A combined deep-learning approach to fully automatic left ventricle segmentation in cardiac magnetic resonance imaging

2019 ◽  
Author(s):  
Ramon A. Moreno ◽  
Marina F. S. de Sá Rebelo ◽  
Talles Carvalho ◽  
Antonildes N. Assunção ◽  
Roberto N. Dantas ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Deep Learning ◽  
Cardiac Magnetic Resonance ◽  
Magnetic Resonance ◽  
Left Ventricle ◽  
Cardiac Magnetic Resonance Imaging ◽  
Learning Approach ◽  
Resonance Imaging ◽  
Fully Automatic ◽  
Ventricle Segmentation

scholarly journals P713 An incidental finding of a completely thrombosed left ventricular pseudoaneurysm: what is the next step?

2020 ◽  
Vol 21 (Supplement_1) ◽  
Author(s):  
S Istratoaie ◽  
A Iliescu ◽  
S Manole ◽  
R Beyer ◽  
D Tudoreanu ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Myocardial Infarction ◽  
Cardiac Magnetic Resonance ◽  
Magnetic Resonance ◽  
Left Ventricle ◽  
Cardiac Magnetic Resonance Imaging ◽  
Multimodality Imaging ◽  
Left Ventricular ◽  
Resonance Imaging ◽  
Left Ventricular Pseudoaneurysm

Abstract Introduction Left ventricular pseudoaneurysm is a rare complication of myocardial infarction. It is the result of ventricular rupture contained by the pericardial adhesions or thrombus. Although echocardiography is suitable as the initial method for diagnosis, multimodality imaging is often required in order to further characterize the pseudoaneurysm morphology and to plan the treatment. Case report A 56-year-old male patient with an old inferior myocardial infarction treated conservatively 6 years ago, was admitted in our department for atypical left laterothoracic pain. Three months before he had the same symptoms and an unexplained paracardiac mass was incidentally diagnosed by transthoracic echocardiography. At that time, he was evaluated by coronary angiography which showed no epicardial coronary artery stenosis. During admission, the ECG showed sinus rhythm, inferior myocardial scarring and right bundle branch block. The laboratory tests revealed cardiac enzymes within normal range, increased D-dimeri and elevated inflammatory markers. The echocardiography showed a nondilated left ventricle (LV) with preserved ejection fraction and akinesia of the inferolateral(IL) LV wall. Attached to the basal IL LV wall, an extensive mass was documented with an echogenic appearance and no color Doppler flow, suggesting a thrombosed pseudoaneurysm. A contrast enhanced computed tomography (CT) scan confirmed the diagnosis, but it was not able to establish whether the pseudoaneurysm was partially or completely thrombosed. For a more accurate morphologic and tissue characterization, a cardiac magnetic resonance imaging(CMR) was subsequently performed, that confirmed the presence of a completely thrombosed pseudoaneurysm, measuring 82x38mm. In this case, a conservative approach was initially suggested by the completely thrombosed chronic pseudoaneurysm (older than 3months and with no Doppler color flow). However, according to the literature a surgical approach should be considered when the pseudoaneurysm dimension is larger than 3 cm. Since the patient refused the surgical intervention, medical treatment was initiated with anticoagulants due to the high embolic risk, betablockers and angiotensin-receptor antagonists to maintain the blood pressure less than 120/80mmHg. At 1 month, his condition was stable. He will be reevaluated in 3 months, to monitor the possible pseudoaneurysm expansion. According to our knowledge, this is the first case of a completely thrombosed pseudoaneurysm described in the literature. Its echocardiographic, CT and CMR appearance is important for the differential diagnosis of all paracardiac masses (tumors, hiatus hernias, etc). Conclusion Completely thrombosed left ventricle pseudoaneurysm remains a challenging diagnosis since its echocardiographic appearance is atypical. Cardiac magnetic resonance imaging has a higher diagnostic yield and can provide important information that may influence the course of treatment. Abstract P713 Figure. LV Pseudoaneurysm-multimodality imaging

Sign in / Sign up

Export Citation Format

Share Document