Automated detection of steno-occlusive lesion on time-of-flight magnetic resonance angiography: an observer performance study.

BACKGROUND AND PURPOSE: Intracranial steno-occlusive lesions are responsible for acute ischemic stroke. However, the clinical benefits of artificial intelligence-based methods for detecting pathologic lesions in intracranial arteries have not been evaluated. We aimed to validate the clinical utility of an artificial intelligence model for detecting steno-occlusive lesions in the intracranial arteries. MATERIALS AND METHODS: Overall, 138 TOF-MRA images were collected from two institutions, which served as internal (n = 62) and external (n = 76) test sets, respectively. Each study was reviewed by five radiologists (two neuroradiologists and three radiology residents) to compare the usage and non-usage of our proposed artificial intelligence model for TOF-MRA interpretation. They identified the steno-occlusive lesions and recorded their reading time. Observer performance was assessed using the area under the Jackknife free-response receiver operating characteristic curve and reading time for comparison. RESULTS: The average area under the Jackknife free-response receiver operating characteristic curve for the five radiologists demonstrated an improvement from 0.70 without artificial intelligence to 0.76 with artificial intelligence (P = .027). Notably, this improvement was most pronounced among the three radiology residents, whose performance metrics increased from 0.68 to 0.76 (P = .002). Despite an increased reading time upon using artificial intelligence, there was no significant change among the readings by radiology residents. Moreover, the use of artificial intelligence resulted in improved inter-observer agreement among the reviewers (the intraclass correlation coefficient increased from 0.734 to 0.752). CONCLUSIONS: Our proposed artificial intelligence model offers a supportive tool for radiologists, potentially enhancing the accuracy of detecting intracranial steno-occlusion lesions on TOF-MRA. Less-experienced readers may benefit the most from this model.ABBREVIATIONS: AI = Artificial intelligence; AUC = Area under the receiver operating characteristic curve; AUFROC = Area under the Jackknife free-response receiver operating characteristic curve; DL = Deep learning; ICC = Intraclass correlation coefficient; IRB = Institutional Review Boards; JAFROC = Jackknife free-response receiver operating characteristic.

Risk Prediction of Emergency Department Visits in Patients With Lung Cancer Using Machine Learning: Retrospective Observational Study.

BACKGROUND: Patients with lung cancer are among the most frequent visitors to emergency departments due to cancer-related problems, and the prognosis for those who seek emergency care is dismal. Given that patients with lung cancer frequently visit health care facilities for treatment or follow-up, the ability to predict emergency department visits based on clinical information gleaned from their routine visits would enhance hospital resource utilization and patient outcomes. OBJECTIVE: This study proposed a machine learning-based prediction model to identify risk factors for emergency department visits by patients with lung cancer. METHODS: This was a retrospective observational study of patients with lung cancer diagnosed at Seoul National University Bundang Hospital, a tertiary general hospital in South Korea, between January 2010 and December 2017. The primary outcome was an emergency department visit within 30 days of an outpatient visit. This study developed a machine learning-based prediction model using a common data model. In addition, the importance of features that influenced the decision-making of the model output was analyzed to identify significant clinical factors. RESULTS: The model with the best performance demonstrated an area under the receiver operating characteristic curve of 0.73 in its ability to predict the attendance of patients with lung cancer in emergency departments. The frequency of recent visits to the emergency department and several laboratory test results that are typically collected during cancer treatment follow-up visits were revealed as influencing factors for the model output. CONCLUSIONS: This study developed a machine learning-based risk prediction model using a common data model and identified influencing factors for emergency department visits by patients with lung cancer. The predictive model contributes to the efficiency of resource utilization and health care service quality by facilitating the identification and early intervention of high-risk patients. This study demonstrated the possibility of collaborative research among different institutions using the common data model for precision medicine in lung cancer.

Automated detection of intracranial aneurysms using skeleton-based 3D patches, semantic segmentation, and auxiliary classification for overcoming data imbalance in brain TOF-MRA.

Accurate and reliable detection of intracranial aneurysms is vital for subsequent treatment to prevent bleeding. However, the detection of intracranial aneurysms can be time-consuming and even challenging, and there is great variability among experts, especially in the case of small aneurysms. This study aimed to detect intracranial aneurysms accurately using a convolutional neural network (CNN) with 3D time-of-flight magnetic resonance angiography (TOF-MRA). A total of 154 3D TOF-MRA datasets with intracranial aneurysms were acquired, and the gold standards were manually drawn by neuroradiologists. We also obtained 113 subjects from a public dataset for external validation. These angiograms were pre-processed by using skull-stripping, signal intensity normalization, and N4 bias correction. The 3D patches along the vessel skeleton from MRA were extracted. Values of the ratio between the aneurysmal and the normal patches ranged from 1:1 to 1:5. The semantic segmentation on intracranial aneurysms was trained using a 3D U-Net with an auxiliary classifier to overcome the imbalance in patches. The proposed method achieved an accuracy of 0.910 in internal validation and external validation accuracy of 0.883 with a 2:1 ratio of normal to aneurysmal patches. This multi-task learning method showed that the aneurysm segmentation performance was sufficient to be helpful in an actual clinical setting.

A Nationwide Web-Based Survey of Neuroradiologists' Perceptions of Artificial Intelligence Software for Neuro-Applications in Korea.

OBJECTIVE: We aimed to investigate current expectations and clinical adoption of artificial intelligence (AI) software among neuroradiologists in Korea. MATERIALS AND METHODS: In April 2022, a 30-item online survey was conducted by neuroradiologists from the Korean Society of Neuroradiology (KSNR) to assess current user experiences, perceptions, attitudes, and future expectations regarding AI for neuro-applications. Respondents with experience in AI software were further investigated in terms of the number and type of software used, period of use, clinical usefulness, and future scope. Results were compared between respondents with and without experience with AI software through multivariable logistic regression and mediation analyses. RESULTS: The survey was completed by 73 respondents, accounting for 21.9% (73/334) of the KSNR members; 72.6% (53/73) were familiar with AI and 58.9% (43/73) had used AI software, with approximately 86% (37/43) using 1-3 AI software programs and 51.2% (22/43) having up to one year of experience with AI software. Among AI software types, brain volumetry software was the most common (62.8% [27/43]). Although 52.1% (38/73) assumed that AI is currently useful in practice, 86.3% (63/73) expected it to be useful for clinical practice within 10 years. The main expected benefits were reducing the time spent on repetitive tasks (91.8% [67/73]) and improving reading accuracy and reducing errors (72.6% [53/73]). Those who experienced AI software were more familiar with AI (adjusted odds ratio, 7.1 [95% confidence interval, 1.81-27.81]; P = 0.005). More than half of the respondents with AI software experience (55.8% [24/43]) agreed that AI should be included in training curriculums, while almost all (95.3% [41/43]) believed that radiologists should coordinate to improve its performance. CONCLUSION: A majority of respondents experienced AI software and showed a proactive attitude toward adopting AI in clinical practice, suggesting that AI should be incorporated into training and active participation in AI development should be encouraged.

Application of symmetry evaluation to deep learning algorithm in detection of mastoiditis on mastoid radiographs.

As many human organs exist in pairs or have symmetric appearance and loss of symmetry may indicate pathology, symmetry evaluation on medical images is very important and has been routinely performed in diagnosis of diseases and pretreatment evaluation. Therefore, applying symmetry evaluation function to deep learning algorithms in interpreting medical images is essential, especially for the organs that have significant inter-individual variation but bilateral symmetry in a person, such as mastoid air cells. In this study, we developed a deep learning algorithm to detect bilateral mastoid abnormalities simultaneously on mastoid anterior-posterior (AP) views with symmetry evaluation. The developed algorithm showed better diagnostic performance in diagnosing mastoiditis on mastoid AP views than the algorithm trained by single-side mastoid radiographs without symmetry evaluation and similar to superior diagnostic performance to head and neck radiologists. The results of this study show the possibility of evaluating symmetry in medical images with deep learning algorithms.

Intracranial steno-occlusive lesion detection on time-of-flight MR angiography using multi-task learning.

Steno-occlusive lesions in intracranial arteries refer to segments of narrowed or occluded blood vessels that increase the risk of ischemic strokes. Steno-occlusive lesion detection is crucial in clinical settings; however, automatic detection methods have hardly been studied. Therefore, we propose a novel automatic method to detect steno-occlusive lesions in sequential transverse slices on time-of-flight magnetic resonance angiography. Our method simultaneously detects lesions while segmenting blood vessels based on end-to-end multi-task learning, reflecting that the lesions are closely related to the connectivity of blood vessels. We design classification and localization modules that can be attached to arbitrary segmentation network. As blood vessels are segmented, both modules simultaneously predict the presence and location of lesions for each transverse slice. By combining outputs from the two modules, we devise a simple operation that boosts the performance of lesion localization. Experimental results show that lesion prediction and localization performance is improved by incorporating blood vessel extraction. Our ablation study demonstrates that the proposed operation enhances lesion localization accuracy. We also verify the effectiveness of multi-task learning by comparing our approach with those that individually detect lesions with extracted blood vessels.

Long-term outcomes and quantitative radiologic analysis of extracranial-intracranial bypass for hemodynamically compromised chronic large artery occlusive disease.

This study aimed to demonstrate the effectiveness of nonemergent extracranial-to-intracranial bypass (EIB) in symptomatic chronic large artery atherosclerotic stenosis or occlusive disease (LAA) through quantitative analysis of computed tomography perfusion (CTP) parameters using RAPID software. We retrospectively analyzed 86 patients who underwent nonemergent EIB due to symptomatic chronic LAA. CTP data obtained preoperatively, immediately postoperatively (PostOp0), and 6 months postoperatively (PostOp6M) after EIB were quantitatively analyzed through RAPID software, and their association with intraoperative bypass flow (BF) was assessed. The clinical outcomes, including neurologic state, incidence of recurrent infarction and complications, were also analyzed. The time-to-maximum (Tmax) > 8 s, > 6 s and > 4 s volumes decreased significantly at PostOp0 and up through PostOp6M (preoperative, 5, 51, and 223 ml (median), respectively; PostOp0, 0, 20.25, and 143 ml, respectively; PostOp6M, 0, 7.5, and 148.5 ml, respectively; p < 0.001, p < 0.001, and p < 0.001, respectively). The postoperative improvement in the Tmax > 6 s and > 4 s volumes was significantly correlated with the BF at PostOp0 and PostOp6M (PostOp0, r = 0.367 (p = 0.001) and r = 0.275 (p = 0.015), respectively; PostOp6M r = 0.511 (p < 0.001) and r = 0.391 (p = 0.001), respectively). The incidence of recurrent cerebral infarction was 4.7%, and there were no major complications that produced permanent neurological impairment. Nonemergent EIB under strict operation indications can be a feasible treatment for symptomatic, hemodynamically compromised LAA patients.

Prediction of hemorrhagic cerebral hyperperfusion syndrome after direct bypass surgery in adult nonhemorrhagic moyamoya disease: combining quantitative parameters on RAPID perfusion CT with clinically related factors.

OBJECTIVE: The aim of this study was to identify predictive factors for hemorrhagic cerebral hyperperfusion syndrome (hCHS) after direct bypass surgery in adult nonhemorrhagic moyamoya disease (non-hMMD) using quantitative parameters on rapid processing of perfusion and diffusion (RAPID) perfusion CT software. METHODS: A total of 277 hemispheres in 223 patients with non-hMMD who underwent combined bypass were retrospectively reviewed. Preoperative volumes of time to maximum (Tmax) > 4 seconds and > 6 seconds were obtained from RAPID analysis of perfusion CT. These quantitative parameters, along with other clinical and angiographic factors, were statistically analyzed to determine the significant predictors for hCHS after bypass surgery. RESULTS: Intra- or postoperative hCHS occurred in 13 hemispheres (4.7%). In 7 hemispheres, subarachnoid hemorrhage occurred intraoperatively, and in 6 hemispheres, intracerebral hemorrhage was detected postoperatively. All hCHS occurred within the 4 days after bypass. Advanced age (OR 1.096, 95% CI 1.039-1.163, p = 0.001) and a large volume of Tmax > 6 seconds (OR 1.011, 95% CI 1.004-1.018, p = 0.002) were statistically significant factors in predicting the risk of hCHS after surgery. The cutoff values of patient age and volume of Tmax > 6 seconds were 43.5 years old (area under the curve [AUC] 0.761) and 80.5 ml (AUC 0.762), respectively. CONCLUSIONS: In adult patients with non-hMMD older than 43.5 years or with a large volume of Tmax > 6 seconds over 80.5 ml, more prudence is required in the decision to undergo bypass surgery and in postoperative management.

Deep learning-based diagnosis of Alzheimer's disease using brain magnetic resonance images: an empirical study.

The limited accessibility of medical specialists for Alzheimer's disease (AD) can make obtaining an accurate diagnosis in a timely manner challenging and may influence prognosis. We investigated whether VUNO Med-DeepBrain AD (DBAD) using a deep learning algorithm can be employed as a decision support service for the diagnosis of AD. This study included 98 elderly participants aged 60 years or older who visited the Seoul Asan Medical Center and the Korea Veterans Health Service. We administered a standard diagnostic assessment for diagnosing AD. DBAD and three panels of medical experts (ME) diagnosed participants with normal cognition (NC) or AD using T1-weighted magnetic resonance imaging. The accuracy (87.1% for DBAD and 84.3% for ME), sensitivity (93.3% for DBAD and 80.0% for ME), and specificity (85.5% for DBAD and 85.5% for ME) of both DBAD and ME for diagnosing AD were comparable; however, DBAD showed a higher trend in every analysis than ME diagnosis. DBAD may support the clinical decisions of physicians who are not specialized in AD; this may enhance the accessibility of AD diagnosis and treatment.

A Longitudinal Study on Attenuated Structural Covariance in Patients With Somatic Symptom Disorder.

Objective: This study was performed to investigate altered regional gray matter volume (rGMV) and structural covariance related to somatic symptom disorder (SSD) and longitudinal changes after treatment. Additionally, this study examined the relationships of structural alteration with its phenotypic subtypes. Methods: Forty-three unmedicated patients with SSD and thirty normal controls completed psychological questionnaires and neurocognitive tests, as well as brain magnetic resonance imaging. Voxel-based morphometry and structural covariances were compared between groups and between subgroups within the SSD group. After 6 months of treatment, SSD patients were followed up for assessments. Results: Patients with SSD exhibited attenuated structural covariances in the pallidal-cerebellar circuit (FDR < 0.05-0.1), as well as regions in the default mode and sensorimotor network (FDR < 0.2), compared to normal controls. The cerebellar rGMVs were negatively correlated with the severity of somatic symptoms. In subgroup analyses, patients with somatic pain showed denser structural covariances between the bilateral superior temporal pole and left angular gyrus, the left middle temporal pole and left angular gyrus, and the left amygdala and right inferior orbitofrontal gyrus, while patients with headache and dizziness had greater structural covariance between the right inferior temporal gyrus and right cerebellum (FDR < 0.1-0.2). After 6 months of treatment, patients showed improved symptoms, however there was no significant structural alteration. Conclusion: The findings suggest that attenuated structural covariance may link to dysfunctional brain network and vulnerability to SSD; they also suggested that specific brain regions and networks may contribute to different subtypes of SSD.

Quantitative radiological analysis and clinical outcomes of urgent EC-IC bypass for hemodynamic compromised patients with acute ischemic stroke.

This study aimed to demonstrate the effectiveness of urgent extracranial-to-intracranial bypass (EIB) in acute ischemic stroke (AIS) through quantitative analysis of computed tomography perfusion (CTP) results using RAPID software. We retrospectively analyzed 41 patients who underwent urgent EIB for AIS under strict operation criteria. The quantitative data from CTP images were reconstructed to analyze changes in pre- and postoperative perfusion status in terms of objective numerical values using RAPID software. Short- and long-term clinical outcomes, including complications and neurological status, were also analyzed. Postoperatively, the volume of time-to-max (Tmax) > 6 s decreased significantly; it continued to improve significantly until 6 months postoperatively (preoperative, 78 ml (median); immediate postoperative, 23 ml; postoperative 6 months, 7 ml; p = 0.000). Ischemic core-penumbra mismatch volumes were also significantly improved until 6 months postoperatively (preoperative, 72 ml (median); immediate postoperative, 23 ml; postoperative 6 months, 5 ml; p = 0.000). In addition, the patients' neurological condition improved significantly (p < 0.001). Only one patient (2.3%) showed progression of infarction. Urgent EIB using strict indications can be a feasible treatment for IAT-ineligible patients with AIS due to large vessel occlusion or stenosis.

Predictive accuracy of T2-FLAIR mismatch sign for the IDH-mutant, 1p/19q noncodeleted low-grade glioma: An updated systematic review and meta-analysis.

BACKGROUND: The T2-fluid-attenuated inversion recovery (FLAIR) mismatch sign, has been considered a highly specific imaging biomarker of IDH-mutant, 1p/19q noncodeleted low-grade glioma. This systematic review and meta-analysis aimed to evaluate the diagnostic performance of T2-FLAIR mismatch sign for prediction of a patient with IDH-mutant, 1p/19q noncodeleted low-grade glioma, and identify the causes responsible for the heterogeneity across the included studies. METHODS: A systematic literature search in the Ovid-MEDLINE and EMBASE databases was performed for studies reporting the relevant topic before November 17, 2020. The pooled sensitivity and specificity values with their 95% confidence intervals were calculated using bivariate random-effects modeling. Meta-regression analyses were also performed to determine factors influencing heterogeneity. RESULTS: For all the 10 included cohorts from 8 studies, the pooled sensitivity was 40% (95% confidence interval [CI] 28-53%), and the pooled specificity was 100% (95% CI 95-100%). In the hierarchic summary receiver operating characteristic curve, the difference between the 95% confidence and prediction regions was relatively large, indicating heterogeneity among the studies. Higgins I2 statistics demonstrated considerable heterogeneity in sensitivity (I2 = 83.5%) and considerable heterogeneity in specificity (I2 = 95.83%). Among the potential covariates, it seemed that none of factors was significantly associated with study heterogeneity in the joint model. However, the specificity was increased in studies with all the factors based on the differences in the composition of the detailed tumors. CONCLUSIONS: The T2-FLAIR mismatch sign is near-perfect specific marker of IDH mutation and 1p/19q noncodeletion.

Diagnostic assessment of magnetic resonance imaging for patients with intralabyrinthine schwannoma: A systematic review.

OBJECTIVES: Recent advancements in high-resolution imaging have improved the diagnostic assessment of magnetic resonance imaging (MRI) for intralabyrinthine schwannoma (ILS). This systematic review aimed to evaluate the diagnostic performance of MRI for patients with ILS. METHODS: Ovid-MEDLINE and EMBASE databases were searched for related studies on the diagnostic performance of MRI for patients with ILS published up to February 10, 2020. The primary endpoint was the diagnostic performance of MRI for ILS. The quality of the enrolled studies was assessed using tailored questionnaires and the Quality Assessment of Diagnostic Accuracy Studies-2 criteria. RESULTS: Overall, 6 retrospective studies that included 122 patients with ILS from a parent population of 364 were included. The sample size, parent population and its composition, reference standard, detailed parameters of MRI, and even the diagnostic methods varied between the studies. The studies had moderate quality. The sensitivity of combination of T2WI and CE-T1WI was over 90%. Relative sensitivity of T2WI comparative to CE-T1WI ranged from 62% to 100%, and the specificity were 100%. CONCLUSIONS: MRI has acceptable diagnostic performance for ILS. There is a need for well-organized research to reduce the factors causing heterogeneity.

Deep Learning-Based Computer-Aided Detection System for Automated Treatment Response Assessment of Brain Metastases on 3D MRI.

BACKGROUND: Although accurate treatment response assessment for brain metastases (BMs) is crucial, it is highly labor intensive. This retrospective study aimed to develop a computer-aided detection (CAD) system for automated BM detection and treatment response evaluation using deep learning. METHODS: We included 214 consecutive MRI examinations of 147 patients with BM obtained between January 2015 and August 2016. These were divided into the training (174 MR images from 127 patients) and test datasets according to temporal separation (temporal test set #1; 40 MR images from 20 patients). For external validation, 24 patients with BM and 11 patients without BM from other institutions were included (geographic test set). In addition, we included 12 MRIs from BM patients obtained between August 2017 and March 2020 (temporal test set #2). Detection sensitivity, dice similarity coefficient (DSC) for segmentation, and agreements in one-dimensional and volumetric Response Assessment in Neuro-Oncology Brain Metastases (RANO-BM) criteria between CAD and radiologists were assessed. RESULTS: In the temporal test set #1, the sensitivity was 75.1% (95% confidence interval [CI]: 69.6%, 79.9%), mean DSC was 0.69 ± 0.22, and false-positive (FP) rate per scan was 0.8 for BM ≥ 5 mm. Agreements in the RANO-BM criteria were moderate (κ, 0.52) and substantial (κ, 0.68) for one-dimensional and volumetric, respectively. In the geographic test set, sensitivity was 87.7% (95% CI: 77.2%, 94.5%), mean DSC was 0.68 ± 0.20, and FP rate per scan was 1.9 for BM ≥ 5 mm. In the temporal test set #2, sensitivity was 94.7% (95% CI: 74.0%, 99.9%), mean DSC was 0.82 ± 0.20, and FP per scan was 0.5 (6/12) for BM ≥ 5 mm. CONCLUSIONS: Our CAD showed potential for automated treatment response assessment of BM ≥ 5 mm.

Classification of true progression after radiotherapy of brain metastasis on MRI using artificial intelligence: a systematic review and meta-analysis.

BACKGROUND: Classification of true progression from nonprogression (eg, radiation-necrosis) after stereotactic radiotherapy/radiosurgery of brain metastasis is known to be a challenging diagnostic task on conventional magnetic resonance imaging (MRI). The scope and status of research using artificial intelligence (AI) on classifying true progression are yet unknown. METHODS: We performed a systematic literature search of MEDLINE and EMBASE databases to identify studies that investigated the performance of AI-assisted MRI in classifying true progression after stereotactic radiotherapy/radiosurgery of brain metastasis, published before November 11, 2020. Pooled sensitivity and specificity were calculated using bivariate random-effects modeling. Meta-regression was performed for the identification of factors contributing to the heterogeneity among the studies. We assessed the quality of the studies using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) criteria and a modified version of the radiomics quality score (RQS). RESULTS: Seven studies were included, with a total of 485 patients and 907 tumors. The pooled sensitivity and specificity were 77% (95% CI, 70-83%) and 74% (64-82%), respectively. All 7 studies used radiomics, and none used deep learning. Several covariates including the proportion of lung cancer as the primary site, MR field strength, and radiomics segmentation slice showed a statistically significant association with the heterogeneity. Study quality was overall favorable in terms of the QUADAS-2 criteria, but not in terms of the RQS. CONCLUSION: The diagnostic performance of AI-assisted MRI seems yet inadequate to be used reliably in clinical practice. Future studies with improved methodologies and a larger training set are needed.

Image Findings of Acute to Subacute Craniocervical Arterial Dissection on Magnetic Resonance Vessel Wall Imaging: A Systematic Review and Proportion Meta-Analysis.

Background and Purpose: This systematic review and meta-analysis aimed to evaluate the pooled proportion of image findings of acute to subacute craniocervical arterial dissection (AD) direct signs on magnetic resonance vessel wall imaging (MR-VWI) and to identify factors responsible for the heterogeneity across the included studies. Methods: A systematic literature search in the Ovid-MEDLINE and EMBASE databases was performed for studies published on the relevant topic before April 14, 2020. Pooled sensitivity and specificity values and their 95% confidence intervals (CIs) were calculated using bivariate random-effects modeling. Meta-regression analyses were also performed to determine factors influencing heterogeneity. Results: Eleven articles with data for 209 patients with acute to subacute craniocervical AD who underwent MR-VWI were included in this systematic review and meta-analysis. The most common findings on MR-VWI were wall hematoma (84%; 95% CI, 71%-92%), abnormal enhancement (72%; 95% CI, 49%-88%), aneurysmal dilatation (71%, 95% CI, 53%-84%), and intimal flap or double lumen signs (49%; 95% CI, 29%-71%). Among the potential covariates of heterogeneity, the presence of contrast-enhanced T1-weighted imaging (CE-T1WI) within the MR-VWI sequence combination significantly affected the pooled proportion of the intimal flap or double lumen signs. Conclusion: Wall hematoma and intimal flap or double lumen signs were the most common and least common direct sign image findings, respectively, on MR-VWI in patients with acute to subacute craniocervical AD. Furthermore, the absence of CE-T1WI in MR-VWI protocol was the cause of heterogeneity for the detection of the intimal flap or double lumen signs. This data may help improve MR-VWI interpretation and enhance the understanding of the radiologic diagnosis of craniocervical AD.

Two-stage deep learning for accelerated 3D time-of-flight MRA without matched training data.

Time-of-flight magnetic resonance angiography (TOF-MRA) is one of the most widely used non-contrast MR imaging methods to visualize blood vessels, but due to the 3-D volume acquisition highly accelerated acquisition is necessary. Accordingly, high quality reconstruction from undersampled TOF-MRA is an important research topic for deep learning. However, most existing deep learning works require matched reference data for supervised training, which are often difficult to obtain. By extending the recent theoretical understanding of cycleGAN from the optimal transport theory, here we propose a novel two-stage unsupervised deep learning approach, which is composed of the multi-coil reconstruction network along the coronal plane followed by a multi-planar refinement network along the axial plane. Specifically, the first network is trained in the square-root of sum of squares (SSoS) domain to achieve high quality parallel image reconstruction, whereas the second refinement network is designed to efficiently learn the characteristics of highly-activated blood flow using double-headed projection discriminator. Extensive experiments demonstrate that the proposed learning process without matched reference exceeds performance of state-of-the-art compressed sensing (CS)-based method and provides comparable or even better results than supervised learning approaches.

Iron-sensitive magnetic resonance imaging in Parkinson's disease: a systematic review and meta-analysis.

OBJECTIVE: To evaluate the diagnostic performance of iron-sensitive sequences targeting the substantia nigra for distinguishing patients with Parkinson's disease from control participants and to identify factors causing heterogeneity. METHODS: A systematic literature search in the Ovid-MEDLINE and EMBASE databases was performed for studies reporting the relevant topic before March 6, 2020. The pooled sensitivity and specificity values with their 95% confidence intervals were calculated using bivariate random-effects modeling. Subgroup and meta-regression analyses were also performed to determine factors influencing heterogeneity affecting the diagnostic performance among the clinical, MRI, and analytic characteristics. RESULTS: A total of 22 articles including 1126 patients with Parkinson's disease and 933 control participants were enrolled in this systematic review and meta-analysis. Of those, 12 studies used objective analyses of quantitative susceptibility measurements, and 10 visually assessed the nigrosome-1 in subjective analyses. Iron-sensitive nigral magnetic resonance imaging showed a pooled sensitivity of 92% (95% confidence interval 88-95%) and a pooled specificity of 90% (95% confidence interval 81-95%). According to subgroup and meta-regression analyses, a longer mean disease duration in patients with Parkinson's disease (≥ 5 years), subjective analysis, a smaller size of pixel (< 0.6 mm2), a larger flip angle (> 15°), a smaller slice thickness (≤ 1 mm), and specific targeting of the substantia nigra pars compacta improved the diagnostic performance. CONCLUSION: Iron-sensitive nigral magnetic resonance imaging had a favorable diagnostic performance in discriminating patients with Parkinson's disease from control participants. Subjective analytic methods remain superior to objective approaches. Further improvements of the spatial resolution and contrast-to-noise ratio to specifically target the nigrosome-1 with objective analytic methods will be needed.

Robustness of Deep Learning Algorithm to Varying Imaging Conditions in Detecting Low Contrast Objects In Computed Tomography Phantom Images: in Comparison to 12 Radiologists.

The performance of deep learning algorithm (DLA) to that of radiologists was compared in detecting low contrast objects in CT phantom images under various imaging conditions. For training, 10,000 images were created using American College of Radiology CT phantom as the background. In half of the images, objects of 3-20 mm size and 5-30 HU contrast difference were generated in random locations. Binary responses were used as the ground truth. For testing, 640 images of Catphan® phantom were used, half of which had objects of either 5 or 9 mm size with 10 HU contrast difference. Twelve radiologists evaluated the presence of objects on a five-point scale. The performances of the DLA and radiologists were compared across different imaging conditions in terms of area under receiver operating characteristics curve (AUC). Multi-reader multi-case AUC and Hanley and McNeil tests were used. We performed post-hoc analysis using bootstrapping and verified that the DLA is less affected by the changing imaging conditions. The AUC of DLA was consistently higher than those of the radiologists across different imaging conditions (p < 0.0001), and it was less affected by varying imaging conditions. The DLA outperformed the radiologists and showed more robust performance under varying imaging conditions.

Comprehensive assessments of the open mouth dynamic maneuver and metal artifact reduction algorithm on computed tomography images of the oral cavity and oropharynx.

OBJECTIVES: To determine the optimal utility of the open mouth maneuver and Metal Artifact Reduction for the Orthopedic Implants (O-MAR) technique for CT of the oral cavity and oropharynx. METHODS: Between July 2017 and May 2019, 59 subjects who underwent both conventional and open mouth head and neck CT scans were included in this retrospective study. All images were reconstructed using the O-MAR algorithm. With conventional CT with/without the O-MAR (CTc_O/CTc) and open mouth CT with/without O-MAR (CTo_O/CTo), one reader measured the noise level in multiple anatomic regions of the oral cavity and oropharynx. Visual scores for the streak artifact and overall subjective image quality were assessed by two independent readers. RESULTS: For the mobile tongue, retromolar trigone, and palatine tonsil, the mean noise was significantly lower, and the mean visual scores were significantly higher, with CTo than with CTc or CTc_O (all, P < 0.001). The mean visual scores were higher with CTo_O than with CTo for the mobile tongue and palatine tonsil (all, P < 0.001). Contrarily, for the mouth floor and tongue base, the mean noise was significantly higher with CTo_O than with CTc or CTc_O, and the mean visual scores were significantly higher with CTc than with CTo or CTo_O (all, P < 0.001). CONCLUSIONS: The open mouth maneuver and O-MAR technique can have different influences on the CT image quality according to the anatomical subsites of the oral cavity and oropharynx.