Tomography

Background: Previous studies suggest that brain atrophy can not only be defined by its morphological extent, but also by the cerebral blood flow (CBF) within a certain area of the brain, including white and gray matter. The aim of this study is to investigate known atrophy patterns in different forms of dementia and to compare segmented brain volumetrics and pulsed arterial spin labeling (pASL) data to explore the correlation between brain maps with atrophy and this non-contrast-enhanced brain-perfusion method. Methods: Our study comprised 17 patients with diagnosed cognitive impairment (five Alzheimer’s disease = AD, five frontotemporal dementia = FTD, seven mild cognitive impairment = MCI) and 19 healthy control subjects (CO). All patients and controls underwent 4D-pASL brain-perfusion MR imaging and T1w MPRAGE. The data were assessed regarding relative brain volume on the basis of 286 brain regions, and absolute and relative cerebral blood flow (CBF/rCBF) were derived from pASL data in the corresponding brain regions. Mini-Mental State Examination (MMSE) was performed to assess cognitive functions. Results: FTD patients demonstrated significant brain atrophy in 43 brain regions compared to CO. Patients with MCI showed significant brain atrophy in 18 brain regions compared to CO, whereas AD patients only showed six brain regions with significant brain atrophy compared to CO. There was good correlation of brain atrophy and pASL perfusion data in five brain regions of patients with diagnosed FTD, especially in the superior temporal gyrus (r = 0.900, p = 0.037), the inferior frontal white matter (pars orbitalis; r = 0.968, p = 0.007) and the thalami (r = 0.810, p = 0.015). Patients with MCI demonstrated a correlation in one brain region (left inferior fronto-occipital fasciculus; r = 0.786, p = 0.036), whereas patients with diagnosed AD revealed no correlation. Conclusions: pASL can detect affected brain regions in cognitive impairment and corresponds with brain atrophy, especially for patients suffering from FTD and MCI. However, there was no correlation of perfusion alterations and brain atrophy in AD. pASL perfusion might thus represent a promising tool for noninvasive brain-perfusion evaluation in specific dementia subtypes as a complimentary imaging-based bio marker in addition to brain volumetry. Full article

Non-traumatic thoracic aorta emergencies are associated with significant morbidity and mortality. Diseases of the intimomedial layers (aortic dissection and variants) have been grouped under the common term of acute aortic syndrome because they are life-threatening conditions clinically indistinguishable on presentation. Patients with aortic dissection may present with a wide variety of symptoms secondary to the pattern of dissection and end organ malperfusion. Other conditions may be seen in patients with acute symptoms, including ruptured and unstable thoracic aortic aneurysm, iatrogenic or infective pseudoaneurysms, aortic fistula, acute aortic thrombus/occlusive disease, and vasculitis. Imaging plays a pivotal role in the patient’s management and care. In the emergency room, chest X-ray is the initial imaging test offering a screening evaluation for alternative common differential diagnoses and a preliminary assessment of the mediastinal dimensions. State-of-the-art multidetector computed tomography angiography (CTA) provides a widely available, rapid, replicable, noninvasive diagnostic imaging with sensitivity approaching 100%. It is an impressive tool in decision-making process with a deep impact on treatment including endovascular or open surgical or conservative treatment. Radiologists must be familiar with the spectrum of these entities to help triage patients appropriately and efficiently. Understanding the imaging findings and proper measurement techniques allow the radiologist to suggest the most appropriate next management step. Full article

Background: The present study examined the relationships among macular microvasculature, retinal structure, and epiretinal membrane (ERM) and explored the utility of optical coherence tomography (OCT) angiography (OCTA) in idiopathic ERM assessment. Methods: The study sample comprised 276 eyes of 276 patients. A total of 154 eyes with ERM and 122 normal (control) eyes were analyzed. Only one eye of each participant was randomly selected for posterior segment imaging. Each patient underwent OCT and OCTA. Images were analyzed with AngioTool 0.6. Results: Foveal avascular zone was significantly smaller in the ERM group (p = 0.044). Average retinal thickness and foveal thickness were significantly higher in the ERM group (both p = 0.001). Moreover, 64 (41.5%) patients exhibited no metamorphopsia, while 46 (29.8%) and 44 (28.7%) patients exhibited moderate and extensive metamorphopsias, respectively. Meanwhile, FAZ was negatively correlated with central retinal thickness in the ERM group. The vessel area (p = 0.0017) and vessel percentage area (p = 0.044) were significantly greater in the ERM group. Conclusions: Changes observed in the superficial plexus in OCTA are related to the severity of metamorphopsia and can be further evaluated to support decision making regarding the surgical management of idiopathic ERM. Full article

Background: SPECT/CT distribution patterns in patients with total knee replacements have previously been correlated with factors such as time of implantation, implant type and alignment. It is unknown, however, if an increased and more extended bone tracer uptake (BTU) in SPECT/CT, representing loading of the joint, correlates with findings from retrieval studies. The aim of this study was to further understand this subject. Materials and Methods: 62 retrieved TKA were included. SPECT/CT was performed prior to revision. Quantitative and qualitative medio-lateral comparisons of BTU intensity and distribution in the tibia were performed. Retrieval analysis was performed with a micro-CT method to assess the thickness differences between medial and lateral sides of polyethylene inserts with symmetrical designs. Results: In the subgroup of TKA with asymmetrical SPECT/CT distribution, there was a significant correlation between retrieval and medical imaging data (p = 0.0355): patients showing a more extended BTU in the medial compartment also had a significantly thinner insert in the medial compartment, and vice versa in the lateral side. Conclusion: This is the first study comparing BTU distribution patterns and retrieval findings. Our results support the hypothesis that SPECT/CT is able to identify bone activity due to implant position and loading. Full article

In this report, we describe the functional imaging findings of systemic artery to pulmonary artery shunt in V/Q SPECT CT imaging. A 63-year-old man with small-cell lung cancer underwent CT pulmonary angiography (CTPA) for suspected acute pulmonary embolism (PE). The CTPA showed an isolated segmental filling defect in the right lower lobe, which was initially interpreted as positive for PE but was actually the consequence of a systemic artery to pulmonary artery shunt due to the recruitment of the bronchial arterial network by the adjacent tumor. A V/Q SPECT/CT scan was also performed, demonstrating a matched perfusion/ventilation defect in the right lower lobe. Full article

In this paper, we present a new regularized image reconstruction method for positron emission tomography (PET), where an adaptive weighted median regularizer is used in the context of a penalized-likelihood framework. The motivation of our work is to overcome the limitation of the conventional median regularizer, which has proven useful for tomographic reconstruction but suffers from the negative effect of removing fine details in the underlying image when the edges occupy less than half of the window elements. The crux of our method is inspired by the well-known non-local means denoising approach, which exploits the measure of similarity between the image patches for weighted smoothing. However, our method is different from the non-local means denoising approach in that the similarity measure between the patches is used for the median weights rather than for the smoothing weights. As the median weights, in this case, are spatially variant, they provide adaptive median regularization achieving high-quality reconstructions. The experimental results indicate that our similarity-driven median regularization method not only improves the reconstruction accuracy, but also has great potential for super-resolution reconstruction for PET. Full article

We aimed to delineate cardiopulmonary interactions in acute bronchiolitis and to evaluate the capacity of a combined cardiopulmonary ultrasonography to predict the need for respiratory support. This was a prospective observational single-center study that includes infants <12 month of age admitted to a hospital due to acute bronchiolitis. All the included patients underwent clinical, laboratory and cardiopulmonary ultrasonographic evaluation at the same time point within 24 h of hospital admission. The existence of significant correlation between cardiac and respiratory parameters was the primary outcome. The association of different cardiopulmonary variables with the need of respiratory support higher than O₂, the length of stay hospitalization, the PICU stay and the duration of respiratory support were a secondary outcome. We enrolled 112 infants (median age 1 (0.5–3) months; 62% males) hospitalized with acute bronchiolitis. Increased values of the pulmonary variables (BROSJOD score, pCO₂ and LUS) showed moderate correlations with NT-proBNP and all echocardiographic parameters indicative of pulmonary hypertension and myocardial dysfunction (Tei index). Up to 36 (32%) infants required respiratory support during the hospitalization. This group presented with higher lung ultrasound score (p < 0.001) and increased values of NT-proBNP (p < 0.001), the Tei index (p < 0.001) and pulmonary artery pressures (p < 0.001). All the analyzed respiratory and cardiac variables showed moderate-to-strong correlations with the LOS of hospitalization and the time of respiratory support. Lung ultrasound and echocardiography showed a moderate-to-strong predictive accuracy for the need of respiratory support in the ROC analysis, with the AUC varying from 0.74 to 0.87. Those cases of bronchiolitis with a worse pulmonary status presented with a more impaired cardiac status. Cardiopulmonary ultrasonography could be a useful tool to easily identify high-risk populations for complicated acute bronchiolitis hospitalization. Full article

Deep learning (DL) has become a remarkably powerful tool for image processing recently. However, the usefulness of DL in positron emission tomography (PET)/computed tomography (CT) for breast cancer (BC) has been insufficiently studied. This study investigated whether a DL model using images with multiple degrees of PET maximum-intensity projection (MIP) images contributes to increase diagnostic accuracy for PET/CT image classification in BC. We retrospectively gathered 400 images of 200 BC and 200 non-BC patients for training data. For each image, we obtained PET MIP images with four different degrees (0°, 30°, 60°, 90°) and made two DL models using Xception. One DL model diagnosed BC with only 0-degree MIP and the other used four different degrees. After training phases, our DL models analyzed test data including 50 BC and 50 non-BC patients. Five radiologists interpreted these test data. Sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were calculated. Our 4-degree model, 0-degree model, and radiologists had a sensitivity of 96%, 82%, and 80–98% and a specificity of 80%, 88%, and 76–92%, respectively. Our 4-degree model had equal or better diagnostic performance compared with that of the radiologists (AUC = 0.936 and 0.872–0.967, p = 0.036–0.405). A DL model similar to our 4-degree model may lead to help radiologists in their diagnostic work in the future. Full article

Objective: To evaluate computed tomography angiography (CTA) data focusing on radiation dose parameters in Thais with Marfan syndrome (MFS) and estimate the distribution of cumulative radiation exposure from CTA surveillance and the risk of cancers. Methods: Between 1st January 2015 and 31st December 2020, we retrospectively evaluated the cumulative CTA radiation doses of MFS patients who underwent CTA at Khon Kaen University Hospital, a leading teaching hospital and advanced tertiary care institution in northeastern Thailand. We utilized the Radiation Risk Assessment Tool (RadRAT) established at the National Cancer Institute in Bethesda, Maryland, to evaluate the risk of cancer-related CTA radiation. Results: The study recruited 29 adult MFS patients who had CTA of the aorta during a 5-year study period with 89 CTA studies. The mean cumulative CTDI vol is 21.5 ± 14.68 mGy, mean cumulative DLP is 682.2 ± 466.7 mGy.cm, the mean baseline future risk for all cancer is 26,134 ± 7601 per 100,000, and the excess lifetime risk for all cancer is 2080.3 ± 1330 per 100,000. The excess lifetime risk of radiation-induced cancer associated with the CTA surveillance study is significantly lower than the risk of aortic dissection or rupture and lower than the baseline future cancer risk. Conclusions: We attempted to quantify the radiation-induced cancer risk from CTA surveillance imaging performed for MFS patients in this study, with all patients receiving a low-risk cumulative radiation dose (less than 1 Gy) and all patients having a low excessive lifetime risk of cancer as a result of CTA. The risk–benefit decision must be made at the point of care, and it entails balancing the benefits of surveillance imaging in anticipating rupture and providing practical, safe treatment, therefore avoiding morbidity and mortality. Full article

The complex anatomy and similarity of imaging features of various pathologies in the pelvis can make accurate radiology interpretation difficult. While prompt recognition of ovarian cancer remains essential, awareness of processes that mimic ovarian tumors can avoid potential misdiagnosis and unnecessary surgery. This article details the female pelvic anatomy and highlights relevant imaging features that mimic extra-ovarian tumors, to help the radiologists accurately build a differential diagnosis of a lesion occupying the adnexa. Full article

The gold standard for the diagnosis of pelvic congestion syndrome (PCS) is venography (VG), although transvaginal ultrasound (TVU) might be a noninvasive, nonionizing alternative. Our aim is to determine whether TVU is an accurate and comparable diagnostic tool for PCS. An observational prospective study including 67 patients was carried out. A TVU was performed on patients, measuring pelvic venous vessels parameters. Subsequentially, a VG was performed, and results were compared for the test calibration of TVU. Out of the 67 patients included, only 51 completed the study and were distributed in two groups according to VG results: 39 patients belonging to the PCS group and 12 to the normal group. PCS patients had a larger venous plexus diameter (15.1 mm vs. 12 mm; p = 0.009) and higher rates of crossing veins in the myometrium (74.35% vs. 33.3%; p = 0.009), reverse or altered flow during Valsalva (58.9% vs. 25%; p = 0.04), and largest pelvic vein ≥ 8 mm (92.3% vs. 25%). The sensitivity and specificity of TVU were 92.3% (95% CI: 78.03–97.99%) and 75% (95% CI: 42.84–93.31%), respectively. In conclusion, transvaginal ultrasonography, with the described methodology, appears to be a promising tool for the diagnosis of PCS, with acceptable sensitivity and specificity. Full article

This study investigated the relationship between sinonasal inflammatory involvement according to the computed tomography (CT) staging system (Lund–Mackay score) with clinical, laboratory, histopathological and prognostic features of chronic rhinosinusitis with nasal polyps (CRSwNP). Seventy-eight patients with CRSwNP who had undergone surgery were enrolled. Total (p = 0.0062), ethmoid (p = 0.0496), sphenoid (p = 0.0335), ostiomeatal complex (OMC) (p = 0.0235) and frontal (p = 0.0164) CT scores were predictive of non-steroidal anti-inflammatory drugs-exacerbated respiratory disease (NERD) in the univariate analysis. Total (p = 0.0022), ethmoid (p = 0.0290), sphenoid (p = 0.0370), frontal (p = 0.0116), maxillary (p = 0.0357) and OMC (p = 0.0058) CT scores were predictve of asthma at the univariate analysis. No significant differences were found between patients with vs. without allergy in terms of total and partial CT scores. High blood eosinophil counts (>0.24 vs. ≤0.24 cells × 10⁹/L) resulted in being associated with total (p = 0.0213), maxillary (p = 0.0227) and ethmoid (p = 0.0491) CT scores in the univariate analysis. Higher ethmoid (p = 0.0006) and total sinonasal (p = 0.0027) CT scores were found to predict histopathologically eosinophil CRSwNPs in the univariate analysis. CT scores did not result as predictive of NSAID-exacerbated respiratory disease, asthma, or blood eosinophil count at the multivariate analysis. Risk of relapse was related to the presence of NERD (p = 0.0207, HR [95% CI] 3.914 [1.232–12.435]), higher preoperative total (HR = 1.098 95%CI: 1.001–1.204, p = 0.0486) and frontal sinus CT scores (HR = 1.555 95%CI: 1.006–1.886, p = 0.0218), but these results were not confirmed by the multivariable analysis. Sinonasal CT scores showed significant differences in this heterogeneous inflammatory condition. Identifying CRSwNP characteristics is necessary to avoid generic treatments with poor outcomes. Full article

Automatic image segmentation plays an important role in the fields of medical image processing so that these fields constantly put forward higher requirements for the accuracy and speed of segmentation. In order to improve the speed and performance of the segmentation algorithm of medical images, we propose a medical image segmentation algorithm based on simple non-iterative clustering (SNIC). Firstly, obtain the feature map of the image by extracting the texture information of it with feature extraction algorithm; Secondly, reduce the image to a quarter of the original image size by downscaling; Then, the SNIC super-pixel algorithm with texture information and adaptive parameters which used to segment the downscaling image to obtain the superpixel mark map; Finally, restore the superpixel labeled image to the original size through the idea of the nearest neighbor algorithm. Experimental results show that the algorithm uses an improved superpixel segmentation method on downscaling images, which can increase the segmentation speed when segmenting medical images, while ensuring excellent segmentation accuracy. Full article

Due to the complex shape of the vertebrae and the background containing a lot of interference information, it is difficult to accurately segment the vertebrae from the computed tomography (CT) volume by manual segmentation. This paper proposes a convolutional neural network for vertebrae segmentation, named Verte-Box. Firstly, in order to enhance feature representation and suppress interference information, this paper places a robust attention mechanism on the central processing unit, including a channel attention module and a dual attention module. The channel attention module is used to explore and emphasize the interdependence between channel graphs of low-level features. The dual attention module is used to enhance features along the location and channel dimensions. Secondly, we design a multi-scale convolution block to the network, which can make full use of different combinations of receptive field sizes and significantly improve the network’s perception of the shape and size of the vertebrae. In addition, we connect the rough segmentation prediction maps generated by each feature in the feature box to generate the final fine prediction result. Therefore, the deep supervision network can effectively capture vertebrae information. We evaluated our method on the publicly available dataset of the CSI 2014 Vertebral Segmentation Challenge and achieved a mean Dice similarity coefficient of 92.18 ± 0.45%, an intersection over union of 87.29 ± 0.58%, and a 95% Hausdorff distance of 7.7107 ± 0.5958, outperforming other algorithms. Full article

Cognitive training-induced neuroplastic brain changes have been reported. This prospective study evaluated whether microscopic fractional anisotropy (μFA) derived from double diffusion encoding (DDE) MRI could detect brain changes following a 4 week cognitive training. Twenty-nine healthy volunteers were recruited and randomly assigned into the training (n = 21) and control (n = 8) groups. Both groups underwent brain MRI including DDE MRI and 3D-T1-weighted imaging twice at an interval of 4–6 weeks, during which the former underwent the training. The training consisted of hour-long dual N-back and attention network tasks conducted five days per week. Training and time-related changes of DDE MRI indices (μFA, fractional anisotropy (FA), and mean diffusivity (MD)) and the gray and white matter volume were evaluated using mixed-design analysis of variance. In addition, any significant imaging indices were tested for correlation with cognitive training-induced task performance changes, using partial correlation analyses. μFA in the left middle frontal gyrus decreased upon the training (53 voxels, uncorrected p < 0.001), which correlated moderately with response time changes in the orienting component of attention (r = −0.521, uncorrected p = 0.032). No significant training and time-related changes were observed for other imaging indices. Thus, μFA can become a sensitive index to detect cognitive training-induced neuroplastic changes. Full article

(1) To explore the potential impact of an AI dual-energy CT (DECT) prototype on decision making and workflows by investigating its capabilities to differentiate COVID-19 from immunotherapy-related pneumonitis. (2) Methods: From 3 April 2020 to 12 February 2021, DECT from biometrically matching patients with COVID-19, pneumonitis, and inconspicuous findings were selected from our clinical routine. Three blinded readers independently scored each pulmonary lobe analogous to CO-RADS. Inter-rater agreement was determined with an intraclass correlation coefficient (ICC). Averaged perfusion metrics per lobe (iodine uptake in mg, volume without vessels in ml, iodine concentration in mg/mL) were extracted using manual segmentation and an AI DECT prototype. A generalized linear mixed model was used to investigate metric validity and potential distinctions at equal CO-RADS scores. Multinomial regression measured the contribution “Reader”, “CO-RADS score”, and “perfusion metrics” to diagnosis. The time to diagnosis was measured for manual vs. AI segmentation. (3) Results: We included 105 patients (62 ± 13 years, mean BMI 27 ± 2). There were no significant differences between manually and AI-extracted perfusion metrics (p = 0.999). Regardless of the CO-RADS score, iodine uptake and concentration per lobe were significantly higher in COVID-19 than in pneumonitis (p < 0.001). In regression, iodine uptake had a greater contribution to diagnosis than CO-RADS scoring (Odds Ratio (OR) = 1.82 [95%CI 1.10–2.99] vs. OR = 0.20 [95%CI 0.14–0.29]). The AI prototype extracted the relevant perfusion metrics significantly faster than radiologists (10 ± 1 vs. 15 ± 2 min, p < 0.001). (4) Conclusions: The investigated AI prototype positively impacts decision making and workflows by extracting perfusion metrics that differentiate COVID-19 from visually similar pneumonitis significantly faster than radiologists. Full article

The purpose of this work is to evaluate the feasibility of performing magnetic resonance fingerprinting (MRF) on older and lower-performance MRI hardware as a means to bring advanced imaging to the aging MRI install base. Phantom and in vivo experiments were performed on a 1.5T Siemens Aera (installed 2015) and 1.5T Siemens Symphony (installed 2002). A 2D spiral MRF sequence for simultaneous T₁/T₂/M₀ mapping was implemented on both scanners with different gradient trajectories to accommodate system specifications. In phantom, for T₁/T₂ values in a physiologically relevant range (T₁: 195–1539 ms; T₂: 20–267 ms), scanners had strong correlation (R² > 0.999) with average absolute percent difference of 8.1% and 10.1%, respectively. Comparison of the two trajectories on the newer scanner showed differences of 2.6% (T₁) and 10.9% (T₂), suggesting a partial explanation of the observed inter-scanner bias. Inter-scanner agreement was better when the same trajectory was used, with differences of 6.0% (T₁) and 4.0% (T₂). Intra-scanner coefficient of variation (CV) of T₁ and T₂ estimates in phantom were <2.0% and in vivo were ≤3.5%. In vivo inter-scanner white matter CV was 4.8% (T₁) and 5.1% (T₂). White matter measurements on the aging scanner after two months were consistent, with differences of 1.9% (T₁) and 3.9% (T₂). In conclusion, MRF is feasible on an aging MRI scanner and required only changes to the gradient trajectory. Full article

Cerebral venous sinus thrombosis (CVST) on non-contrast CT (NCCT) is often challenging to detect. We retrospectively selected 41 children and 36 adults with confirmed CVST and two age-matched control groups with comparable initial symptoms. We evaluated NCCT placing four small circular ROIs in standardized regions of the cerebral dural venous system. The mean and maximum HU values were considered from each ROI, and the relative percentage variations were calculated (mean % variation and maximum % variation). We compared the highest measured value to the remaining three HU values through an ad-hoc formula based on the assumption that the thrombosed sinus has higher attenuation compared with the healthy sinuses. Percentage variations were employed to reflect how the attenuation of the thrombosed sinus deviates from the unaffected counterparts. The attenuation of the affected sinus was increased in patients with CVST, and consequently both the mean % and maximum % variations were increased. A mean % variation value of 12.97 and a maximum % variation value of 10.14 were found to be useful to distinguish patients with CVST from healthy subjects, with high sensitivity and specificity. Increased densitometric values were present in the site of venous thrombosis. A systematic, blind evaluation of the brain venous system can assist radiologists in identifying patients who need or do not need further imaging. Full article

Evaluation of Parkinsonian Syndromes (PS) with Ioflupane iodine-123 dopamine transporter single photon emission computed tomography (DaT-SPECT), in conjunction with history and clinical examination, aids in diagnosis. FDA-approved, semi-quantitative software, DaTQUANT^TM (GE Healthcare, Chicago, IL, USA) is available to assist in interpretation. This study aims to evaluate the optimal variables and thresholds of DaTQUANT to yield the optimal diagnostic accuracy. It is a retrospective review with three different patient populations. DaT-SPECT images from all three study groups were evaluated using DaTQUANT^TM software, and both single and multi-variable logistic regression were used to model PS status. The optimal models were chosen via accuracy, sensitivity, and specificity, then evaluated on the other study groups. Among single variable models, the posterior putamen yielded the highest accuracy (84% to 95%), while balancing sensitivity and specificity. Multi-variable models did not substantially improve the accuracy. When the optimal single variable models for each group were used to evaluate the remaining two groups, comparable results were achieved. In typical utilization of DaT-SPECT for differentiation between nigrostriatal degenerative disease (NSDD) and non-NSDD, the posterior putamen was the single variable that yielded the highest accuracy across three different patient populations. The posterior putamen’s recommended thresholds for DaTQUANT are SBR ≤ 1.0, z-score of ≤−1.8 and percent deviation ≤ −0.34. Full article