Purpose: To develop an artificial intelligence (AI) solution for automated segmentation and analysis of joint cardiac MRI short-axis T1 and T2 mapping.
Materials And Methods: In this retrospective study, a joint T1 and T2 mapping sequence was used to acquire 4240 maps from 807 patients across two hospitals between March and November 2020. Five hundred nine maps from 94 consecutive patients were assigned to a holdout testing set. A convolutional neural network was trained to segment the endocardial and epicardial contours with use of an edge probability estimation approach. Training labels were segmented by an expert cardiologist. Predicted contours were processed to yield mapping values for each of the 16 American Heart Association segments. Network segmentation performance and segment-wise measurements on the testing set were compared with those of two experts on the holdout testing set. The AI model was fully integrated using open-source software to run on MRI scanners.
Results: A total of 3899 maps (92%) were deemed artifact-free and suitable for human segmentation. AI segmentation closely matched that of each expert (mean Dice coefficient, 0.82 ± 0.07 [SD] vs expert 1 and 0.86 ± 0.06 vs expert 2) and compared favorably with interexpert agreement (Dice coefficient, 0.84 ± 0.06 for expert 1 vs expert 2). AI-derived segment-wise values for native T1, postcontrast T1, and T2 mapping correlated with expert-derived values ( = 0.96, 0.98, and 0.87, respectively, vs expert 1, and 0.97, 0.99, and 0.92 vs expert 2) and fell within the range of interexpert reproducibility ( = 0.97, 0.99, and 0.90, respectively). The AI model has since been deployed at two hospitals, enabling automated inline analysis.
Conclusion: Automated inline analysis of joint T1 and T2 mapping allows accurate segment-wise tissue characterization, with performance equivalent to that of human experts. MRI, Neural Networks, Cardiac, Heart © RSNA, 2022.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9885378 | PMC |
| http://dx.doi.org/10.1148/ryai.220050 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Automatic cleaning in acoustic ejection mass spectrometry: Enhancing the system robustness for large-scale high-throughput analysis of complex samples.
SLAS Technol
December 2024
SCIEX, Concord, Ontario L4K 4V8, Canada. Electronic address:
The rapid evolution of high-throughput mass spectrometry (HT-MS) technologies has positioned MS as a pivotal analytical tool across diverse disciplines. Its significance is particularly pronounced in high-throughput drug discovery and development, where MS plays a critical role throughout various phases. Acoustic ejection mass spectrometry (AEMS) is a recent addition to the HT-MS landscape, showcasing a balanced performance high analytical throughput and high data quality.
View Article and Find Full Text PDFCompact Arterial Monitoring Device Use in Resuscitative Endovascular Balloon Occlusion of the Aorta (REBOA): A Simple Validation Study in Swine.
Cureus
October 2024
Department of Anesthesiology, Uniformed Services University of the Health Sciences, Bethesda, USA.
Article Synopsis
- - Hemorrhage is the main cause of preventable death in trauma situations, leading to military and civilian advancements in medical practices, particularly through the use of tourniquets to manage extremity bleeding and save lives.
- - While tourniquets have significantly decreased deaths from bleeding in military settings, noncompressible hemorrhage still poses a major risk, especially before patients receive definitive medical care.
- - The study explores using a small, disposable pressure monitor during resuscitative endovascular balloon occlusion of the aorta (REBOA) to enhance blood pressure monitoring, facilitate better resuscitation practices, and reduce the need for blood products in extreme environments.
Facilitating Absorption Spectroscopy of Strongly Absorbing Fluids: A High-Throughput Approach.
J Phys Chem A
November 2024
Dynamics and Structure Formation - Herzig Group, University of Bayreuth, Universitätsstraße 30, Bayreuth 95447, Germany.
Measuring absorption spectra of strongly absorbing solutions is usually only possible by diluting the solution. However, the absorption spectra can be heavily influenced by concentration-dependent interactions, so dilution leads to misleading results. To enable reliable absorption measurements of concentrated solutions, we introduce in this work thinning fluid film spectroscopy (TFFS), a simple and highly automatable method.
View Article and Find Full Text PDFAutomated detection of motion artifacts in brain MR images using deep learning.
NMR Biomed
January 2025
Columbia Magnetic Resonance Research Center, Columbia University in the City of New York, New York, New York, USA.
Article Synopsis
- * A 2D convolutional neural network (CNN) was trained on synthesized data and tested on various datasets, achieving notable precision (85%) and recall (80%) in identifying motion issues.
- * The model demonstrated excellent agreement with a radiologist's assessments (93%) and correlates strongly with an image quality metric, aiming to streamline the quality assessment process, especially in low-resource environments.
On the Use of Reflection Polarized Optical Microscopy for Rapid Comparison of Crystallinity and Phase Segregation of P3HT:PCBM Thin Films.
Macromol Rapid Commun
December 2024
Physical Sciences and Engineering Division (PSE), KAUST Solar Center (KSC), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955-6900, Kingdom of Saudi Arabia.
Rapid, nondestructive characterization techniques for evaluating the degree of crystallinity and phase segregation of organic semiconductor blend thin films are highly desired for in-line, automated optoelectronic device fabrication facilities. Here, it is demonstrated that reflection polarized optical microscopy (POM), a simple technique capable of imaging local anisotropy of materials, is capable of determining the relative degree of crystallinity and phase segregation of thin films of polymer:fullerene blends. While previous works on POM of organic semiconductors have largely employed the transmission geometry, it is demonstrated that reflection POM provides 3× greater contrast.
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!