find Full Text PDFCan artificial intelligence improve patient educational material readability? A systematic review and narrative synthesis.
Intern Med J
December 2024
Department of Interventional Neuroradiology, Austin Health, Melbourne, Victoria, Australia.
Enhancing patient comprehension of their health is crucial in improving health outcomes. The integration of artificial intelligence (AI) in distilling medical information into a conversational, legible format can potentially enhance health literacy. This review aims to examine the accuracy, reliability, comprehensiveness and readability of medical patient education materials (PEMs) simplified by AI models.
View Article and Find Full Text PDFImpact of annotation imperfections and auto-curation for deep learning-based organ-at-risk segmentation.
Phys Imaging Radiat Oncol
October 2024
Amsterdam UMC location Vrije Universiteit Amsterdam, Department of Radiation Oncology, De Boelelaan 1117, Amsterdam, the Netherlands.
Background And Purpose: Segmentation imperfections (noise) in radiotherapy organ-at-risk segmentation naturally arise from specialist experience and image quality. Using clinical contours can result in sub-optimal convolutional neural network (CNN) training and performance, but manual curation is costly. We address the impact of simulated and clinical segmentation noise on CNN parotid gland (PG) segmentation performance and provide proof-of-concept for an easily implemented auto-curation countermeasure.
View Article and Find Full Text PDFDeep learning models are used to minimize the number of polyps that goes unnoticed by the experts and to accurately segment the detected polyps during interventions. Although state-of-the-art models are proposed, it remains a challenge to define representations that are able to generalize well and that mediate between capturing low-level features and higher-level semantic details without being redundant. Another challenge with these models is that they are computation and memory intensive, which can pose a problem with real-time applications.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!