Cardiac Masses and Where to Find Them
Insaf Chouarfia , Ilenia Monaco , Mounia Sedrati , Fatima S. Bouhaik , Valeria Trivelloni , Stefano Salvini , Iheb Guefrachi , Yassine Bencharef , Dario Bottigliero
The Heart Surgery Forum ›› 2025, Vol. 28 ›› Issue (10) : 47574
Cardiac masses pose a significant diagnostic challenge, requiring a structured imaging-based approach. Echocardiography represents the first-line and most essential diagnostic tool, providing a rapid, non-invasive, and cost-effective method for detecting and characterizing intracardiac lesions. While metastatic involvement is the most frequent cause of secondary cardiac masses, the primary tumors are predominantly benign. However, distinguishing between tumors, thrombi, and pseudotumors often necessitates advanced imaging techniques, such as cardiac magnetic resonance imaging (MRI) or computed tomography (CT). Meanwhile, in addition to the diagnostic role, imaging techniques are essential for risk stratification and guiding therapeutic decisions. Thus, a multidisciplinary approach integrating multiple imaging modalities is crucial for optimizing patient management and improving outcomes.
cardiac tumors / cardiac thrombi / echocardiography / cardiac magnetic resonance imaging / computed tomography / positron emission tomography / multimodality imaging / tissue characterization / diagnostic algorithms / emerging technologies in cardiac imaging
3.2.2.1 Anatomical and Tissue Characterization Sequences
T1-weighted sequences effectively delineate anatomy and are particularly useful for identifying fat-containing masses such as lipomas, which appear hyperintense due to their short T1 relaxation time. These sequences also provide excellent definition of mass morphology, delineating precise borders and relationships with adjacent structures. T1-weighted sequences with fat suppression can further enhance diagnostic specificity by confirming the presence of adipose tissue within masses, a feature characteristic of certain entities such as lipomas or liposarcomas.
T2-weighted imaging helps assess fluid content within masses, with cysts and myxomas typically displaying high signal intensity due to their high water content. This sequence is particularly valuable for differentiating solid from cystic components and for identifying edematous changes within solid masses. Recent studies demonstrated that T2 signal intensity ratios (comparing mass signal to skeletal muscle) greater than 3.0 were highly specific (92%) for myxomas and cysts [11, 18].
Additionally, T2-weighted sequences with fat suppression can highlight areas of inflammation or edema, providing insights into the biological activity of the mass.
First-pass perfusion sequences evaluate the vascularity of masses by tracking the initial passage of gadolinium contrast through the cardiac chambers and myocardium. These dynamic sequences capture contrast enhancement patterns, with malignant tumors generally showing earlier and more pronounced enhancement compared to benign lesions or thrombi. Quantitative analysis of perfusion parameters, including time-to-peak enhancement and wash-in/wash-out kinetics, provides additional diagnostic information. Malignant masses typically demonstrate rapid enhancement with early washout, while benign lesions often show more gradual enhancement patterns.
3.2.2.2 Advanced Tissue Characterization Techniques
Late gadolinium enhancement (LGE) patterns are particularly valuable in mass characterization. Thrombi characteristically appear as filling defects with no enhancement, even on delayed imaging, whereas most tumors demonstrate variable degrees of enhancement. Mousavi et al. (2019) [11] demonstrated that specific LGE patterns correlate strongly with histopathological features, enabling more accurate pre-procedural diagnostic assessment. Their study of 145 patients showed that a combination of T1/T2 mapping and LGE achieved high diagnostic accuracy in distinguishing malignant from benign masses. Specific enhancement patterns have been associated with particular pathologies: heterogeneous enhancement with central hypoenhancement suggests necrotic areas commonly seen in malignancies, while peripheral rim enhancement is more characteristic of certain benign lesions or inflammatory processes. Fig. 2 demonstrates MRI appearance of left ventricular apical thrombus.
Parametric mapping techniques, including native T1, post-contrast T1, and T2 mapping, represent significant advances in tissue characterization. These techniques provide quantitative values that reflect intrinsic tissue properties, allowing for objective assessment beyond visual interpretation. Lavall et al. (2023) [19] demonstrated that these techniques can differentiate infiltrative processes, such as cardiac amyloidosis or sarcoidosis, from discrete masses with high specificity.
Their analysis found that native T1 values exceeding 1341 ms at 3.0 T were 100% sensitive and 97% specific for amyloidosis, while focal masses typically demonstrated heterogeneous T1 values. Pazos-López et al. (2014) [18] further established the value of parametric mapping in characterizing various cardiac masses, providing quantitative thresholds that can guide clinical decision-making. Recent works established that extracellular volume fraction (ECV) derived from T1 mapping was significantly higher in malignant tumors compared to benign masses, providing additional quantitative information for tissue characterization.
Feature tracking analysis, which assesses myocardial deformation through strain and strain rate measurements, provides insights into functional consequences of cardiac masses. This technique can evaluate whether a mass is adherent to or infiltrating the myocardium by detecting impaired regional deformation. Recent studies have demonstrated that strain patterns can differentiate infiltrative processes from space- occupying lesions with high sensitivity.
Diffusion-weighted imaging (DWI) measures the Brownian motion of water molecules and has shown utility in distinguishing malignant from benign cardiac masses. Recent studies have found that malignant lesions typically exhibit restricted diffusion with lower apparent diffusion coefficient (ADC) values compared to benign masses. The restricted diffusion observed in malignant lesions reflects their higher cellularity, nuclear-to-cytoplasmic ratio, and reduced extracellular space.
3.2.2.3 Standardization and Clinical Implementation
Grazzini et al. (2023) [20] published comprehensive guidelines on CMR protocols for cardiac mass evaluation, emphasizing the importance of standardized acquisition techniques and reporting templates to optimize diagnostic yield. The recommended protocol includes:
- Cine imaging in multiple planes (short-axis, 2-chamber, 3-chamber, and 4-chamber views);
- Black-blood T1-weighted imaging with and without fat suppression;
- T2-weighted imaging with and without fat suppression;
- First-pass perfusion imaging;
- Early and late gadolinium enhancement imaging;
- Parametric mapping (T1 native, T1 post-contrast, T2, and ECV calculation);
- Optional sequences including tagging, 4D flow, and diffusion-weighted imaging.
These protocols have been widely adopted and have significantly improved the reproducibility of CMR findings across different centers. The standardized reporting templates include essential elements such as mass location, size, morphology, signal characteristics across all sequences, enhancement patterns, functional impact, and differential diagnosis based on imaging features.
Recent technological advances in CMR, including higher field strengths (3T), improved spatial and temporal resolution, and accelerated acquisition techniques, have further enhanced the diagnostic capabilities of this modality. These improvements allow for more detailed characterization of small masses and better differentiation of mass components, further solidifying CMR’s position as the gold standard for cardiac mass evaluation.
3.2.3.1 Technical Advances and Imaging Protocols
Modern multi-detector CT (MDCT) scanners, with 64 to 320 detector rows, provide detailed anatomical information with isotropic spatial resolution approaching 0.5 mm and temporal resolution below 100 ms. These technical parameters allow for precise delineation of cardiac masses and their anatomical relationships. Lopez-Mattei et al. (2023) [22] published comprehensive protocols for cardiac CT acquisition and interpretation specifically optimized for mass evaluation. Their recommended protocol includes:
- Electrocardiography (ECG)-gated acquisition to minimize cardiac motion artifacts;
- Multiphasic imaging (including arterial, venous, and delayed phases) to optimize mass characterization;
- Thin-slice reconstruction (1 mm) with multiplanar reformations;
- Specific contrast administration protocols tailored to the suspected pathology;
- Dose-reduction techniques including tube current modulation, iterative reconstruction algorithms, and prospective ECG-gating when appropriate.
These protocols have been widely adopted in clinical practice and have significantly improved the diagnostic yield of cardiac CT for mass evaluation. Depending on the clinical question, Lopez-Mattei and colleagues [22] recommend tailoring the acquisition parameters to maximize diagnostic information while minimizing radiation exposure. For example, for suspected thrombi, a delayed phase acquisition (approximately 2 minutes after contrast administration) significantly improves detection sensitivity.
3.2.3.2 Advanced CT Technologies for Tissue Characterization
Dual-energy CT (DECT) significantly enhances tissue characterization by analyzing how tissues attenuate X-rays at different energy levels. Eberhard et al. (2024) [23] demonstrated that DECT improved diagnostic confidence compared to conventional CT. Studies have shown that DECT achieved 88% diagnostic accuracy for differentiating benign from malignant masses, with 93% specificity for lipomatous tumors.
Spectral CT, an evolution of DECT, further enhances tissue discrimination. Hong et al. (2018) [24] demonstrated in 41 patients that dual-energy CT achieved 67% sensitivity and 79% specificity for differentiating thrombi from neoplasms. Their analysis revealed significantly lower normalized iodine concentration in thrombi compared to neoplasms (1.79 0.26 vs. 2.98 0.23, p = 0.002), yielding an area under the ROC curve of 0.77. The utility of CT for thrombus identification is demonstrated in Fig. 3, which shows the characteristic imaging features that enable reliable differentiation from cardiac tumors.
3.2.3.3 Clinical Applications and Diagnostic Performance
CT coronary angiography offers the unique advantage of simultaneously evaluating coronary artery disease and cardiac masses, an important consideration in preoperative planning [25]. D’Angelo et al. (2020) [26] conducted a cohort study involving 60 patients that demonstrated CT achieved high sensitivity and specificity for differentiating between benign and malignant cardiac masses. Studies have identified several CT features highly predictive of malignancy:
- Irregular borders (odds ratio [OR] 4.8, 95% CI: 2.3–9.7);
- Heterogeneous enhancement (OR 3.6, 95% CI: 1.9–6.8);
- Involvement of more than one cardiac chamber (OR 5.2, 95% CI: 2.6–10.1);
- Contrast enhancement 50 Hounsfield units (OR 2.7, 95% CI: 1.4–5.3);
- Invasion of adjacent structures (OR 7.3, 95% CI: 3.4–15.9).
Additionally, CT excels in detecting calcification within masses, a feature often associated with specific pathologies such as fibromas or hemangiomas. The presence and pattern of calcification can be pathognomonic for certain entities—for example, the “popcorn” pattern of calcification often seen in cardiac fibromas.
Lopez-Mattei et al. (2021) [27] established a systematic approach to the differential diagnosis of cardiac masses on CT, providing radiologists with specific imaging features that distinguish between various pathologies. Their comprehensive analysis categorized cardiac masses based on location, enhancement patterns, and associated features, creating a structured algorithmic approach to diagnosis. For example:
- Left atrial masses with low attenuation and no enhancement most likely represent thrombi;
- Intramyocardial masses with fat attenuation suggest lipomas;
- Pericardial masses with high attenuation on non-contrast CT typically represent hematomas or calcified lesions;
- Masses involving the right heart with invasion into the pulmonary veins are highly suspicious for angiosarcoma.
This systematic approach has improved diagnostic accuracy and communication between radiologists and clinicians, facilitating more precise diagnosis and treatment planning.
3.2.3.4 Radiation Dose Considerations and Mitigation Strategies
While radiation exposure remains a concern with CT imaging, dose-reduction strategies and iterative reconstruction techniques have substantially mitigated this limitation. Modern cardiac CT protocols for mass evaluation typically deliver effective doses in the range of 2–5 mSv, significantly lower than earlier-generation scanners. Important dose-reduction strategies include:
- Tube current modulation, which adjusts radiation output based on patient size and anatomical region;
- Prospective ECG-gating, limiting radiation exposure to specific phases of the cardiac cycle;
- Iterative reconstruction algorithms, which maintain image quality while reducing radiation dose by 30–60%;
- Targeted protocols focusing only on the region of interest rather than the entire thorax.
Joudar et al. (2025) [21] reported that these optimization techniques have reduced the average radiation dose for cardiac mass evaluation over the past decade, making CT a more acceptable option for initial evaluation and follow-up imaging.
3.2.3.5 Integration With Other Imaging Modalities
CT findings often complement those from other imaging modalities, particularly in cases where CMR is contraindicated or provides inconclusive results. The integration of CT with echocardiography and, when available, nuclear imaging techniques provides a comprehensive assessment of cardiac masses. Current evidence suggests that the optimal role of cardiac CT in the diagnostic algorithm includes [22, 24]:
- First-line imaging for suspected calcified masses;
- Evaluation of patients with contraindications to CMR (e.g., pacemakers, claustrophobia);
- Assessment of coronary artery involvement by cardiac masses;
- Characterization of masses with inconclusive findings on echocardiography or CMR;
- Follow-up of known masses when CMR is not available or contraindicated.
This strategic use of cardiac CT optimizes its diagnostic value, while recognizing that CMR remains superior for characterizing masses that lack calcification or fatty components.
| [1] |
Paolisso P, Bergamaschi L, Angeli F, Belmonte M, Foà A, Canton L, et al. Cardiac Magnetic Resonance to Predict Cardiac Mass Malignancy: The CMR Mass Score. Circulation. Cardiovascular Imaging. 2024; 17: e016115. https://doi.org/10.1161/CIRCIMAGING.123.016115. |
| [2] |
Aghayev A, Cheezum MK, Steigner ML, Mousavi N, Padera R, Barac A, et al. Multimodality imaging to distinguish between benign and malignant cardiac masses. Journal of Nuclear Cardiology: Official Publication of the American Society of Nuclear Cardiology. 2022; 29: 1504–1517. https://doi.org/10.1007/s12350-021-02790-9. |
| [3] |
Colin GC, Gerber BL, Amzulescu M, Bogaert J. Cardiac myxoma: a contemporary multimodality imaging review. The International Journal of Cardiovascular Imaging. 2018; 34: 1789–1808. https://doi.org/10.1007/s10554-018-1396-z. |
| [4] |
Martineau P, Dilsizian V, Pelletier-Galarneau M. Incremental Value of FDG-PET in the Evaluation of Cardiac Masses. Current Cardiology Reports. 2021; 23: 78. https://doi.org/10.1007/s11886-021-01509-z. |
| [5] |
Mankad R, Herrmann J. Cardiac tumors: echo assessment. Echo Research and Practice. 2016; 3: R65–R77. https://doi.org/10.1530/ERP-16-0035. |
| [6] |
Butany J, Nair V, Naseemuddin A, Nair GM, Catton C, Yau T. Cardiac tumours: diagnosis and management. The Lancet. Oncology. 2005; 6: 219–228. https://doi.org/10.1016/S1470-2045(05)70093-0. |
| [7] |
Tyebally S, Chen D, Bhattacharyya S, Mughrabi A, Hussain Z, Manisty C, et al. Cardiac Tumors: JACC CardioOncology State-of-the-Art Review. JACC. CardioOncology. 2020; 2: 293–311. https://doi.org/10.1016/j.jaccao.2020.05.009. |
| [8] |
Haider I, Ullah H, Fatima M, Karim MS, Haq FU, Majid A, et al. Tissue characterization of benign cardiac tumors by cardiac magnetic resonance imaging, a review of core imaging protocol and benign cardiac tumors. Frontiers in Cardiovascular Medicine. 2023; 10: 1009411. https://doi.org/10.3389/fcvm.2023.1009411. |
| [9] |
Prabhu N, DeCara JM. Imaging Cardiac Masses in Patients with Cancer. The Israel Medical Association Journal: IMAJ. 2022; 24: 186–190. |
| [10] |
Kurmann R, El-Am E, Ahmad A, Abbasi MA, Mazur P, Akiki E, et al. Cardiac Masses Discovered by Echocardiogram; What to Do Next? Structural Heart: the Journal of the Heart Team. 2023; 7: 100154. https://doi.org/10.1016/j.shj.2022.100154. |
| [11] |
Mousavi N, Cheezum MK, Aghayev A, Padera R, Vita T, Steigner M, et al. Assessment of Cardiac Masses by Cardiac Magnetic Resonance Imaging: Histological Correlation and Clinical Outcomes. Journal of the American Heart Association. 2019; 8: e007829. https://doi.org/10.1161/JAHA.117.007829. |
| [12] |
Lyon AR, López-Fernández T, Couch LS, Asteggiano R, Aznar MC, Bergler-Klein J, et al. 2022 ESC Guidelines on cardio-oncology developed in collaboration with the European Hematology Association (EHA), the European Society for Therapeutic Radiology and Oncology (ESTRO) and the International Cardio-Oncology Society (IC-OS). European Heart Journal. 2022; 43: 4229–4361. https://doi.org/10.1093/eurheartj/ehac244. |
| [13] |
L’Angiocola PD, Donati R. Cardiac Masses in Echocardiography: A Pragmatic Review. Journal of Cardiovascular Echography. 2020; 30: 5–14. https://doi.org/10.4103/jcecho.jcecho_2_20. |
| [14] |
Wang Q, Wang B, Zhang X, Zhong X, Chang S, Yang J, et al. The usefulness of contrast echocardiography in the evaluation of cardiac masses: a multicenter study. BMC Cardiovascular Disorders. 2024; 24: 43. https://doi.org/10.1186/s12872-024-03708-2. |
| [15] |
Paolisso P, Foà A, Bergamaschi L, Graziosi M, Rinaldi A, Magnani I, et al. Echocardiographic Markers in the Diagnosis of Cardiac Masses. Journal of the American Society of Echocardiography: Official Publication of the American Society of Echocardiography. 2023; 36: 464–473.e2. https://doi.org/10.1016/j.echo.2022.12.022. |
| [16] |
Gatti M, D’Angelo T, Muscogiuri G, Dell’aversana S, Andreis A, Carisio A, et al. Cardiovascular magnetic resonance of cardiac tumors and masses. World Journal of Cardiology. 2021; 13: 628–649. https://doi.org/10.4330/wjc.v13.i11.628. |
| [17] |
Bonnes J, Brink M, Nijveldt R. How to evaluate cardiac masses by cardiovascular magnetic resonance parametric mapping? European Heart Journal. Cardiovascular Imaging. 2023; 24: 1605–1607. https://doi.org/10.1093/ehjci/jead221. |
| [18] |
Pazos-López P, Pozo E, Siqueira ME, García-Lunar I, Cham M, Jacobi A, et al. Value of CMR for the differential diagnosis of cardiac masses. JACC. Cardiovascular Imaging. 2014; 7: 896–905. https://doi.org/10.1016/j.jcmg.2014.05.009. |
| [19] |
Lavall D, Vosshage NH, Geßner R, Stöbe S, Ebel S, Denecke T, et al. Native T1 mapping for the diagnosis of cardiac amyloidosis in patients with left ventricular hypertrophy. Clinical Research in Cardiology: Official Journal of the German Cardiac Society. 2023; 112: 334–342. https://doi.org/10.1007/s00392-022-02005-2. |
| [20] |
Grazzini G, Pradella S, Rossi A, Basile RP, Ruggieri M, Galli D, et al. Practical Guide to Interpreting Cardiac Magnetic Resonance in Patients with Cardiac Masses. Journal of Cardiovascular Development and Disease. 2023; 10: 229. https://doi.org/10.3390/jcdd10060229. |
| [21] |
Joudar I, Nasri S, Kamaoui I, Skiker I. Cardiac computed tomography for cardiac masses: a necessity or a luxury? The Journal of International Medical Research. 2025; 53: 3000605241306604. https://doi.org/10.1177/03000605241306604. |
| [22] |
Lopez-Mattei J, Yang EH, Baldassarre LA, Agha A, Blankstein R, Choi AD, et al. Cardiac computed tomographic imaging in cardio-oncology: An expert consensus document of the Society of Cardiovascular Computed Tomography (SCCT). Endorsed by the International Cardio-Oncology Society (ICOS). Journal of Cardiovascular Computed Tomography. 2023; 17: 66–83. https://doi.org/10.1016/j.jcct.2022.09.002. |
| [23] |
Eberhard M, Lisi C, Moser LJ, Mergen V, Klambauer K, Uçar E, et al. Advanced myocardial characterization and function with cardiac CT. The International Journal of Cardiovascular Imaging. 2024. https://doi.org/10.1007/s10554-024-03229-1. (online ahead of print) |
| [24] |
Hong YJ, Hur J, Han K, Im DJ, Suh YJ, Lee HJ, et al. Quantitative Analysis of a Whole Cardiac Mass Using Dual-Energy Computed Tomography: Comparison with Conventional Computed Tomography and Magnetic Resonance Imaging. Scientific Reports. 2018; 8: 15334. https://doi.org/10.1038/s41598-018-33635-0. |
| [25] |
Erbay MI, Manubolu VS, Stein-Merlob AF, Ferencik M, Mamas MA, Lopez-Mattei J, et al. Integration and Potential Applications of Cardiovascular Computed Tomography in Cardio-Oncology. Current Cardiology Reports. 2025; 27: 51. https://doi.org/10.1007/s11886-025-02206-x. |
| [26] |
D’Angelo EC, Paolisso P, Vitale G, Foà A, Bergamaschi L, Magnani I, et al. Diagnostic Accuracy of Cardiac Computed Tomography and 18-F Fluorodeoxyglucose Positron Emission Tomography in Cardiac Masses. JACC. Cardiovascular Imaging. 2020; 13: 2400–2411. https://doi.org/10.1016/j.jcmg.2020.03.021. |
| [27] |
Lopez-Mattei JC, Yang EH, Ferencik M, Baldassarre LA, Dent S, Budoff MJ. Cardiac Computed Tomography in Cardio-Oncology: JACC: CardioOncology Primer. JACC. CardioOncology. 2021; 3: 635–649. https://doi.org/10.1016/j.jaccao.2021.09.010. |
| [28] |
Kim K, Ko WS, Kim SJ. Diagnostic test accuracies of F-18 FDG PET for characterisation of cardiac masses compared to conventional imaging techniques: systematic review and meta-analysis. The British Journal of Radiology. 2022; 95: 20210263. https://doi.org/10.1259/bjr.20210263. |
| [29] |
Rizzo A, Viccaro V, Pavon AG, Leo LA, Treglia G. [18F]FDG PET imaging in the differentiation of cardiac masses: an updated systematic review and dual Meta-Analysis of diagnostic performance and parameter variability. European Journal of Nuclear Medicine and Molecular Imaging. 2025; 52: 3379–3394. https://doi.org/10.1007/s00259-025-07289-w. |
| [30] |
Lucinian YA, Martineau P, Abikhzer G, Harel F, Pelletier-Galarneau M. Novel tracers to assess myocardial inflammation with radionuclide imaging. Journal of Nuclear Cardiology: Official Publication of the American Society of Nuclear Cardiology. 2024; 42: 102012. https://doi.org/10.1016/j.nuclcard.2024.102012. |
| [31] |
Kazimierczyk R, Kaminski KA, Nekolla SG. Cardiac PET/MRI: Recent Developments and Future Aspects. Seminars in Nuclear Medicine. 2024; 54: 733–746. https://doi.org/10.1053/j.semnuclmed.2024.05.007. |
| [32] |
Lemasle M, Lavie Badie Y, Cariou E, Fournier P, Porterie J, Rousseau H, et al. Contribution and performance of multimodal imaging in the diagnosis and management of cardiac masses. The International Journal of Cardiovascular Imaging. 2020; 36: 971–981. https://doi.org/10.1007/s10554-020-01774-z. |
| [33] |
Angeli F, Bodega F, Bergamaschi L, Armillotta M, Amicone S, Canton L, et al. Multimodality Imaging in the Diagnostic Work-Up of Patients With Cardiac Masses: JACC: CardioOncology State-of-the-Art Review. JACC. CardioOncology. 2024; 6: 847–862. https://doi.org/10.1016/j.jaccao.2024.09.006. |
| [34] |
Wang TKM, Klein AL. Multi-Modality Cardiac Imaging for Pericardial Diseases: A Contemporary Review. Reviews in Cardiovascular Medicine. 2022; 23: 336. https://doi.org/10.31083/j.rcm2310336. |
| [35] |
De Garate E, Ghosh Dastidar A, Baritussio A, Scatteia A, Amadu A, Venuti G, et al. Clinical impact of cardiovascular magnetic resonance on the management of acutely hospitalised patients. Heart. 2016; 102: A4. https://doi.org/10.1136/heartjnl-2016-309668.7. |
| [36] |
Hegde VA, Biederman RW, Mikolich JR. Cardiovascular Magnetic Resonance Imaging-Incremental Value in a Series of 361 Patients Demonstrating Cost Savings and Clinical Benefits: An Outcome-Based Study. Clinical Medicine Insights. Cardiology. 2017; 11: 1179546817710026. https://doi.org/10.1177/1179546817710026. |
| [37] |
Kalapos A, Szabó L, Dohy Z, Kiss M, Merkely B, Gyires-Tóth B, et al. Automated T1 and T2 mapping segmentation on cardiovascular magnetic resonance imaging using deep learning. Frontiers in Cardiovascular Medicine. 2023; 10: 1147581. https://doi.org/10.3389/fcvm.2023.1147581. |
| [38] |
Gearhart A, Goto S, Deo RC, Powell AJ. An Automated View Classification Model for Pediatric Echocardiography Using Artificial Intelligence. Journal of the American Society of Echocardiography: Official Publication of the American Society of Echocardiography. 2022; 35: 1238–1246. https://doi.org/10.1016/j.echo.2022.08.009. |
| [39] |
Lee JW, Park CH, Im DJ, Lee KH, Kim TH, Han K, et al. CT-based radiomics signature for differentiation between cardiac tumors and a thrombi: a retrospective, multicenter study. Scientific Reports. 2022; 12: 8173. https://doi.org/10.1038/s41598-022-12229-x. |
| [40] |
Polidori T, De Santis D, Rucci C, Tremamunno G, Piccinni G, Pugliese L, et al. Radiomics applications in cardiac imaging: a comprehensive review. La Radiologia Medica. 2023; 128: 922–933. https://doi.org/10.1007/s11547-023-01658-x. |
/
| 〈 |
|
〉 |