Heterogeneity Measurement of Cardiac Tissues Leveraging Uncertainty Information from Image Segmentation

Huang, Zhensheng; Gan, Yu; Lye, Theresa H.; Zhang, Haofeng; Laine, Andrew F.; Angelini, Elsa D.; Hendon, Christine P.

doi:10.1007/978-3-030-59710-8_76

Cited by 4 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We performed molecular docking simulations on the RNA-dependent RNA polymerase (RdRp) (Protein Data Bank ID: 6M71), and the main protease (Mpro) (PDB ID: 6Y2E) of SARS-CoV-2, using default settings in the Protein–Ligand ANT System (PLANTS) (Korb et al 2009 ), as described elsewhere (Copertino et al 2021 b). The ligand docking sites were specified, respectively, as the catalytic sites determined by Zhang et al ( 2020 a) (Gln189) and Gao et al ( 2020 ) (Asp623), using an estimated radius of 10 Å around the specified residues. The resulting protein–ligand scores (PLANTS scores), calculated using the CHEMPLP algorithm, reflect the energy change when ligands and proteins come together, with values less than − 80.00 suggesting effective ligand–protein interactions.…”

Section: Methodsmentioning

confidence: 99%

Identifying FDA-approved drugs with multimodal properties against COVID-19 using a data-driven approach and a lung organoid model of SARS-CoV-2 entry

Duarte

Copertino

Íñiguez

et al. 2021

Mol Med

View full text Add to dashboard Cite

Background Vaccination programs have been launched worldwide to halt the spread of COVID-19. However, the identification of existing, safe compounds with combined treatment and prophylactic properties would be beneficial to individuals who are waiting to be vaccinated, particularly in less economically developed countries, where vaccine availability may be initially limited. Methods We used a data-driven approach, combining results from the screening of a large transcriptomic database (L1000) and molecular docking analyses, with in vitro tests using a lung organoid model of SARS-CoV-2 entry, to identify drugs with putative multimodal properties against COVID-19. Results Out of thousands of FDA-approved drugs considered, we observed that atorvastatin was the most promising candidate, as its effects negatively correlated with the transcriptional changes associated with infection. Atorvastatin was further predicted to bind to SARS-CoV-2’s main protease and RNA-dependent RNA polymerase, and was shown to inhibit viral entry in our lung organoid model. Conclusions Small clinical studies reported that general statin use, and specifically, atorvastatin use, are associated with protective effects against COVID-19. Our study corroborrates these findings and supports the investigation of atorvastatin in larger clinical studies. Ultimately, our framework demonstrates one promising way to fast-track the identification of compounds for COVID-19, which could similarly be applied when tackling future pandemics.

show abstract

Section: Methodsmentioning

confidence: 99%

Identifying FDA-approved drugs with multimodal properties against COVID-19 using a data-driven approach and a lung organoid model of SARS-CoV-2 entry

Duarte

Copertino

Íñiguez

et al. 2021

Mol Med

View full text Add to dashboard Cite

show abstract

“…Their model contained a multi-modal training process that involved both en-face OCT and optical coherence tomography angiography (OCTA) to provide the intensity and geometric profiles. [23] trained a fully supervised segmentation network for cardiac tissue segmentation and used model uncertainty to estimate tissue heterogeneity. Existing work mainly relies upon fully supervised learning techniques.…”

Section: Related Workmentioning

confidence: 99%

“…Current automated analysis on cardiac OCT images is mostly based on fully supervised learning models [22], [23], [34]. These models were limited and suffered from the drawback of manual workload in the labeling process.…”

mentioning

confidence: 99%

Cardiac Adipose Tissue Segmentation via Image-Level Annotations

Huang

Gan

Lye

et al. 2023

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

Automatically identifying the structural substrates underlying cardiac abnormalities can potentially provide real-time guidance for interventional procedures. With the knowledge of cardiac tissue substrates, the treatment of complex arrhythmias such as atrial fibrillation and ventricular tachycardia can be further optimized by detecting arrhythmia substrates to target for treatment (i.e., adipose) and identifying critical structures to avoid. Optical coherence tomography (OCT) is a real-time imaging modality that aids in addressing this need. Existing approaches for cardiac image analysis mainly rely on fully supervised learning techniques, which suffer from the drawback of workload on labor-intensive annotation process of pixelwise labeling. To lessen the need for pixel-wise labeling, we develop a two-stage deep learning framework for cardiac adipose tissue segmentation using image-level annotations on OCT images of human cardiac substrates. In particular, we integrate class activation mapping with superpixel segmentation to solve the sparse tissue seed challenge raised in cardiac tissue segmentation. Our study bridges the gap between the demand on automatic tissue analysis and the lack of high-quality pixel-wise annotations. To the best of our knowledge, this is the first study that attempts to address cardiac tissue segmentation on OCT images via weakly supervised learning techniques. Within an invitro human cardiac OCT dataset, we demonstrate that our weakly supervised approach on image-level annotations achieves comparable performance as fully supervised methods trained on pixel-wise annotations.

show abstract

Trustworthy clinical AI solutions: A unified review of uncertainty quantification in Deep Learning models for medical image analysis

Lambert,

Forbes,

Doyle

et al. 2024

Artificial Intelligence in Medicine

View full text Add to dashboard Cite

Heterogeneity Measurement of Cardiac Tissues Leveraging Uncertainty Information from Image Segmentation

Cited by 4 publications

References 25 publications

Identifying FDA-approved drugs with multimodal properties against COVID-19 using a data-driven approach and a lung organoid model of SARS-CoV-2 entry

Identifying FDA-approved drugs with multimodal properties against COVID-19 using a data-driven approach and a lung organoid model of SARS-CoV-2 entry

Cardiac Adipose Tissue Segmentation via Image-Level Annotations

Trustworthy clinical AI solutions: A unified review of uncertainty quantification in Deep Learning models for medical image analysis

Contact Info

Product

Resources

About