Magnetic resonance angiography imaging of pulmonary embolism using agents with blood pool properties as an alternative to computed tomography to avoid radiation exposure

Artificial intelligence-based pulmonary embolism classification: Development and validation using real-world data

This paper presents an artificial intelligence-based classification model for the detection of pulmonary embolism in computed tomography angiography. The proposed model, developed from public data and validated on a large dataset from a tertiary hospital, uses a two-dimensional approach that integrates temporal series to classify each slice of the examination and make predictions at both slice and examination levels. The training process consists of two stages: first using a convolutional neural network InceptionResNet V2 and then a recurrent neural network long short-term memory model. This approach achieved an accuracy of 93% at the slice level and 77% at the examination level. External validation using a hospital dataset resulted in a precision of 86% for positive pulmonary embolism cases and 69% for negative pulmonary embolism cases. Notably, the model excels in excluding pulmonary embolism, achieving a precision of 73% and a recall of 82%, emphasizing its clinical value in reducing unnecessary interventions. In addition, the diverse demographic distribution in the validation dataset strengthens the model's generalizability. Overall, this model offers promising potential for accurate detection and exclusion of pulmonary embolism, potentially streamlining diagnosis and improving patient outcomes.

MR Angiography of Pulmonary Vasculature

Pulmonary MR angiography (MRA) is a useful alternative to computed tomographic angiography (CTA) for the study of the pulmonary vasculature. For pulmonary hypertension and partial anomalous pulmonary venous return, a cardiac MR imaging and the pulmonary MRA are useful for flow quantification and planning treatment. For the diagnosis of pulmonary embolism (PE), MRA-PE has been shown to have non-inferior outcomes at 6 months when compared with CTA-PE. Over the last 15 years, pulmonary MRA has become a routine and reliable examination for the workup of pulmonary hypertension and the primary diagnosis of PE at the University of Wisconsin.

18F-PEG1-Vinyl Sulfone-Labeled Red Blood Cells as Positron Emission Tomography Agent to Image Intra-Abdominal Bleeding

¹⁸F-Labeled blood pool agents (BPAs) have attracted great attention for identifying bleeding sites. However, many BPAs are not sufficiently evaluated partially due to the limitations of labeling methods. In our previous work, we noticed that ¹⁸F-PEG1-vinyl sulfone (¹⁸F-VS) could efficiently label red blood cells (RBCs) ex vivo and in situ. However, its application as BPA is not fully evaluated. In this study, we systematically explored the feasibility of using ¹⁸F-VS-labeled RBCs as a positron emission tomography (PET) BPA for intra-abdominal bleeding diagnosis. In brief, we first optimized the labeling conditions, which lead to an 80% labeling yield of RBCs after incubating with ¹⁸F-VS in phosphate-buffered saline (PBS) at 37°C for 20 min. ¹⁸F-VS-labeled RBCs were found to be stable in vitro, which could simplify its transportation/storage for in vivo applications. In normal rat PET study, the cardiovascular system could be clearly imaged up to 5 h post injection (p.i.). An intra-abdominal hemorrhage rat model demonstrated that the ¹⁸F-VS-labeled RBCs clearly showed the dynamic changes of extravascular radioactivity due to intra-abdominal hemorrhage. Validation in the model of gastrointestinal bleeding clearly demonstrated the great potential of using ¹⁸F-VS-labeled RBCs as a BPA, which could be further evaluated in future studies.

Combined MR Imaging for Pulmonary Embolism and Deep Venous Thrombosis by Contrast-enhanced MR Volume Interpolated Body Examination

MR pulmonary angiography (MRPA) combined with indirect MR venography (MRV) was attempted by using 3D contrast-enhanced MR volume interpolated body examination (VIBE) sequence. Agreement rate for deep venous thrombosis (DVT) detection between MRV and duplex sonography (DUS) was evaluated; the potential of this method for venous thromoembolism (VTE) was also investigated. Thirty-four patients with DUS-identified DVT were enrolled in this study. MRI was performed after a single administration of Gadopentetate dimeglumine. Fat-suppressed 3D VIBE was applied for visualizing pulmonary arteries, abdominal veins, pelvic and leg veins, ranging from lung apex to ankle level. Two radiologists observed the MR images in consensus, recorded the location and number of emboli. MRV images were assessed based on per-vein segment. The agreement rate between MRV and DUS for venous segment-to-segment comparison was analyzed by Wilcoxon rank sum test. All the patients were diagnosed as having DVT by MRV. MRV detected 55 more venous segments with thrombi than DUS based on per-vein segment analysis. Twenty-three patients with pulmonary embolism (PE) were detected by MRPA. Twenty-one patients underwent both pulmonary CT angiography and MRPA, and consistency for PE detection was 100%. Total examination time of the combined MR protocol was 7 min for each patient. The contrast-enhanced VIBE sequence proves to be a feasible and reliable method for VTE diagnosis in one-stop MR scanning procedure, and contrast-enhanced VIBE performs better to depict DVT than DUS on per-vein segment basis.

Fat-suppressed magnetic resonance volume interpolated examination for deep venous thrombosis compared with duplex sonography

The aim of the present study was to evaluate magnetic resonance venography (MRV) scanned by breath-hold volume interpolated body examination with spectral fat saturation (VIBE-fs), combined with Dixon fat-suppressed VIBE (VIBE-Dixon) by using a 1.5T MR scanner for detecting deep venous thrombosis (DVT) compared with duplex sonography. A total of 31 patients with DVT were identified using duplex sonography and were enrolled in the present study for MRV examination, from the inferior vena cava to the ankle level after injection of gadopentetate dimeglumine. Venous segment-to-segment comparison was assessed for DVT detection between MRV and duplex sonography. A total of two radiologists separately performed subjective image quality assessment using a 5-point scale. Cohen's κ coefficient, Wilcoxon rank sum test and intraclass correlation coefficient values were used for statistical analysis. Of the 303 evaluated vein segments, duplex sonography identified 119 (39.3%; 119/303) venous segments with thrombus, while MRV detected 170 (56.1%; 170/303) venous segments with thrombus. The diagnostic agreement rate of DVT between duplex sonography and MRV was poor in the deep femoral vein and anterior tibial veins, while it was excellent in the inferior vena cava (IVC), common iliac vein, external iliac vein, femoral vein, popliteal vein, posterior tibial veins and peroneal veins. In addition, poor reliability was detected in the deep femoral vein, anterior tibial veins and peroneal veins, but good to excellent reliability was observed in IVC, common iliac vein, external iliac vein, femoral vein, popliteal vein and posterior tibial veins. Furthermore, image quality scores of each venous segment between the two radiologists indicated no statistical difference. Therefore, MRV scanned using VIBE-fs for the suprainguinal and VIBE-Dixon for the infrainguinal region may be a useful method for detecting DVT compared with duplex sonography. The results of present study proved this MR protocol to be a beneficial alternative imaging modality for the detection of DVT when duplex sonography is inadequate or not able to be performed.

Advances in Contrast Agents for Contrast-Enhanced Magnetic Resonance Imaging

INTRODUCTION

Magnetic resonance imaging (MRI) is a well-established medical invention in modern medical technology diagnosis. It is a nondestructive, versatile, and sensitive technique with a high spatial resolution for medical diagnosis. However, MRI has some limitations in differentiating certain tissues, particularly tiny blood vessels, pathological to healthy tissues, specific tumors, and inflammatory conditions such as arthritis, atherosclerosis, and multiple sclerosis. The contrast agent (CA) assisted imaging is the best possible solution to resolve the limitations of MRI.

METHOD

The literature review was carried out using the keywords, "MRI, T1&T2 relaxation, MRI CAs, delivery and adverse effects, classification of CAs." The tools used for the literature search were PubMed, Scopus, and Google Scholar.

RESULT AND DISCUSSION

The literature findings focus on MRI technique, limitations, and possible solutions. Primarily, the review focuses on the mechanism of CAs in image formation with detailed explanations of T1 and T2 relaxations, the mechanism of the MRI-CA image formations. This review presents the adverse effects of CA as well as available marketed formulations and recent patents to extent complete information about the MRI-CA.

CONCLUSION

MRI generates detailed visual information of various tissues with high resolution and contrast. The proton present in the biological fluid plays a crucial role in MR image formation, and it is unable to distinguish pathological conditions in many cases. The CAs are the best solution to resolve the limitation by interacting with native protons. The present review discusses the mechanism of CAs in contrast enhancement and its broad classification with the latest literature. Furthermore, the article presents information about CA biodistribution and adverse effects. The review concludes with an appropriate solution for adverse effects and presents the future prospective for researchers to develop advanced formulations.