Quantitative clinical nuclear cardiology, part 2: Evolving/emerging applications

Slomka, Piotr J.; Moody, Jonathan B.; Miller, Robert J.H.; Renaud, Jennifer; Ficaro, Edward P.; Garcia, Ernest

doi:10.2967/jnumed.120.242537

Cited by 9 publications

(4 citation statements)

References 78 publications

(107 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine Learning (ML)-based PM interpretation already surpassed the clinical interpretation of SPECT MPI for MACE prediction 19 . The incorporation of AI techniques to standardize and automate processing of PET MPI could further improve cardiovascular risk stratification 20,21 to noninvasively support clinical decision for using coronary revascularization 22 and guide the clinical management of patients with suspected coronary artery disease. This would provide an accurate and systematic assessment of tissue perfusion hemodynamics in a one-stop-shop.…”

Section: Introductionmentioning

confidence: 99%

Predicting MACE from [82Rb] PET: Can AI outperformmore traditional quantitative assessment of themyocardial perfusion ?

Bors,

Abler,

Matthieu

et al. 2023

Preprint

View full text Add to dashboard Cite

Assessing the individual risk of Major Adverse Cardiac Events (MACE) is of major importance as cardiovascular diseases remain the leading cause of death worldwide. Quantitative Myocardial Perfusion Imaging (MPI) parameters such as stress Myocardial Blood Flow (sMBF) or Myocardial Flow Reserve (MFR) constitutes the gold standard for prognosis assessment. We propose a systematic investigation of the value of Artificial Intelligence (AI) to leverage [82Rb] Silicon PhotoMultiplier (SiPM) PET MPI for MACE prediction. We establish a general pipeline for AI model validation to assess and compare the performance of global (i.e. average of the entire MPI signal), regional (17 segments), radiomics and Convolutional Neural Network (CNN) models leveraging various MPI signals on a dataset of 234 patients. Results showed that all regional AI models significantly outperformed the global model (p < 0.001), where the best AUC of 73.9% (CI: 72.5-75.3) was obtained with a CNN model. A regional AI model based on MBF averages from 17 segments fed to a Logistic Regression (LR) constituted an excellent trade-off between model simplicity and performance, achieving an AUC of 73.4% (CI: 72.3-74.7). A radiomics model based on intensity features revealed that the global average was the least important feature when compared to other aggregations of the MPI signal over the myocardium. We conclude that AI models can allow better personalized prognosis assessment for MACE.

show abstract

Section: Introductionmentioning

confidence: 99%

Predicting MACE from [82Rb] PET: Can AI outperformmore traditional quantitative assessment of themyocardial perfusion ?

Bors,

Abler,

Matthieu

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…In addition to physical examination and baseline cardiological assessment using echocardiography, the cardio-oncologist can revert to a broad spectrum of nuclear cardiological diagnostic workup [ 22 , 23 ]. Many nuclear medicine imaging techniques can be used to detect cardiovascular toxicity of which the most promising and commonly used modalities will be discussed in this review article.…”

Section: Introductionmentioning

confidence: 99%

Shining Damaged Hearts: Immunotherapy-Related Cardiotoxicity in the Spotlight of Nuclear Cardiology

Kersting

Settelmeier

Mavroeidi

et al. 2022

IJMS

View full text Add to dashboard Cite

The emerging use of immunotherapies in cancer treatment increases the risk of immunotherapy-related cardiotoxicity. In contrast to conventional chemotherapy, these novel therapies have expanded the forms and presentations of cardiovascular damage to a broad spectrum from asymptomatic changes to fulminant short- and long-term complications in terms of cardiomyopathy, arrythmia, and vascular disease. In cancer patients and, particularly, cancer patients undergoing (immune-)therapy, cardio-oncological monitoring is a complex interplay between pretherapeutic risk assessment, identification of impending cardiotoxicity, and post-therapeutic surveillance. For these purposes, the cardio-oncologist can revert to a broad spectrum of nuclear cardiological diagnostic workup. The most promising commonly used nuclear medicine imaging techniques in relation to immunotherapy will be discussed in this review article with a special focus on the continuous development of highly specific molecular markers and steadily improving methods of image generation. The review closes with an outlook on possible new developments of molecular imaging and advanced image evaluation techniques in this exciting and increasingly growing field of immunotherapy-related cardiotoxicity.

show abstract

“…Machine Learning (ML)-based PM interpretation already surpassed the clinical interpretation of SPECT MPI for MACE prediction 19 . The incorporation of AI techniques to standardize and automate processing of PET MPI could further improve cardiovascular risk stratification 20,21 to noninvasively support clinical decision for using coronary revascularization 9 and guide the clinical management of patients with suspected coronary artery disease. This would provide an accurate and systematic assessment of tissue perfusion hemodynamics in a one-stop-shop.…”

mentioning

confidence: 99%

Comparing various AI approaches to traditional quantitative assessment of the myocardial perfusion in [82Rb] PET for MACE prediction

Bors,

Abler,

Dietz

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Assessing the individual risk of Major Adverse Cardiac Events (MACE) is of major importance as cardiovascular diseases remain the leading cause of death worldwide. Quantitative Myocardial Perfusion Imaging (MPI) parameters such as stress Myocardial Blood Flow (sMBF) or Myocardial Flow Reserve (MFR) constitutes the gold standard for prognosis assessment. We propose a systematic investigation of the value of Artificial Intelligence (AI) to leverage [$$^{82}$$ 82 Rb] Silicon PhotoMultiplier (SiPM) PET MPI for MACE prediction. We establish a general pipeline for AI model validation to assess and compare the performance of global (i.e. average of the entire MPI signal), regional (17 segments), radiomics and Convolutional Neural Network (CNN) models leveraging various MPI signals on a dataset of 234 patients. Results showed that all regional AI models significantly outperformed the global model ($$p<0.001$$ p < 0.001 ), where the best AUC of 73.9% (CI 72.5–75.3) was obtained with a CNN model. A regional AI model based on MBF averages from 17 segments fed to a Logistic Regression (LR) constituted an excellent trade-off between model simplicity and performance, achieving an AUC of 73.4% (CI 72.3–74.7). A radiomics model based on intensity features revealed that the global average was the least important feature when compared to other aggregations of the MPI signal over the myocardium. We conclude that AI models can allow better personalized prognosis assessment for MACE.

show abstract

Quantitative clinical nuclear cardiology, part 2: Evolving/emerging applications

Cited by 9 publications

References 78 publications

Predicting MACE from [82Rb] PET: Can AI outperformmore traditional quantitative assessment of themyocardial perfusion ?

Predicting MACE from [82Rb] PET: Can AI outperformmore traditional quantitative assessment of themyocardial perfusion ?

Shining Damaged Hearts: Immunotherapy-Related Cardiotoxicity in the Spotlight of Nuclear Cardiology

Comparing various AI approaches to traditional quantitative assessment of the myocardial perfusion in [82Rb] PET for MACE prediction

Contact Info

Product

Resources

About