Magnetic Resonance Imaging Angiography Research Articles

Thoracic aortic pseudoaneurysm: Inside its pathophysiology.

Our aim was to investigate pathophysiology of pseudoaneurysm of the thoracic aorta, an acute or chronic pathology, respectively, secondary to blunt thoracic trauma and aortitis, or complicating a deep penetrating aortic ulcer, intraparietal hematoma, aortic aneurysm, and even aortic graft, often with atherosclerosis as a common background. Given the relative rarity of this disease, an "inductive" retrospective method made it possible to retrieve clinical, radiological, and histopathological elements, which were mutually compared and validated through a "deductive" process of reinterpretation. We have identified three main structural constituents of this disease: a cavity, a single blood entry port, communicating with the aortic lumen, and a pseudocapsule. It is often caused by a chronic degenerative pathology of the intima and medial layers of the aorta, typically involving elastic fibers and smooth muscle cells, with possible intermediate stages of deep aortic ulcer or intraparietal hematoma. Otherwise, the acute onset may be secondary to acute aortitis or aortic injury. Today, thanks to the current angiographic tools represented by 3-D high resolution multidetector CT and MRI angiography, the diagnosis of thoracic aortic pseudoaneurysm is easier, as well as its surgical indications.

Advances in the diagnosis of multiorgan involvement in systemic sclerosis: a focus on MRI.

Systemic sclerosis (SSc) is a rare chronic multisystem autoimmune disease characterized by endothelial dysfunction, tissue hypoxia, and diffuse organ fibrosis. MRI provides a radiation free approach to noninvasively assess the key manifestations of SSc in multiple organs. The purpose of this review is to summarize recent advances in MRI techniques to provide diagnostic and prognostic information in patients with SSc. MRI can probe processes that play a key role in the development of SSc-related complications, including neointima proliferation, fibrosis, and hypoxia. Feature tracking and parametric mapping MRI can detect cardiac involvement at the subclinical level. Contrast-free MRI angiography with Digital Artery Volume Index (DAVIX) assessment allow comprehensive assessment of hand involvement. T1 mapping and BOLD imaging can assess SSc effects on skeletal muscle, and lung MRI is becoming a key method for imaging of interstitial lung disease. As a new exciting application, the sodium content of the skin can be quantified by 23Na MRI reflective of glycosaminoglycan content. Recent advances in MRI provide a unique opportunity to study the key pathophysiologic processes and clinical manifestations of SSc in multiple organs noninvasively, which can pave the way for the development of effective therapies.

Imaging of Carotid Stenosis: Where Are We Standing? Comparison of Multiparametric Ultrasound, CT Angiography, and MRI Angiography, with Recent Developments.

Atherosclerotic disease of the carotid arteries is a crucial risk factor in predicting the likelihood of future stroke events. In addition, emerging studies suggest that carotid stenosis may also be an indicator of plaque load on coronary arteries and thus have a correlation with the risk of acute cardiovascular events. Furthermore, although in symptomatic patients the degree of stenosis is the main morphological parameter studied, recent evidence suggests, especially in asymptomatic patients, that plaque vulnerability should also be evaluated as an emerging and significant imaging parameter. The reference diagnostic methods for the evaluation of carotid stenosis are currently ultrasonography, magnetic resonance imaging (MRI), and computed tomography angiography (CTA). In addition, other more invasive methods such as 123I-metaiodobenzylguanidine (MIBG) scintigraphy and PET-CT, as well as digital subtraction angiography, can be used. Each method has advantages and disadvantages, and there is often some confusion in their use. For example, the usefulness of MRI is often underestimated. In addition, implementations for each method have been developed over the years and are already enabling a significant increase in diagnostic accuracy. The purpose of our study is to make an in-depth analysis of all the methods in use and in particular their role in the diagnostic procedure of carotid stenosis, also discussing new technologies.

Perfusion parameter map generation from TOF-MRA in stroke using generative adversarial networks

PurposeTo generate perfusion parameter maps from Time-of-flight magnetic resonance angiography (TOF-MRA) images using artificial intelligence to provide an alternative to traditional perfusion imaging techniques. Materials and methodsThis retrospective study included a total of 272 patients with cerebrovascular diseases; 200 with acute stroke (from 2010 to 2018), and 72 with steno-occlusive disease (from 2011 to 2014). For each patient the TOF MRA image and the corresponding Dynamic susceptibility contrast magnetic resonance imaging (DSC-MRI) were retrieved from the datasets. The authors propose an adapted generative adversarial network (GAN) architecture, 3D pix2pix GAN, that generates common perfusion maps (CBF, CBV, MTT, TTP, Tmax) from TOF-MRA images. The performance was evaluated by the structural similarity index measure (SSIM). For a subset of 20 patients from the acute stroke dataset, the Dice coefficient was calculated to measure the overlap between the generated and real hypoperfused lesions with a time-to-maximum (Tmax) > 6 s. ResultsThe GAN model exhibited high visual overlap and performance for all perfusion maps in both datasets: acute stroke (mean SSIM 0.88–0.92, mean PSNR 28.48–30.89, mean MAE 0.02–0.04 and mean NRMSE 0.14–0.37) and steno-occlusive disease patients (mean SSIM 0.83–0.98, mean PSNR 23.62–38.21, mean MAE 0.01–0.05 and mean NRMSE 0.03–0.15). For the overlap analysis for lesions with Tmax>6 s, the median Dice coefficient was 0.49. ConclusionOur AI model can successfully generate perfusion parameter maps from TOF-MRA images, paving the way for a non-invasive alternative for assessing cerebral hemodynamics in cerebrovascular disease patients. This method could impact the stratification of patients with cerebrovascular diseases. Our results warrant more extensive refinement and validation of the method.

Small vessel disease burden and prognosis of recent subcortical ischaemic stroke differ by parent artery atherosclerosis.

Parent artery atherosclerosis is an important aetiology of recent subcortical ischaemic stroke (RSIS). However, comparisons of RSIS with different degrees of parent artery atherosclerosis are lacking. Prospectively collected data from our multicentre cohort (all were tertiary centres) of the Stroke Imaging Package Study between 2015 and 2017 were retrospectively reviewed. The patients with RSIS defined as a single clinically relevant diffusion-weighted imaging positive lesion in the territory of lenticulostriate arteries were categorized into three subgroups: (1) normal middle cerebral artery (MCA) on magnetic resonance angiography and high-resolution magnetic resonance imaging (HR-MRI); (2) low-grade MCA atherosclerosis (normal or <50% stenosis on magnetic resonance angiography and with MCA plaques on HR-MRI); (3) steno-occlusive MCA atherosclerosis (stenosis ≥50% or occlusion). The primary outcome was 90-day functional dependence (modified Rankin Scale score >2). The clinical and imaging findings were compared between subgroups. A total of 239 patients (median age 60.0 [52.0-67.0] years, 72% male) were enrolled, including 140 with normal MCA, 64 with low-grade MCA atherosclerosis and 35 with steno-occlusive MCA atherosclerosis. Patients with steno-occlusive MCA atherosclerosis had the largest infarct volume. Low-grade MCA atherosclerosis was independently associated with cerebral microbleeding, more severe perivascular spaces in basal ganglia and higher total cerebral small vessel disease burden. Low-grade MCA atherosclerosis was an independent determinant of 90-day functional dependence (odds ratio 3.897; 95% confidence interval 1.309-11.604). Our study suggested RSIS with varying severity of parent artery atherosclerosis exhibits distinctive clinical and neuroimaging characteristics, with low-grade MCA atherosclerosis associating with higher cerebral small vessel disease burden and worse prognosis.

Investigation of distributed learning for automated lesion detection in head MR images

In this study, we investigated the application of distributed learning, including federated learning and cyclical weight transfer, in the development of computer-aided detection (CADe) software for (1) cerebral aneurysm detection in magnetic resonance (MR) angiography images and (2) brain metastasis detection in brain contrast-enhanced MR images. We used datasets collected from various institutions, scanner vendors, and magnetic field strengths for each target CADe software. We compared the performance of multiple strategies, including a centralized strategy, in which software development is conducted at a development institution after collecting de-identified data from multiple institutions. Our results showed that the performance of CADe software trained through distributed learning was equal to or better than that trained through the centralized strategy. However, the distributed learning strategies that achieved the highest performance depend on the target CADe software. Hence, distributed learning can become one of the strategies for CADe software development using data collected from multiple institutions.

Internal carotid artery stenosis and occlusion based on machine learning: A comparative analysis of high-resolution MRI and CT angiography

ObjectiveTo explore the clinical application value of the imaging model of carotid artery stenosis occlusion based on machine learning. MethodsA comprehensive analysis was conducted on the medical records, MRI, and CT angiography images of 188 patients diagnosed with severe carotid artery stenosis. Specifically, the training cohort comprised 131 patients, including 107 males and 24 females, with an average age of 68 ± 8 years. A distinct verification group included 57 patients, 50 males, and 7 females, averaging 67 ± 8 years old. For the study, the volume of interest was manually outlined slice by slice, following the contour of the carotid plaque on cross-sectional images. The Radiomics package of Python software was used to extract image omics features. Feature screening was performed using intra - and inter-group correlation coefficients, redundancy analysis, minimum absolute contraction, and selection operator (LASSO) regression analysis. ResultsFour image omics features were selected based on the training set to construct the prediction model. Among the six machine learning models, the logistic regression model showed high and stable diagnostic efficiency, with an AUC of 0.872, sensitivity of 100.0%, and specificity of 66.2% in the training set. The AUC of the validation set was 0.867, the sensitivity was 83.3%, and the specificity was 78.8%. The calibration curve and DCA show that the logistic regression model has a reasonable degree of calibration and clinical application value. ConclusionPredictive MRI and CT angiography models based on machine learning have their respective application values in risk stratification of patients with severe asymptomatic carotid stenosis.

Retrospective case series of vertebral artery injuries associated with cervical spine trauma

BackgroundVertebral artery injuries (VAI) can occur due to cervical spine trauma. VAI can prove a serious complication and potentially compromise vascular supply to the posterior aspect of the brain. Currently, there is a paucity of evidence with regards to incidence, management, and outcomes for these patients. The purpose of this study is to investigate and elucidate the incidence of VAI associated with cervical trauma at a national tertiary referral centre for spinal pathology, their respective management, and associated outcomes. MethodsA retrospective review was conducted from 2012 to 2021 to identify patients with VAI secondary to cervical spine trauma. Demographic, clinical, and radiological data was collected to identify common traits in injury characteristics and management. Results1013 spine patients presented to our institution across the 10-year period. 739/1013 (72.9 %) were trauma patients. 42/739 (5.7 %) were imaged for suspected VAI secondary to trauma. There were 14/739 (1.9 %) confirmed VAI. All patients had CT-angiography for diagnosis. Four of the confirmed VAI patients (28.6 %) had additional MR-angiography imaging. Right-side was the most common side of VAI injury (7/14; 50 %), followed by left (5/14; 35.7 %) and bilateral (2/14; 14.3 %) injuries. 8/14 (57.1 %) patients were prescribed anti-thrombotic therapy. Acute mortality within 3-months was noted to be 2/14 (14.3 %) and occurred at 49 days and 57 days respectively. ConclusionVAI associated with cervical spine injury is rare in occurrence. However, it can be associated with high morbidity and mortality. As such, a multi-disciplinary approach to care is integral to ensuring good outcomes in these patients.

Predicting impaired cerebrovascular reactivity and risk of hyperperfusion syndrome in carotid artery stenosis using BeamSAT magnetic resonance imaging

Pencil-beam presaturation (BeamSAT) magnetic resonance imaging (MRI) produces selective magnetic resonance angiography (MRA) images of specific arteries, including the unilateral internal carotid artery (ICA-selective MRA) or vertebral artery (VA-selective MRA). We evaluate the influence of flow pattern, visualized using BeamSAT MRI, on preoperative cerebral hemodynamic status and postoperative hyperperfusion syndrome (HPS). Patients undergoing carotid artery stenting or carotid endarterectomy were categorized into two groups to evaluate flow pattern. Patients with neither crossflow on BeamSAT MRI nor mismatch in middle cerebral artery (MCA) signal intensity between ICA-selective and conventional MRA were classified into Group I, comprising 29 patients. Group II included all other patients comprising 19 patients, who were suspected of experiencing changes in intracranial flow patterns. Cerebral blood flow and cerebrovascular reactivity (CVR) were assessed using single-photon emission computed tomography, and potential HPS symptoms were retrospectively assessed by chart review. Preoperative ipsilateral CVR was significantly lower in Group II than in Group I (18.0% ± 20.0% vs. 48.3% ± 19.5%; P < 0.0001). Group II showed significantly impaired CVR (odds ratio 17.7, 95% confidence interval 1.82–171; P = 0.013) in multivariate analysis. The partial areas under the curve of the BeamSAT logistic model (0.843) were significantly larger than those of the conventional logistic model (0.626) over the range of high sensitivity (0.6–1) (P = 0.04). The incidence of postoperative HPS symptoms was significantly higher in Group II than in Group I (8/19 vs. 1/29; P = 0.001). BeamSAT MRI may be a valuable and non-invasive tool for assessing cerebral hemodynamics and predicting postoperative HPS.

Shortened Cerebral Circulation Time Predicts Resistance to Obliteration in High-Flow Brain Arteriovenous Malformations After Stereotactic Radiosurgery.

Treatment selection for brain arteriovenous malformations (BAVMs) is complicated by BAVM size, location, and hemodynamics. Quantitative digital subtraction angiography is used to quantify the hemodynamic impact of BAVMs on cerebral circulation. This study investigated the association between cerebral circulation time and the complete obliteration (CO) rate of BAVMs after stereotactic radiosurgery (SRS). We analyzed the data of 143 patients who underwent SRS for BAVMs between January 2011 and December 2019 in our institute. Their pre-SRS magnetic resonance imaging and angiography images were analyzed to acquire BAVM characteristics and quantitative digital subtraction angiography parameters. Modified cerebral circulation time (mCCT) was defined as the time difference between the bolus arrival time of the ipsilateral cavernous internal carotid artery and that of the parietal vein, as determined from the lateral view of images obtained using digital subtraction angiography. Cox regression with hazard ratios and Kaplan-Meier analyses were conducted to determine the associations between the parameters and BAVM CO after SRS. Of the 143 patients, 101 (70.6%) achieved BAVM CO. According to the multivariate analyses, an increased mCCT (hazard ratio: 1.24, P = .041) was the independent factor associated with BAVM CO after adjustment for age, sex, hemorrhagic presentation, a BAVM volume of >5 cm3, and a margin dose of >18 Gy. Individuals with an mCCT of ≤2.32 s had a lower 36-month probability of BAVM CO than did those with an mCCT of >2.32 s (44.1% ± 6.8% vs 63.3% ± 5.6%, P = .034). The hemodynamic impact of high-flow BAVM demonstrated by a shortened mCCT is associated with a lower BAVM CO rate after SRS.

Multidetector computer tomography and magnetic resonance imaging of double superior mesenteric veins: A case report

BACKGROUND This study aimed to describe the findings of double superior mesenteric veins (SMVs), a rare anatomical variation, on multidetector computer tomography (MDCT) and magnetic resonance imaging (MRI) images. CASE SUMMARY We describe the case of a 34-year-old male, who underwent both MDC and MRI examinations of the upper abdomen because of liver cirrhosis. MDCT and MRI angiography images of the upper abdomen revealed an anatomic variation of the superior mesenteric vein (SMV), the double SMVs. CONCLUSION The double SMVs are a congenital abnormality without potential clinical manifestation. Physicians need to be aware of this anatomical variation during abdominal surgery to avoid iatrogenic injury.

Neglected Superior Ophthalmic Vein Enlargement before Delayed Symptom of Carotid-Cavernous Fistula in a Blowout Fracture: A Case Report and Literature Review.

Carotid-cavernous fistula (CCF) is a rare condition. However, it should be suspected when there are traumatic facial fractures, because if not diagnosed, it can lead to permanent damage such as blindness. Traumatic CCF often presents delayed symptoms, and delayed diagnosis without prompt treatment can lead to permanent injuries in optic and cranial nerves III, IV, V, and VI as well as intracranial hemorrhage. The routine initial modality for patients with suspected facial bone fractures is noncontrast computed tomography (CT) to identify any fracture lines and check for intracranial hemorrhage. We report a post-traumatic CCF case with a 4-day symptom delay, where left superior ophthalmic vein (SOV) enlargement was observed on the routine noncontrast facial CT with ipsilateral orbital wall fracture. When the patient first presented to the emergency room (ER), we did not detect vein enlargement on CT. Afterwards, the patient developed delayed symptoms of CCF and was readmitted to the ER. When we reanalyzed the first CT scan, an enlarged SOV was confirmed. The diagnosis was confirmed via magnetic resonance imaging angiography, and the patient was successfully treated with embolization of the fistula. Thus, we recommend reviewing ophthalmic vein enlargement that is readily identifiable through noncontrast CT for patients injured by craniofacial trauma to suspect the presence of delayed CCF at their initial presentation.

Artificial intelligence-based automatic nidus segmentation of cerebral arteriovenous malformation on time-of-flight magnetic resonance angiography

ObjectiveAccurate nidus segmentation and quantification have long been challenging but important tasks in the clinical management of Cerebral Arteriovenous Malformation (CAVM). However, there are still dilemmas in nidus segmentation, such as difficulty defining the demarcation of the nidus, observer-dependent variation and time consumption. The aim of this study isto develop an artificial intelligence model to automatically segment the nidus on Time-Of-Flight Magnetic Resonance Angiography (TOF-MRA) images. MethodsA total of 92patients with CAVM who underwent both TOF-MRA and DSA examinations were enrolled. Two neurosurgeonsmanually segmented the nidusonTOF-MRA images,which were regarded as theground-truth reference. AU-Net-basedAImodelwascreatedfor automatic nidus detectionand segmentationonTOF-MRA images. ResultsThe meannidus volumes of the AI segmentationmodeland the ground truthwere 5.427 ± 4.996 and 4.824 ± 4.567 mL,respectively. The meandifference in the nidus volume between the two groups was0.603 ± 1.514 mL,which wasnot statisticallysignificant (P = 0.693). The DSC,precision and recallofthe testset were 0.754 ± 0.074, 0.713 ± 0.102 and 0.816 ± 0.098, respectively. The linear correlation coefficient of the nidus volume betweenthesetwo groupswas 0.988, p < 0.001. ConclusionThe performance of the AI segmentationmodel is moderate consistent with that of manual segmentation. This AI model has great potential in clinical settings, such as preoperative planning, treatment efficacy evaluation, riskstratification and follow-up.

Review of the role of imaging in the diagnosis of priapism.

Imaging has a specific role in the diagnosis and management of priapism. The primary imaging modality is ultrasound with colour Doppler (CDUS) which can accurately assess the hemodynamics of the cavernosal arteries. This is particularly useful in equivocal cases and can help differentiate ischemic from non-ischemic priapism as well as confirm the presence and location of arterio-venous fistulae post penile trauma. Furthermore, CDUS is invaluable in the post treatment follow up of non-ischemic priapism. Contrast enhanced magnetic resonance imaging (MRI) can demonstrate the extent of cavernosal necrosis in ischemic priapism and in conjunction with computer tomography (CT) has an important role in excluding underlying malignancy. MRI and CT angiography are used to evaluate pudendal arterial anatomy, which can be extremely variable and aids in the management of non-ischemic priapism. In selected cases of non-ischemic priapism, catheter angiography and transcatheter embolization of arteriovenous fistulae is an effective treatment. This review will examine the specific roles of different imaging modalities in the subtypes of priapism as well as highlight some of the pitfalls encountered in imaging.

Deep Learning-Based High-Resolution Magnetic Resonance Angiography (MRA) Generation Model for 4D Time-Resolved Angiography with Interleaved Stochastic Trajectories (TWIST) MRA in Fast Stroke Imaging.

The purpose of this study is to improve the qualitative and quantitative image quality of the time-resolved angiography with interleaved stochastic trajectories technique (4D-TWIST-MRA) using deep neural network (DNN)-based MR image reconstruction software. A total of 520 consecutive patients underwent 4D-TWIST-MRA for ischemic stroke or intracranial vessel stenosis evaluation. Four-dimensional DNN-reconstructed MRA (4D-DNR) was generated using commercially available software (SwiftMR v.3.0.0.0, AIRS Medical, Seoul, Republic of Korea). Among those evaluated, 397 (76.3%) patients received concurrent time-of-flight MRA (TOF-MRA) to compare the signal-to-noise ratio (SNR), image quality, noise, sharpness, vascular conspicuity, and degree of venous contamination with a 5-point Likert scale. Two radiologists independently evaluated the detection rate of intracranial aneurysm in TOF-MRA, 4D-TWIST-MRA, and 4D-DNR in separate sessions. The other 123 (23.7%) patients received 4D-TWIST-MRA due to a suspicion of acute ischemic stroke. The confidence level and decision time for large vessel occlusion were evaluated in these patients. In qualitative analysis, 4D-DNR demonstrated better overall image quality, sharpness, vascular conspicuity, and noise reduction compared to 4D-TWIST-MRA. Moreover, 4D-DNR exhibited a higher SNR than 4D-TWIST-MRA. The venous contamination and aneurysm detection rates were not significantly different between the two MRA images. When compared to TOF-MRA, 4D-CE-MRA underestimated the aneurysm size (2.66 ± 0.51 vs. 1.75 ± 0.62, p = 0.029); however, 4D-DNR showed no significant difference in size compared to TOF-MRA (2.66 ± 0.51 vs. 2.10 ± 0.41, p = 0.327). In the diagnosis of large vessel occlusion, 4D-DNR showed a better confidence level and shorter decision time than 4D-TWIST-MRA. DNN reconstruction may improve the qualitative and quantitative image quality of 4D-TWIST-MRA, and also enhance diagnostic performance for intracranial aneurysm and large vessel occlusion.

Giant cavernous aneurysms occluded by aneurysmal thrombosis, calcification, parent artery occlusion: A case report and review of literature

BACKGROUND Patients with giant intracranial aneurysms (GIAs) are at a high risk of rupture, morbidity, and mortality even after surgical or endovascular treatment. We described a case of a spontaneously occluded GIA secondary to gradual growth of the GIA, continuously progressed aneurysmal thrombosis, complete aneurysmal calcification and complete occlusion of the parent artery-the right internal carotid artery (RICA). CASE SUMMARY A 72-year-old female patient complained of sudden pain in her right eye upon admission to our hospital. She had been diagnosed with a GIA [30 mm (axial) × 38 mm (coronal) × 28 mm (sagittal)] containing an aneurysmal thrombus located in the cavernous sinus segment of RICA diagnosed by magnetic resonance imaging (MRI), enhanced MRI, and magnetic resonance angiography more than 14 years ago. Later, with slow growth of the cavernous carotid GIA, aneurysmal thrombosis progressed continuously, spontaneous occlusion of the RICA, complete aneurysmal calcification, and occlusion of the GIA occurred gradually. She had no history of subarachnoid hemorrhage but missed the chance for endovascular therapy at an early stage. As a result, she was left with severe permanent sequelae from the injuries to the right cranial nerves II, III, IV, V1/V2, and VI. CONCLUSION The risk of rupture of the cavernous carotid GIAs was relatively low and possibly further be reduced by the stasis flow and spontaneous occlusion of the parent artery internal carotid artery (ICA) induced by the mass effect of the cavernous carotid GIAs and the extremely rare aneurysmal calcification. However, nowadays, it is advisable to recommend early endovascular treatment for the cavernous carotid GIAs to prevent injuries to the surrounding intracranial nerves and occlusion of the ICA, mainly caused by the mass effect of the cavernous carotid GIAs.

Anomalous Left Coronary Artery from the Pulmonary Artery (ALCAPA) Syndrome: Pathophysiology, Clinical Manifestations, and Management Strategies

Anomalous Left Coronary Artery from the Pulmonary Artery (ALCAPA) syndrome is a rare congenital cardiac anomaly characterized by the abnormal origin of the left coronary artery from the pulmonary artery instead of the aortic root. This anomaly results in compromised myocardial perfusion, particularly under conditions of increased oxygen demand, leading to myocardial ischemia, left ventricular dysfunction, and potentially severe complications such as mitral regurgitation, heart failure, and sudden cardiac death. The pathophysiology of ALCAPA involves a complex interplay of coronary artery steal phenomenon, collateral vessel development, and ventricular remodeling. Early diagnosis is critical, often achieved through echocardiography, cardiac MRI, or coronary angiography, which helps in planning surgical intervention. The mainstay of treatment is surgical correction to re-establish a dual coronary artery system, with techniques including direct reimplantation of the anomalous coronary artery into the aorta or the creation of an intrapulmonary tunnel. This article provides a comprehensive review of the current understanding of ALCAPA syndrome, emphasizing its pathophysiology, clinical presentation, diagnostic modalities, and surgical management, along with outcomes and prognostic factors.

Patient-specific cerebral 3D vessel model reconstruction using deep learning

Three-dimensional vessel model reconstruction from patient-specific magnetic resonance angiography (MRA) images often requires some manual maneuvers. This study aimed to establish the deep learning (DL)-based method for vessel model reconstruction. Time of flight MRA of 40 patients with internal carotid artery aneurysms was prepared, and three-dimensional vessel models were constructed using the threshold and region-growing method. Using those datasets, supervised deep learning using 2D U-net was performed to reconstruct 3D vessel models. The accuracy of the DL-based vessel segmentations was assessed using 20 MRA images outside the training dataset. The dice coefficient was used as the indicator of the model accuracy, and the blood flow simulation was performed using the DL-based vessel model. The created DL model could successfully reconstruct a three-dimensional model in all 60 cases. The dice coefficient in the test dataset was 0.859. Of note, the DL-generated model proved its efficacy even for large aneurysms (> 10 mm in their diameter). The reconstructed model was feasible in performing blood flow simulation to assist clinical decision-making. Our DL-based method could successfully reconstruct a three-dimensional vessel model with moderate accuracy. Future studies are warranted to exhibit that DL-based technology can promote medical image processing.Graphical

Synthesizing 4D Magnetic Resonance Angiography From 3D Time-of-Flight Using Deep Learning: A Feasibility Study.

Objective and background This study aimed to develop a deep convolutional neural network (DCNN) model capable of generating synthetic 4D magnetic resonance angiography (MRA) from 3D time-of-flight (TOF) images, allowing estimation of temporal changes in arterial flow. TOF MRA provides static information about arterial structures through maximum intensity projection (MIP) processing, but it does not capture the dynamic information of contrast agent circulation, which is lost during MIP processing. Considering the principles of TOF, it is hypothesized that dynamic information about arterial blood flow is latent within TOF signals. Although arterial spin labeling (ASL) can extract dynamic arterial information, ASL MRA has drawbacks, such as longer imaging times and lower spatial resolution than TOF MRA. This study's primary aim is to extend the utility of TOF MRA by training a machine-learning model on paired TOF and ASL data to extract latent dynamic information from TOF signals. Methods A DCNN combining a modified U-Net and a long-short-term memory (LSTM) network was trained on a dataset of 13 subjects (11 menand two women, aged 42-77 years) using paired 3D TOF MRA and 4D ASL MRA images. Subjects had no history of cerebral vessel occlusion or significant stenosis. The dataset was acquired using a 3T MRI system with a 32-channel head coil. Preprocessing involved resampling and intensity normalization of TOF and ASL images, followed by data augmentation and arterial mask generation. The model learned to extract flow information from TOF images and generate 8-phase 4D MRA images. The precision of flow estimation was evaluated using the coefficient of determination (R²) and Bland-Altman analysis. A board-certified neuroradiologist validated the quality of the images and the absence of significant stenosis in the major cerebral arteries. Results The generated 4D MRA images closely resembled the ground-truth ASL MRA data, with R² values of 0.92, 0.85, and 0.84 for the internal carotid artery (ICA), proximal middle cerebral artery (MCA), and distal MCA, respectively. Bland-Altman analysis revealed a systematic error of -0.06, with 95% agreement limits ranging from -0.18 to 0.12. Additionally, the model successfully identified flow abnormalities in a subject with left MCA stenosis, displaying a delayed peak and subsequent flattening distal to the stenosis, indicative of reduced blood flow. Visualization of the predicted arterial flow overlaid on the original TOF MRA images highlighted the spatial progression and dynamics of the flow. Conclusions The DCNN model effectively generated synthetic 4D MRA images from TOF images, demonstrating its potential to estimate temporal changes in arterial flow accurately. This non-invasive technique offers a promising alternative to conventional methods for visualizing and evaluating healthy and pathological flow dynamics. It has significant potential to improve the diagnosis and treatment of cerebrovascular diseases by providing detailed temporal flow information without the need for contrast agents or invasive procedures. The practical implementation of this model could enable the extraction of dynamic cerebral blood flow information from routine brain MRI examinations, contributing to the early diagnosis and management of cerebrovascular disorders.

Deep learning-based platform performs high detection sensitivity of intracranial aneurysms in 3D brain TOF-MRA: An external clinical validation study

PurposeTo evaluate the diagnostic efficacy of a developed artificial intelligence (AI) platform incorporating deep learning algorithms for the automated detection of intracranial aneurysms in time-of-flight (TOF) magnetic resonance angiography (MRA). MethodThis retrospective study encompassed 3D TOF MRA images acquired between January 2023 and June 2023, aiming to validate the presence of intracranial aneurysms via our developed AI platform. The manual segmentation results by experienced neuroradiologists served as the “gold standard”. Following annotation of MRA images by neuroradiologists using InferScholar software, the AI platform conducted automatic segmentation of intracranial aneurysms. Various metrics including accuracy (ACC), balanced ACC, area under the curve (AUC), sensitivity (SE), specificity (SP), F1 score, Brier Score, and Net Benefit were utilized to evaluate the generalization of AI platform. Comparison of intracranial aneurysm identification performance was conducted between the AI platform and six radiologists with experience ranging from 3 to 12 years in interpreting MR images. Additionally, a comparative analysis was carried out between radiologists’ detection performance based on independent visual diagnosis and AI-assisted diagnosis. Subgroup analyses were also performed based on the size and location of the aneurysms to explore factors impacting aneurysm detectability. Results510 patients were enrolled including 215 patients (42.16 %) with intracranial aneurysms and 295 patients (57.84 %) without aneurysms. Compared with six radiologists, the AI platform showed competitive discrimination power (AUC, 0.96), acceptable calibration (Brier Score loss, 0.08), and clinical utility (Net Benefit, 86.96 %). The AI platform demonstrated superior performance in detecting aneurysms with an overall SE, SP, ACC, balanced ACC, and F1 score of 91.63 %, 92.20 %, 91.96 %, 91.92 %, and 90.57 % respectively, outperforming the detectability of the two resident radiologists. For subgroup analysis based on aneurysm size and location, we observed that the SE of the AI platform for identifying tiny (diameter＜3mm), small (3 mm ≤ diameter＜5mm), medium (5 mm ≤ diameter＜7mm) and large aneurysms (diameter ≥ 7 mm) was 87.80 %, 93.14 %, 95.45 %, and 100 %, respectively. Furthermore, the SE for detecting aneurysms in the anterior circulation was higher than that in the posterior circulation. Utilizing the AI assistance, six radiologists (i.e., two residents, two attendings and two professors) achieved statistically significant improvements in mean SE (residents: 71.40 % vs. 88.37 %; attendings: 82.79 % vs. 93.26 %; professors: 90.07 % vs. 97.44 %; P < 0.05) and ACC (residents: 85.29 % vs. 94.12 %; attendings: 91.76 % vs. 97.06 %; professors: 95.29 % vs. 98.82 %; P < 0.05) while no statistically significant change was observed in SP. Overall, radiologists’ mean SE increased by 11.40 %, mean SP increased by 1.86 %, and mean ACC increased by 5.88 %, mean balanced ACC promoted by 6.63 %, mean F1 score grew by 7.89 %, and Net Benefit rose by 12.52 %, with a concurrent decrease in mean Brier score declined by 0.06. ConclusionsThe deep learning algorithms implemented in the AI platform effectively detected intracranial aneurysms on TOF-MRA and notably enhanced the diagnostic capabilities of radiologists. This indicates that the AI-based auxiliary diagnosis model can provide dependable and precise prediction to improve the diagnostic capacity of radiologists.