1
|
Clerc OF, Vijayakumar S, Dorbala S. Radionuclide Imaging of Cardiac Amyloidosis: An Update and Future Aspects. Semin Nucl Med 2024; 54:717-732. [PMID: 38960850 DOI: 10.1053/j.semnuclmed.2024.05.012] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/17/2024] [Revised: 05/27/2024] [Accepted: 05/30/2024] [Indexed: 07/05/2024]
Abstract
Cardiac amyloidosis (CA) is caused by the misfolding, accumulation and aggregation of proteins into large fibrils in the extracellular compartment of the myocardium, leading to restrictive cardiomyopathy, heart failure and death. The major forms are transthyretin (ATTR) CA and light-chain (AL) CA, based on the respective precursor protein. Each of them requires early diagnosis for a timely treatment initiation that will improve patient outcomes. For this, radionuclide imaging is essentially used as single-photon emission computed tomography (SPECT) with bone-avid radiotracers or as positron emission tomography (PET) with amyloid-binding radiotracers. Both offer unprecedented specificity for the diagnostic of CA. SPECT has even revolutionized the diagnosis of ATTR-CA by making it non-invasive. Indeed, SPECT has now entered the standard diagnostic pathway to CA and has led to earlier diagnosis of the disease. SPECT also modified the epidemiology of ATTR-CA, highlighting that the disease is much more frequent than previously believed, and showing that ATTR-CA plays a substantial role in HFpEF and aortic stenosis, particularly among elderly patients. In parallel, amyloid-binding radiotracers for PET have accumulated a substantial amount of evidence, but are not approved for clinical use in CA yet. Further studies are needed to refine acquisition protocols and validate results in broader populations. Unlike bone-avid SPECT radiotracers, PET radiotracers have been specifically created to bind to amyloid fibrils. Thus, PET is the only imaging method that is truly specific for amyloid deposits and very sensitive to any amyloid type. Indeed, PET can not only detect ATTR-CA, but also AL-CA and rare hereditary forms. For both SPECT and PET, advances in quantitation of myocardial uptake have generated more granular and reproducible findings, paving the way for progress in earlier diagnosis, risk stratification and therapeutic response monitoring. Encouraging findings have shown that SPECT and PET are sensitive to early CA when other diagnostic methods are negative. Both radionuclide imaging techniques can predict adverse outcomes, but more evidence is needed to determine how to use them in conjunction with usual prognostic staging scores. Studies on follow-up imaging after therapy suggested that SPECT and PET can capture myocardial changes in CA, but again, more data are needed to meaningfully interpret such changes. Based on all these promising results, radionuclide imaging has the potential to further impact the landscape of CA in diagnosis, prognosis and follow-up, but also to substantially contribute to the assessment of novel therapies that will improve the lives of patients with CA.
Collapse
Affiliation(s)
- Olivier F Clerc
- Division of Nuclear Medicine, Department of Radiology, Brigham and Women's Hospital, Boston, MA; Cardiac Amyloidosis Program, Division of Cardiology, Department of Medicine, Brigham and Women's Hospital, Boston, MA
| | - Shilpa Vijayakumar
- Division of Nuclear Medicine, Department of Radiology, Brigham and Women's Hospital, Boston, MA; Cardiac Amyloidosis Program, Division of Cardiology, Department of Medicine, Brigham and Women's Hospital, Boston, MA
| | - Sharmila Dorbala
- Division of Nuclear Medicine, Department of Radiology, Brigham and Women's Hospital, Boston, MA; Cardiac Amyloidosis Program, Division of Cardiology, Department of Medicine, Brigham and Women's Hospital, Boston, MA; CV Imaging Program, Cardiovascular Division and Department of Radiology, Brigham and Women's Hospital, Boston, MA.
| |
Collapse
|
2
|
Lau ATC, DiDomenico RJ, Kim K. Cost-effectiveness of systematic screening and treatment of transthyretin amyloid cardiomyopathy (ATTR-CM) in patients with heart failure with preserved ejection fraction (HFpEF) in United States. Int J Cardiol 2024; 398:131598. [PMID: 37979789 DOI: 10.1016/j.ijcard.2023.131598] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/06/2023] [Revised: 10/25/2023] [Accepted: 11/14/2023] [Indexed: 11/20/2023]
Abstract
BACKGROUND Transthyretin amyloid cardiomyopathy (ATTR-CM) is an underdiagnosed cause of heart failure in clinical practice. 99mTc-pyrophosphate scintigraphy (PYP-scan) improves the accuracy of ATTR-CM detection, enabling timely initiation of tafamidis, a drug that slows the progression of ATTR-CM and lowers the risk of adverse cardiac events. PYP-scans, serum free light-chain (FLC) test and immunofixation electrophoresis (IFE) are critical components of a systematic screening. We assessed the cost-effectiveness of universal systematic screening (USS) compared to standard-of-care (SoC) selected clinical referrals for the systematic screening in patients aged 60 years or older with heart failure with preserved ejection fraction (HFpEF) and ventricular wall thickness of at least 12 mm. METHODS Two screening strategies, USS versus SoC screening for ATTR-CM were compared in a model-based assessment. Treatment decisions were based upon the accuracy of each screening strategy, which was followed by Markov state transitions across New York Heart Association (NYHA) functional classes and death. Model inputs were identified from a literature review. We calculated lifetime cost in 2022 US dollars and quality adjusted life-years (QALYs) of each strategy. The primary outcome was the incremental cost-effectiveness ratio (ICER). RESULTS The USS was associated with a significant increase in lifetime costs ($124,380 vs. $70,412) and modest improvement in QALYs (4.42 QALYs vs 4.36 QALYs). The ICER for the USS was $919,509 per QALY gained. ICER was sensitive to the age at the time of ATTR-CM diagnosis, true prevalence rate of ATTR-CM, and daily cost of tafamidis. CONCLUSIONS Owing to the high cost of treatment with tafamidis, USS along with PYP scan for ATTR-CM in older HFpEF patients with ventricular wall thickening is unlikely to become a cost-effective strategy at a liberal WTP threshold.
Collapse
Affiliation(s)
- Anson T C Lau
- Department of Pharmacy Systems, Outcomes and Policy, University of Illinois Chicago, Chicago, IL, United States of America
| | - Robert J DiDomenico
- Department of Pharmacy Practice, University of Illinois Chicago, Chicago, IL, United States of America; Center for Pharmacoepidemiology and Pharmacoeconomic Research, University of Illinois Chicago, Chicago, IL, United States of America
| | - Kibum Kim
- Department of Pharmacy Systems, Outcomes and Policy, University of Illinois Chicago, Chicago, IL, United States of America; Center for Pharmacoepidemiology and Pharmacoeconomic Research, University of Illinois Chicago, Chicago, IL, United States of America.
| |
Collapse
|
3
|
Usanase N, Uzun B, Ozsahin DU, Ozsahin I. A look at radiation detectors and their applications in medical imaging. Jpn J Radiol 2024; 42:145-157. [PMID: 37733205 DOI: 10.1007/s11604-023-01486-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/01/2023] [Accepted: 08/28/2023] [Indexed: 09/22/2023]
Abstract
The effectiveness and precision of disease diagnosis and treatment have increased, thanks to developments in clinical imaging over the past few decades. Science is developing and progressing steadily in imaging modalities, and effective outcomes are starting to show up as a result of the shorter scanning periods needed as well as the higher-resolution images generated. The choice of one clinical device over another is influenced by technical disparities among the equipment, such as detection medium, shorter scan time, patient comfort, cost-effectiveness, accessibility, greater sensitivity and specificity, and spatial resolution. Lately, computational algorithms, artificial intelligence (AI), in particular, have been incorporated with diagnostic and treatment techniques, including imaging systems. AI is a discipline comprised of multiple computational and mathematical models. Its applications aided in manipulating sophisticated data in imaging processes and increased imaging tests' accuracy and precision during diagnosis. Computed tomography (CT), positron emission tomography (PET), and Single Photon Emission Computed Tomography (SPECT) along with their corresponding radiation detectors have been reviewed in this study. This review will provide an in-depth explanation of the above-mentioned imaging modalities as well as the radiation detectors that are their essential components. From the early development of these medical instruments till now, various modifications and improvements have been done and more is yet to be established for better performance which calls for a necessity to capture the available information and record the gaps to be filled for better future advances.
Collapse
Affiliation(s)
- Natacha Usanase
- Operational Research Centre in Healthcare, Near East University, Mersin 10, Nicosia, Turkey.
| | - Berna Uzun
- Operational Research Centre in Healthcare, Near East University, Mersin 10, Nicosia, Turkey
- Department of Statistics, Carlos III Madrid University, Getafe, Madrid, Spain
| | - Dilber Uzun Ozsahin
- Operational Research Centre in Healthcare, Near East University, Mersin 10, Nicosia, Turkey
- Medical Diagnostic Imaging Department, College of Health Sciences, University of Sharjah, Sharjah, United Arab Emirates
| | - Ilker Ozsahin
- Operational Research Centre in Healthcare, Near East University, Mersin 10, Nicosia, Turkey
- Brain Health Imaging Institute, Department of Radiology, Weill Cornell Medicine, New York, NY, 10065, USA
| |
Collapse
|
4
|
Alwan L, Benz DC, Cuddy SAM, Dobner S, Shiri I, Caobelli F, Bernhard B, Stämpfli SF, Eberli F, Reyes M, Kwong RY, Falk RH, Dorbala S, Gräni C. Current and Evolving Multimodality Cardiac Imaging in Managing Transthyretin Amyloid Cardiomyopathy. JACC Cardiovasc Imaging 2024; 17:195-211. [PMID: 38099914 DOI: 10.1016/j.jcmg.2023.10.010] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/29/2023] [Revised: 10/06/2023] [Accepted: 10/18/2023] [Indexed: 01/29/2024]
Abstract
Amyloid transthyretin (ATTR) amyloidosis is a protein-misfolding disease characterized by fibril accumulation in the extracellular space that can result in local tissue disruption and organ dysfunction. Cardiac involvement drives morbidity and mortality, and the heart is the major organ affected by ATTR amyloidosis. Multimodality cardiac imaging (ie, echocardiography, scintigraphy, and cardiac magnetic resonance) allows accurate diagnosis of ATTR cardiomyopathy (ATTR-CM), and this is of particular importance because ATTR-targeting therapies have become available and probably exert their greatest benefit at earlier disease stages. Apart from establishing the diagnosis, multimodality cardiac imaging may help to better understand pathogenesis, predict prognosis, and monitor treatment response. The aim of this review is to give an update on contemporary and evolving cardiac imaging methods and their role in diagnosing and managing ATTR-CM. Further, an outlook is presented on how artificial intelligence in cardiac imaging may improve future clinical decision making and patient management in the setting of ATTR-CM.
Collapse
Affiliation(s)
- Louhai Alwan
- Department of Cardiology, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland
| | - Dominik C Benz
- Amyloidosis Program, Division of Cardiology, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA; Division of Nuclear Medicine, Department of Radiology, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA; CV Imaging Program, Cardiovascular Division, Department of Medicine and Radiology, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA; Cardiac Imaging, Department of Cardiology and Nuclear Medicine, Zurich University Hospital, Zurich, Switzerland
| | - Sarah A M Cuddy
- Amyloidosis Program, Division of Cardiology, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA; Division of Nuclear Medicine, Department of Radiology, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA; CV Imaging Program, Cardiovascular Division, Department of Medicine and Radiology, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA
| | - Stephan Dobner
- Department of Cardiology, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland
| | - Isaac Shiri
- Department of Cardiology, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland
| | - Federico Caobelli
- University Clinic of Nuclear Medicine, Inselspital, Bern University Hospital, Switzerland
| | - Benedikt Bernhard
- Department of Cardiology, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland; CV Imaging Program, Cardiovascular Division, Department of Medicine and Radiology, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA
| | - Simon F Stämpfli
- Department of Cardiology, Heart Centre Lucerne, Luzerner Kantonsspital, Lucerne, Switzerland
| | - Franz Eberli
- Department of Cardiology, Triemli Hospital (Triemlispital), Zurich, Switzerland
| | - Mauricio Reyes
- Insel Data Science Center, Inselspital, Bern University Hospital, Bern, Switzerland; Artificial Intelligence in Medical Imaging, ARTORG Center for Biomedical Research, University of Bern, Bern, Switzerland
| | - Raymond Y Kwong
- CV Imaging Program, Cardiovascular Division, Department of Medicine and Radiology, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA
| | - Rodney H Falk
- Amyloidosis Program, Division of Cardiology, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA
| | - Sharmila Dorbala
- Amyloidosis Program, Division of Cardiology, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA; Division of Nuclear Medicine, Department of Radiology, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA; CV Imaging Program, Cardiovascular Division, Department of Medicine and Radiology, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA
| | - Christoph Gräni
- Department of Cardiology, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland.
| |
Collapse
|
5
|
Motwani M. 2022 Artificial intelligence primer for the nuclear cardiologist. J Nucl Cardiol 2023; 30:2441-2453. [PMID: 35854041 DOI: 10.1007/s12350-022-03049-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/15/2022] [Accepted: 06/14/2022] [Indexed: 10/17/2022]
Abstract
Driven by advances in computing power, the past decade has seen rapid developments in artificial intelligence (AI) which now offers potential enhancements to every aspect of nuclear cardiology workflow including acquisition, reconstruction, segmentation, direct image analysis, and interpretation; as well as facilitating clinical and imaging big-data integration for superior personalized risk stratification. To understand the relevance and potential of AI in their field, this review provides a primer for nuclear cardiologists in 2022. The aim is to explain terminology and provide a summary of key current implementations, challenges, and future aspirations of AI-based enhancements to nuclear cardiology.
Collapse
Affiliation(s)
- Manish Motwani
- Department of Cardiology, Manchester Heart Institute, Manchester Royal Infirmary, Manchester Heart Centre, Manchester University NHS Foundation Trust, Oxford Road, Manchester, UK.
- Institute of Cardiovascular Science, University of Manchester, Manchester, UK.
| |
Collapse
|
6
|
Bargagna F, De Santi LA, Martini N, Genovesi D, Favilli B, Vergaro G, Emdin M, Giorgetti A, Positano V, Santarelli MF. Bayesian Convolutional Neural Networks in Medical Imaging Classification: A Promising Solution for Deep Learning Limits in Data Scarcity Scenarios. J Digit Imaging 2023; 36:2567-2577. [PMID: 37787869 PMCID: PMC10584795 DOI: 10.1007/s10278-023-00897-8] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/13/2023] [Revised: 08/04/2023] [Accepted: 08/07/2023] [Indexed: 10/04/2023] Open
Abstract
Deep neural networks (DNNs) have already impacted the field of medicine in data analysis, classification, and image processing. Unfortunately, their performance is drastically reduced when datasets are scarce in nature (e.g., rare diseases or early-research data). In such scenarios, DNNs display poor capacity for generalization and often lead to highly biased estimates and silent failures. Moreover, deterministic systems cannot provide epistemic uncertainty, a key component to asserting the model's reliability. In this work, we developed a probabilistic system for classification as a framework for addressing the aforementioned criticalities. Specifically, we implemented a Bayesian convolutional neural network (BCNN) for the classification of cardiac amyloidosis (CA) subtypes. We prepared four different CNNs: base-deterministic, dropout-deterministic, dropout-Bayesian, and Bayesian. We then trained them on a dataset of 1107 PET images from 47 CA and control patients (data scarcity scenario). The Bayesian model achieved performances (78.28 (1.99) % test accuracy) comparable to the base-deterministic, dropout-deterministic, and dropout-Bayesian ones, while showing strongly increased "Out of Distribution" input detection (validation-test accuracy mismatch reduction). Additionally, both the dropout-Bayesian and the Bayesian models enriched the classification through confidence estimates, while reducing the criticalities of the dropout-deterministic and base-deterministic approaches. This in turn increased the model's reliability, also providing much needed insights into the network's estimates. The obtained results suggest that a Bayesian CNN can be a promising solution for addressing the challenges posed by data scarcity in medical imaging classification tasks.
Collapse
Affiliation(s)
- Filippo Bargagna
- University of Pisa, Pisa, Italy.
- Fondazione G. Monasterio CNR - Regione Toscana, Pisa, Italy.
| | - Lisa Anita De Santi
- University of Pisa, Pisa, Italy
- Fondazione G. Monasterio CNR - Regione Toscana, Pisa, Italy
| | - Nicola Martini
- Fondazione G. Monasterio CNR - Regione Toscana, Pisa, Italy
| | - Dario Genovesi
- Fondazione G. Monasterio CNR - Regione Toscana, Pisa, Italy
| | | | | | - Michele Emdin
- Scuola Universitaria Superiore 'S. Anna', Pisa, Italy
| | | | | | | |
Collapse
|
7
|
Deep Learning to Classify AL versus ATTR Cardiac Amyloidosis MR Images. Biomedicines 2023; 11:biomedicines11010193. [PMID: 36672702 PMCID: PMC9855341 DOI: 10.3390/biomedicines11010193] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/29/2022] [Revised: 12/18/2022] [Accepted: 01/11/2023] [Indexed: 01/13/2023] Open
Abstract
The aim of this work was to compare the classification of cardiac MR-images of AL versus ATTR amyloidosis by neural networks and by experienced human readers. Cine-MR images and late gadolinium enhancement (LGE) images of 120 patients were studied (70 AL and 50 TTR). A VGG16 convolutional neural network (CNN) was trained with a 5-fold cross validation process, taking care to strictly distribute images of a given patient in either the training group or the test group. The analysis was performed at the patient level by averaging the predictions obtained for each image. The classification accuracy obtained between AL and ATTR amyloidosis was 0.750 for cine-CNN, 0.611 for Gado-CNN and between 0.617 and 0.675 for human readers. The corresponding AUC of the ROC curve was 0.839 for cine-CNN, 0.679 for gado-CNN (p < 0.004 vs. cine) and 0.714 for the best human reader (p < 0.007 vs. cine). Logistic regression with cine-CNN and gado-CNN, as well as analysis focused on the specific orientation plane, did not change the overall results. We conclude that cine-CNN leads to significantly better discrimination between AL and ATTR amyloidosis as compared to gado-CNN or human readers, but with lower performance than reported in studies where visual diagnosis is easy, and is currently suboptimal for clinical practice.
Collapse
|
8
|
Wechalekar AD, Fontana M, Quarta CC, Liedtke M. AL Amyloidosis for Cardiologists: Awareness, Diagnosis, and Future Prospects: JACC: CardioOncology State-of-the-Art Review. JACC CardioOncol 2022; 4:427-441. [PMID: 36444232 PMCID: PMC9700258 DOI: 10.1016/j.jaccao.2022.08.009] [Citation(s) in RCA: 19] [Impact Index Per Article: 9.5] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/05/2022] [Revised: 08/05/2022] [Accepted: 08/09/2022] [Indexed: 11/17/2022] Open
Abstract
Amyloid light chain (AL) amyloidosis is a rare, debilitating, often fatal disease. Symptoms of cardiomyopathy are common presenting features, and patients often are referred to cardiologists. Cardiac amyloid infiltration is the leading predictor of death. However, the variable presentation and perceived rarity of the disease frequently lead to delay in suspecting amyloidosis as a cause of heart failure, leading to misdiagnoses and a marked delay in diagnosis, with devastating consequences for the patient. A median time from symptom onset to correct diagnosis of about 2 years is often too long when median survival from diagnosis for patients with AL amyloidosis and cardiomyopathy is 4 months to 2 years. The authors highlight the challenges to diagnosis, identify gaps in the current knowledge, and summarize novel treatments on the horizon to raise awareness about the critical need for early recognition of symptoms and diagnosis of AL amyloidosis aimed at accelerating treatment and improving outcomes for patients.
Collapse
Key Words
- AL amyloidosis
- AL, amyloid light chain
- ASCT, autologous stem cell transplantation
- ATTR, transthyretin
- CMR, cardiac magnetic resonance imaging
- CR, complete response
- CyBorD, cyclophosphamide-bortezomib-dexamethasone
- FLC, free light chain
- Ig, immunoglobulin
- LGE, late gadolinium enhancement
- NT-proBNP, N-terminal pro–brain natriuretic peptide
- PCD, plasma cell dyscrasia
- QoL, quality of life
- VGPR, very good partial response
- awareness
- diagnosis
- future therapies
Collapse
Affiliation(s)
| | - Marianna Fontana
- National Amyloidosis Centre, London, United Kingdom
- Royal Free London NHS Foundation Trust, London, United Kingdom
| | - C. Cristina Quarta
- Alexion Pharmaceuticals, AstraZeneca Rare Disease, Boston, Massachusetts, USA
| | - Michaela Liedtke
- Stanford Amyloid Center, Stanford University School of Medicine, Stanford, California, USA
| |
Collapse
|
9
|
Nuclear Molecular Imaging of Disease Burden and Response to Treatment for Cardiac Amyloidosis. BIOLOGY 2022; 11:biology11101395. [PMID: 36290299 PMCID: PMC9598720 DOI: 10.3390/biology11101395] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 08/17/2022] [Revised: 09/20/2022] [Accepted: 09/21/2022] [Indexed: 11/16/2022]
Abstract
Simple Summary Cardiac amyloidosis (CA) is characterized by extracellular infiltration and deposition of amyloid fibrils primarily derived from the circulating transthyretin protein (TTR) or immunoglobulin light chain (AL). With the development of non-invasive diagnostic approaches and the emergence of new pharmacotherapeutic treatments for CA, the transformative effects of bone scintigraphy have been important in diagnosing TTR-CA. Positron emission tomography (PET) imaging is another promising, non-invasive option for the diagnosis of CA and may help differentiate between ATTR and AL amyloidosis. Bone-seeking single-photon emission tomography/computed tomography (SPECT/CT) quantification and amyloid-targeting PET imaging could be useful as a new strategy for disease burden and therapy monitoring to provide more insights into therapy response assessed by quantifying the amyloid burden in CA. Abstract Cardiac amyloidosis (CA) is a heterogeneous group of diseases in which extracellular insoluble amyloid proteins are deposited in specific organs and tissues locally or systemically, thereby interfering with physiological function. Transthyretin protein (TTR) and light chain (AL) amyloidosis are the most common types of cardiac amyloidosis. Radionuclide bone scintigraphy has recently become the most common non-invasive test for the diagnosis of TTR-CA but is of limited value for the diagnosis of AL-CA. PET has proved promising for the diagnosis of CA and its applications are expected to expand in the future. This review summarizes the current bone scintigraphy and amyloid-targeting Positron emission tomography (PET) imaging, the binding imaging properties of radiotracers, and the values of diagnosis, prognosis, and monitoring therapy response in CA.
Collapse
|
10
|
Miller RJH, Huang C, Liang JX, Slomka PJ. Artificial intelligence for disease diagnosis and risk prediction in nuclear cardiology. J Nucl Cardiol 2022; 29:1754-1762. [PMID: 35508795 DOI: 10.1007/s12350-022-02977-8] [Citation(s) in RCA: 5] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/29/2022] [Accepted: 03/29/2022] [Indexed: 10/18/2022]
Abstract
Artificial intelligence (AI) techniques have emerged as a highly efficient approach to accurately and rapidly interpret diagnostic imaging and may play a vital role in nuclear cardiology. In nuclear cardiology, there are many clinical, stress, and imaging variables potentially available, which need to be optimally integrated to predict the presence of obstructive coronary artery disease (CAD) or predict the risk of cardiovascular events. In spite of clinical awareness of a large number of potential variables, it is difficult for physicians to integrate multiple features consistently and objectively. Machine learning (ML) is particularly well suited to integrating this vast array of information to provide patient-specific predictions. Deep learning (DL), a branch of ML characterized by a multi-layered convolutional model architecture, can extract information directly from images and identify latent image features associated with a specific prediction. This review will discuss the latest AI applications to disease diagnosis and risk prediction in nuclear cardiology with a focus on potential clinical applications.
Collapse
Affiliation(s)
- Robert J H Miller
- Division of Artificial Intelligence in Medicine, Departments of Medicine, Imaging and Biomedical Sciences, Cedars-Sinai Medical Center, 8700 Beverly Blvd, Suite Metro 203, Los Angeles, CA, 90048, USA
- Department of Cardiac Sciences, University of Calgary and Libin Cardiovascular Institute, Calgary, AB, Canada
| | - Cathleen Huang
- Division of Artificial Intelligence in Medicine, Departments of Medicine, Imaging and Biomedical Sciences, Cedars-Sinai Medical Center, 8700 Beverly Blvd, Suite Metro 203, Los Angeles, CA, 90048, USA
| | - Joanna X Liang
- Division of Artificial Intelligence in Medicine, Departments of Medicine, Imaging and Biomedical Sciences, Cedars-Sinai Medical Center, 8700 Beverly Blvd, Suite Metro 203, Los Angeles, CA, 90048, USA
| | - Piotr J Slomka
- Division of Artificial Intelligence in Medicine, Departments of Medicine, Imaging and Biomedical Sciences, Cedars-Sinai Medical Center, 8700 Beverly Blvd, Suite Metro 203, Los Angeles, CA, 90048, USA.
| |
Collapse
|
11
|
Halme HL, Ihalainen T, Suomalainen O, Loimaala A, Mätzke S, Uusitalo V, Sipilä O, Hippeläinen E. Convolutional neural networks for detection of transthyretin amyloidosis in 2D scintigraphy images. EJNMMI Res 2022; 12:27. [PMID: 35524861 PMCID: PMC9079204 DOI: 10.1186/s13550-022-00897-9] [Citation(s) in RCA: 8] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/04/2022] [Accepted: 04/10/2022] [Indexed: 11/10/2022] Open
Abstract
BACKGROUND Transthyretin amyloidosis (ATTR) is a progressive disease which can be diagnosed non-invasively using bone avid [99mTc]-labeled radiotracers. Thus, ATTR is also an occasional incidental finding on bone scintigraphy. In this study, we trained convolutional neural networks (CNN) to automatically detect and classify ATTR from scintigraphy images. The study population consisted of 1334 patients who underwent [99mTc]-labeled hydroxymethylene diphosphonate (HMDP) scintigraphy and were visually graded using Perugini grades (grades 0-3). A total of 47 patients had visual grade ≥ 2 which was considered positive for ATTR. Two custom-made CNN architectures were trained to discriminate between the four Perugini grades of cardiac uptake. The classification performance was compared to four state-of-the-art CNN models. RESULTS Our CNN models performed better than, or equally well as, the state-of-the-art models in detection and classification of cardiac uptake. Both models achieved area under the curve (AUC) ≥ 0.85 in the four-class Perugini grade classification. Accuracy was good in detection of negative vs. positive ATTR patients (grade < 2 vs grade ≥ 2, AUC > 0.88) and high-grade cardiac uptake vs. other patients (grade < 3 vs. grade 3, AUC = 0.94). Maximum activation maps demonstrated that the automated deep learning models were focused on detecting the myocardium and not extracardiac features. CONCLUSION Automated convolutional neural networks can accurately detect and classify different grades of cardiac uptake on bone scintigraphy. The CNN models are focused on clinically relevant image features. Automated screening of bone scintigraphy images using CNN could improve the early diagnosis of ATTR.
Collapse
Affiliation(s)
- Hanna-Leena Halme
- Clinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of Helsinki, Helsinki, Finland
| | - Toni Ihalainen
- Clinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of Helsinki, Helsinki, Finland
| | - Olli Suomalainen
- Heart and Lung Center, Helsinki University Hospital and University of Helsinki, Helsinki, Finland
| | - Antti Loimaala
- Clinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of Helsinki, Helsinki, Finland.,Research Program in Systems Oncology, Faculty of Medicine, University of Helsinki, Helsinki, Finland
| | - Sorjo Mätzke
- Clinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of Helsinki, Helsinki, Finland
| | - Valtteri Uusitalo
- Clinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of Helsinki, Helsinki, Finland
| | - Outi Sipilä
- Clinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of Helsinki, Helsinki, Finland
| | - Eero Hippeläinen
- Clinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of Helsinki, Helsinki, Finland. .,Department of Physics, University of Helsinki, Helsinki, Finland.
| |
Collapse
|
12
|
Miller RJH, Singh A, Dey D, Slomka P. Artificial Intelligence and Cardiac PET/Computed Tomography Imaging. PET Clin 2021; 17:85-94. [PMID: 34809873 DOI: 10.1016/j.cpet.2021.06.011] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
Artificial intelligence is an important technology, with rapidly expanding applications for cardiac PET. We review the common terminology, including methods for training and testing, which are fundamental to understanding artificial intelligence. Next, we highlight applications to improve image acquisition, reconstruction, and segmentation. Computed tomographic imaging is commonly acquired in conjunction with PET and various artificial intelligence methods have been applied, including methods to automatically extract anatomic information or generate synthetic attenuation images. Last, we describe methods to automate disease diagnosis or risk stratification. This summary highlights the current and future clinical applications of artificial intelligence to cardiovascular PET imaging.
Collapse
Affiliation(s)
- Robert J H Miller
- Department of Cardiac Sciences, University of Calgary, GAA08 HRIC, 3230 Hospital Drive NW, Calgary AB, T2N 4Z6, Canada
| | - Ananya Singh
- Departments of Imaging and Medicine, Cedars-Sinai Medical Center, 8700 Beverly Blvd, Suite Metro 203, Los Angeles, CA 90048, USA
| | - Damini Dey
- Departments of Imaging and Medicine, Cedars-Sinai Medical Center, 8700 Beverly Blvd, Suite Metro 203, Los Angeles, CA 90048, USA
| | - Piotr Slomka
- Departments of Imaging and Medicine, Cedars-Sinai Medical Center, 8700 Beverly Blvd, Suite Metro 203, Los Angeles, CA 90048, USA.
| |
Collapse
|
13
|
18F-Florbetaben: a potential nuclear protagonist in the diagnosis of cardiac amyloidosis. Clin Transl Imaging 2021. [DOI: 10.1007/s40336-021-00459-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/20/2022]
|