Tomography

Background: mpMRI assesses prostate lesions through their PI-RADS score. The primary goal of this prospective study was to demonstrate the correlation of PI-RADS v2 score and the volume of a lesion with the presence and clinical significance of prostate cancer (PCa). The secondary goal was to determine the extent of additionally PCa in inconspicuous areas. Methods: All 157 patients underwent a perineal MRI/TRUS-fusion prostate biopsy. Targeted biopsies as well as a systematic biopsy were performed. The presence of PCa in the probes was specified by the ISUP grading system. Results: In total, 258 lesions were biopsied. Of the PI-RADS 3 lesions, 24% were neoplastic. This was also true for 36.9% of the PI-RADS 4 lesions and for 59.5% of the PI-RADS 5 lesions. Correlation between PI-RADS v2 scores and lesion volume was significant (p < 0.00001). In the non-suspicious mpMRI areas carcinoma was revealed in 19.7% of the patients. Conclusions: The study shows that the PI-RADS v2 score and the lesion volume correlate with the presence and clinical significance of PCa. However, there are two major points to consider: First, there is a high number of false positive findings. Second, inconspicuous mpMRI areas revealed PCa. Full article

Background: To evaluate the clinical utility of an Artificial Intelligence (AI) radiology solution, Quantib Prostate, for prostate cancer (PCa) lesions detection on multiparametric Magnetic Resonance Images (mpMRI). Methods: Prostate mpMRI exams of 108 patients were retrospectively studied. The diagnostic performance of an expert radiologist (>8 years of experience) and of an inexperienced radiologist aided by Quantib software were compared. Three groups of patients were assessed: patients with positive mpMRI, positive target biopsy, and/or at least one positive random biopsy (group A, 73 patients); patients with positive mpMRI and a negative biopsy (group B, 14 patients), and patients with negative mpMRI who did not undergo biopsy (group-C, 21 patients). Results: In group A, the AI-assisted radiologist found new lesions with positive biopsy correlation, increasing the diagnostic PCa performance when compared with the expert radiologist, reaching an SE of 92.3% and a PPV of 90.1% (vs. 71.7% and 84.4%). In group A, the expert radiologist found 96 lesions on 73 mpMRI exams (17.7% PIRADS3, 56.3% PIRADS4, and 26% PIRADS5). The AI-assisted radiologist found 121 lesions (0.8% PIRADS3, 53.7% PIRADS4, and 45.5% PIRADS5). At biopsy, 33.9% of the lesions were ISUP1, 31.4% were ISUP2, 22% were ISUP3, 10.2% were ISUP4, and 2.5% were ISUP5. In group B, where biopsies were negative, the AI-assisted radiologist excluded three lesions but confirmed all the others. In group-C, the AI-assisted radiologist found 37 new lesions, most of them PIRADS 3, with 32.4% localized in the peripherical zone and 67.6% in the transition zone. Conclusions: Quantib software is a very sensitive tool to use specifically in high-risk patients (high PIRADS and high Gleason score). Full article

Breast-conserving surgery (BCS) with negative resection margins decreases the locoregional recurrence rate. Breast cancer size is one of the main determinants of Tumor-Node-Metastasis (TNM) staging. Our study aimed to investigate the accuracy of supine 3D automated breast ultrasound (3D ABUS) compared to prone 3D ABUS in the evaluation of tumor size in breast cancer patient candidates for BCS. In this prospective two-center study (Groups 1 and 2), we enrolled patients with percutaneous biopsy-proven early-stage breast cancer, in the period between June 2019 and May 2020. Patients underwent hand-held ultrasound (HHUS), contrast-enhanced magnetic resonance imaging (CE-MRI) and 3D ABUS—supine 3D ABUS in Group 1 and prone 3D ABUS in Group 2. Histopathological examination (HE) was considered the reference standard. Bland–Altman analysis and plots were used. Eighty-eight patients were enrolled. Compared to prone, supine 3D ABUS showed better agreement with HE, with a slight tendency toward underestimation (mean difference of −2 mm). Supine 3D ABUS appears to be a useful tool and more accurate than HHUS in the staging of breast cancer. Full article

Peritumoral brain edema (PBE) is common in intracranial meningiomas (IM) and can increase their morbidity. It is not uncommon for a neurosurgeon to confront meningiomas with a large proportion of PBE independently from the site and size of the contrast-enhancing lesion with increased surgical risks. We performed a retrospective review of 216 surgically-treated patients suffering from IM. We recorded clinical, biological, and radiological data based on the rate of tumor and edema volume and divided the patients into a group with high Edema/Tumor ratio and a group with a low ratio. We investigated how the ratio of edema/lesion may affect the outcome. Multivariate analysis was performed for the two groups. Smokers were found to be more likely to belong to the high-rate group. The edema/tumor ratio did not affect the surgical radicality; however, independently of the biological sub-type, WHO grading, and EOR, a higher frequency of recurrence is shown in patients with a high edema/tumor ratio (70.5% vs. 8.4%. p < 0.01). There is evidence to suggest that the blood-brain barrier (BBB) damage from smoke could play a role in an increased volume of PBE. The present study demonstrates that IMs showing a high PBE ratio to tumor volume at diagnosis are associated with a smoking habit and a higher incidence of recurrence independently of their biological type and grading. Full article

Amide proton transfer (APT)-weighted MRI is a promising molecular imaging technique that has been employed in clinic for detection and grading of brain tumors. MTR_asym, the quantification method of APT, is easily influenced by B₀ inhomogeneity and causes artifacts. Current model-free interpolation methods have enabled moderate B₀ correction for middle offsets, but have performed poorly at limbic offsets. To address this shortcoming, we proposed a practical B₀ correction approach that is suitable under time-limited sparse acquisition scenarios and for B₁ ≥ 1 μT under 3T. In this study, this approach employed a simplified Lorentzian model containing only two pools of symmetric water and asymmetric solutes, to describe the Z-spectral shape with wide and ‘invisible’ CEST peaks. The B₀ correction was then performed on the basis of the fitted two-pool Lorentzian lines, instead of using conventional model-free interpolation. The approach was firstly evaluated on densely sampled Z-spectra data by using the spline interpolation of all acquired 16 offsets as the gold standard. When only six offsets were available for B₀ correction, our method outperformed conventional methods. In particular, the errors at limbic offsets were significantly reduced (n = 8, p < 0.01). Secondly, our method was assessed on the six-offset APT data of nine brain tumor patients. Our MTR_asym (3.5 ppm), using the two-pool model, displayed a similar contrast to the vendor-provided B₀-orrected MTR_asym (3.5 ppm). While the vendor failed in correcting B₀ at 4.3 and 2.7 ppm for a large portion of voxels, our method enabled well differentiation of B₀ artifacts from tumors. In conclusion, the proposed approach could alleviate analysis errors caused by B₀ inhomogeneity, which is useful for facilitating the comprehensive metabolic analysis of brain tumors. Full article

We reviewed the reported imaging findings of myocarditis in the literature following COVID-19 vaccination on cardiac imaging by a literature search in online databases, including Scopus, Medline (PubMed), Web of Science, Embase (Elsevier), and Google Scholar. In total, 532 cases of myocarditis after COVID-19 vaccination were reported (462, 86.8% men and 70, 13.2% women, age range 12 to 80) with the following distribution: Pfizer-BioNTech: 367 (69%), Moderna: 137 (25.8%), AstraZeneca: 12 (2.3%), Janssen/Johnson & Johnson: 6 (1.1%), COVAXIN: 1 (0.1%), and unknown mRNA vaccine: 9 (1.7%). The distribution of patients receiving vaccine dosage was investigated. On cardiac MR Imaging, late intravenous gadolinium enhancement (LGE) was observed mainly in the epicardial/subepicardial segments (90.8%, 318 of 350 enhancing segments), with the dominance of inferolateral segment and inferior walls. Pericardial effusion was reported in 13.1% of cases. The vast majority of patients (94%, 500 of 532) were discharged from the hospital except for 4 (0.7%) cases. Post-COVID-19 myocarditis was most commonly reported in symptomatic men after the second or third dose, with CMRI findings including LGE in 90.8% of inferior and inferolateral epicardial/subepicardial segments. Most cases were self-limited. Full article

The aim of this study was to investigate the features of partial anomalous left pulmonary artery (PALPA) and differences between cases with posterior versus anterior a nomalous vessels in relation to the tracheobronchial tree. We hypothesized that statistical significance was dependent on the course of the anomalous vessel due to airway compression in the posterior type. This study included cases obtained from the literature (n = 33) and an institution teaching file (n = 2). Information collected: age, sex, medical history, additional anomalies, anomalous vessel course, and respiratory symptoms. Data were analyzed with independent samples t-test and Fisher’s exact test. PALPAs were more commonly anterior than posterior. Mean age: 5.3 years (SD = 12.4) for anterior and 6.8 years (SD = 18.5) for posterior (p = 0.77). Respiratory symptoms: 20% of anterior and 60% of posterior cases (p = 0.032). Tracheobronchial anomalies: 35% of anterior and 60% of posterior cases (p = 0.182). Non-cardiac and non-tracheobronchial anomalies: 30% of anterior and 47% of posterior cases (p = 0.511). Kabuki syndrome: 25% of anterior and 6.7% of posterior cases (p = 0.207). In conclusion, respiratory symptoms were the only significant difference between anterior and posterior PALPA types. Full article

The High-Resolution Computed Tomography (HRCT) detection and diagnosis of diffuse lung disease is primarily based on the recognition of a limited number of specific abnormal findings, pattern combinations or their distributions, as well as anamnesis and clinical information. Since texture recognition has a very high accuracy percentage if a complex network approach is used, this paper aims to implement such a technique customized for diffuse interstitial lung diseases (DILD). The proposed procedure translates HRCT lung imaging into complex networks by taking samples containing a secondary lobule, converting them into complex networks and analyzing them in three dimensions: emphysema, ground glass opacity, and consolidation. This method was evaluated on a 60-patient lot and the results showed a clear, quantifiable difference between healthy and affected lungs. By deconstructing the image on three pathological axes, the method offers an objective way to quantify DILD details which, so far, have only been analyzed subjectively. Full article

A brain tumor is the growth of abnormal cells in certain brain tissues with a high mortality rate; therefore, it requires high precision in diagnosis, as a minor human judgment can eventually cause severe consequences. Magnetic Resonance Image (MRI) serves as a non-invasive tool to detect the presence of a tumor. However, Rician noise is inevitably instilled during the image acquisition process, which leads to poor observation and interferes with the treatment. Computer-Aided Diagnosis (CAD) systems can perform early diagnosis of the disease, potentially increasing the chances of survival, and lessening the need for an expert to analyze the MRIs. Convolutional Neural Networks (CNN) have proven to be very effective in tumor detection in brain MRIs. There have been multiple studies dedicated to brain tumor classification; however, these techniques lack the evaluation of the impact of the Rician noise on state-of-the-art deep learning techniques and the consideration of the scaling impact on the performance of the deep learning as the size and location of tumors vary from image to image with irregular shape and boundaries. Moreover, transfer learning-based pre-trained models such as AlexNet and ResNet have been used for brain tumor detection. However, these architectures have many trainable parameters and hence have a high computational cost. This study proposes a two-fold solution: (a) Multi-Scale CNN (MSCNN) architecture to develop a robust classification model for brain tumor diagnosis, and (b) minimizing the impact of Rician noise on the performance of the MSCNN. The proposed model is a multi-class classification solution that classifies MRIs into glioma, meningioma, pituitary, and non-tumor. The core objective is to develop a robust model for enhancing the performance of the existing tumor detection systems in terms of accuracy and efficiency. Furthermore, MRIs are denoised using a Fuzzy Similarity-based Non-Local Means (FSNLM) filter to improve the classification results. Different evaluation metrics are employed, such as accuracy, precision, recall, specificity, and F1-score, to evaluate and compare the performance of the proposed multi-scale CNN and other state-of-the-art techniques, such as AlexNet and ResNet. In addition, trainable and non-trainable parameters of the proposed model and the existing techniques are also compared to evaluate the computational efficiency. The experimental results show that the proposed multi-scale CNN model outperforms AlexNet and ResNet in terms of accuracy and efficiency at a lower computational cost. Based on experimental results, it is found that our proposed MCNN2 achieved accuracy and F1-score of 91.2% and 91%, respectively, which is significantly higher than the existing AlexNet and ResNet techniques. Moreover, our findings suggest that the proposed model is more effective and efficient in facilitating clinical research and practice for MRI classification. Full article

Background: The diagnosis of acute spondylodiscitis can be very difficult because clinical onset symptoms are highly variable. The reference examination is MRI, but very often the first diagnostic investigation performed is CT, given its high availability in the acute setting. CT allows rapid evaluation of other alternative diagnoses (e.g., fractures), but scarce literature is available to evaluate the accuracy of CT, and in particular of multi-detector computed tomography (MDCT), in the diagnosis of suspected spondylodiscitis. The aim of our study was to establish MDCT accuracy and how this diagnostic method could help doctors in the depiction of acute spondylodiscitis in an emergency situation by comparing the diagnostic performance of MDCT with MRI, which is the gold standard. Methods: We searched our radiological archive for all MRI examinations of patients who had been studied for a suspicion of acute spondylodiscitis in the period between January 2017 and January 2021 (n = 162). We included only patients who had undergone MDCT examination prior to MRI examination (n = 25). The overall diagnostic value of MDCT was estimated, using MRI as the gold standard. In particular, the aim of our study was to clarify the effectiveness of CT in radiological cases that require immediate intervention (stage of complications). Therefore, the radiologist, faced with a negative CT finding, can suggest an elective (not urgent) MRI with relative serenity and without therapeutic delays. Results: MDCT allowed identification of the presence of acute spondylodiscitis in 13 of 25 patients. Specificity and positive predictive value were 100% for MDCT, while sensitivity and negative predictive value were 68% and 50%, respectively, achieving an overall accuracy of 76%. In addition, MDCT allowed the identification of paravertebral abscesses (92%), fairly pathognomonic lesions of spondylodiscitis pathology. Conclusions: The MDCT allows identification of the presence of acute spondylodiscitis in the Emergency Department (ED) with a satisfactory accuracy. In the case of a positive CT examination, this allows therapy to be started immediately and reduces complications. However, we suggest performing an elective MRI examination in negative cases in which pathological findings are hard to diagnose with CT alone. Full article

Up to 30% of ischemic stroke cases are due to large vessel occlusion (LVO), causing significant morbidity. Studies have shown that the collateral circulation of patients with acute ischemic stroke (AIS) secondary to LVO can predict their clinical and radiological outcomes. The aim of this study is to identify baseline patient characteristics that can help predict the collateral status of these patients for improved triage. In this IRB approved retrospective study, consecutive patients presenting with AIS secondary to anterior circulation LVO were identified between September 2019 and August 2021. The baseline patient characteristics, laboratory values, imaging features and outcomes were collected using a manual chart review. From the 181 consecutive patients initially reviewed, 54 were confirmed with a clinical diagnosis of AIS and anterior circulation LVO. In patients with poor collateral status, the body mass index (BMI) was found to be significantly lower compared to those with good collateral status (26.4 ± 5.6 vs. 31.7 ± 12.3; p = 0.045). BMI of >35 kg/m² was found to predict the presence of good collateral status. Age was found to be significantly higher (70.5 ± 9.6 vs. 58.9 ± 15.6; p = 0.034) in patients with poor collateral status and M1 strokes associated with older age and BMI. Full article

In this case series, we describe a novel ultrasound (US)-guided cervical interlaminar epidural steroid injections (CILESIs) procedure that does not depend on the loss-of-resistance method for epidural space identification. A needle is introduced into three US-identified structures (triple bar sign), the interspinal ligament, ligamentum flavum, and dura mater. The injectants are monitored using superb microvascular imaging during injection. Here, we demonstrate the use of US-guided CILESIs in nine cases and propose the use of sonography, rather than conventional methods, for easier and safer cervical epidural injections. Sonography for direct visualization of cervical epidural injection may allow for outpatient injections. Full article

The human olfactory bulb (OB) has a laminar structure. The segregation of cell populations in the OB image poses a significant challenge because of indistinct boundaries of the layers. Standard 3D visualization tools usually have a low resolution and cannot provide the high accuracy required for morphometric analysis. X-ray phase contrast tomography (XPCT) offers sufficient resolution and contrast to identify single cells in large volumes of the brain. The numerous microanatomical structures detectable in XPCT image of the OB, however, greatly complicate the manual delineation of OB neuronal cell layers. To address the challenging problem of fully automated segmentation of XPCT images of human OB morphological layers, we propose a new pipeline for tomographic data processing. Convolutional neural networks (CNN) were used to segment XPCT image of native unstained human OB. Virtual segmentation of the whole OB and an accurate delineation of each layer in a healthy non-demented OB is mandatory as the first step for assessing OB morphological changes in smell impairment research. In this framework, we proposed an effective tool that could help to shed light on OB layer-specific degeneration in patients with olfactory disorder. Full article

Introduction. Coronavirus SARS-CoV-2, the causative agent of COVID-19, primarily causes a respiratory tract infection that is not limited to respiratory distress syndrome, but it is also implicated in other body systems. Systemic complications were reported due to an exaggerated inflammatory response, which involves severe alveolar damage in the lungs and exacerbates the hypercoagulation that leads to venous thrombosis, ischemic attack, vascular dysfunction and infarction of visceral abdominal organs. Some complications are related to anticoagulant drugs that are administrated to stabilize hypercoagulability, but increase the risk of bleeding, hematoma and hemorrhage. The aim of this study is to report the diagnostic role of CT in the early diagnosis and management of patients with severe COVID-19 complications through the most interesting cases in our experience. Material and Methods. The retrospective analysis of patients studied for COVID-19 in our institution and hospitals, which are part of the university training network, was performed. Cases. Pneumomediastinum, cortical kidney necrosis, splenic infarction, cerebral ischemic stroke, thrombosis of the lower limb and hematomas are the most major complications that are reviewed in this study. Conclusions. Since the onset of the COVID-19 pandemic, the CT imaging modality with its high sensitivity and specificity remains the preferred imaging choice to diagnose early the different complications associated with COVID-19, such as thrombosis, ischemic stroke, infarction and pneumomediastinum, and their management, which significantly improved the outcomes. Full article

Manufacturing a customized mastectomy bra, using medical images obtained for breast cancer treatment, could be suggested as an alternative instead of the anthropometric method. However, the breast shape of a medical image is deformed from the anthropometric method as the measurement posture is different between the anthropometric method for making clothes and the medical image. As a breast consists of adipose tissues and a few muscles without bones, there is a possibility that a bra can be manufactured if the volume is constant. Therefore, a hypothesis was established that the volume of the breast would be constant, even if the measurement methods were different. As a result of the comparison of 3D-SIM and PPM by MRI, 18 items could be measured simultaneously. Nine items showed differences according to the measurement method. The next step in the case of 3D-SIM was calculating the volume by separating the breast shape into a cone and a hemispherical shape; in the case of MRI, an ellipsoidal volume formula was applied. A t-test was performed on the results obtained, showing no significant difference. Therefore, it was proven that the volume of the breast does not change despite the difference in the measurement and the measurement method. Full article

Organ volume measurements are a key metric for managing ADPKD (the most common inherited renal disease). However, measuring organ volumes is tedious and involves manually contouring organ outlines on multiple cross-sectional MRI or CT images. The automation of kidney contouring using deep learning has been proposed, as it has small errors compared to manual contouring. Here, a deployed open-source deep learning ADPKD kidney segmentation pipeline is extended to also measure liver and spleen volumes, which are also important. This 2D U-net deep learning approach was developed with radiologist labeled T2-weighted images from 215 ADPKD subjects (70% training = 151, 30% validation = 64). Additional ADPKD subjects were utilized for prospective (n = 30) and external (n = 30) validations for a total of 275 subjects. Image cropping previously optimized for kidneys was included in training but removed for the validation and inference to accommodate the liver which is closer to the image border. An effective algorithm was developed to adjudicate overlap voxels that are labeled as more than one organ. Left kidney, right kidney, liver and spleen labels had average errors of 3%, 7%, 3%, and 1%, respectively, on external validation and 5%, 6%, 5%, and 1% on prospective validation. Dice scores also showed that the deep learning model was close to the radiologist contouring, measuring 0.98, 0.96, 0.97 and 0.96 on external validation and 0.96, 0.96, 0.96 and 0.95 on prospective validation for left kidney, right kidney, liver and spleen, respectively. The time required for manual correction of deep learning segmentation errors was only 19:17 min compared to 33:04 min for manual segmentations, a 42% time saving (p = 0.004). Standard deviation of model assisted segmentations was reduced to 7, 5, 11, 5 mL for right kidney, left kidney, liver and spleen respectively from 14, 10, 55 and 14 mL for manual segmentations. Thus, deep learning reduces the radiologist time required to perform multiorgan segmentations in ADPKD and reduces measurement variability. Full article

The emergence of the COVID-19 pandemic over a relatively brief interval illustrates the need for rapid data-driven approaches to facilitate clinical decision making. We examined a machine learning process to predict inpatient mortality among COVID-19 patients using clinical and chest radiographic data. Modeling was performed with a de-identified dataset of encounters prior to widespread vaccine availability. Non-imaging predictors included demographics, pre-admission clinical history, and past medical history variables. Imaging features were extracted from chest radiographs by applying a deep convolutional neural network with transfer learning. A multi-layer perceptron combining 64 deep learning features from chest radiographs with 98 patient clinical features was trained to predict mortality. The Local Interpretable Model-Agnostic Explanations (LIME) method was used to explain model predictions. Non-imaging data alone predicted mortality with an ROC-AUC of 0.87 ± 0.03 (mean ± SD), while the addition of imaging data improved prediction slightly (ROC-AUC: 0.91 ± 0.02). The application of LIME to the combined imaging and clinical model found HbA1c values to contribute the most to model prediction (17.1 ± 1.7%), while imaging contributed 8.8 ± 2.8%. Age, gender, and BMI contributed 8.7%, 8.2%, and 7.1%, respectively. Our findings demonstrate a viable explainable AI approach to quantify the contributions of imaging and clinical data to COVID mortality predictions. Full article