Tomography

Objective: This study aimed to assess the diagnostic value of contrast-enhanced ultrasound (CEUS) time–intensity curve (TIC) parameters in detecting non-ossified thyroid cartilage invasion in patients with laryngeal squamous cell carcinoma (SCC). Methods: A CEUS TIC analysis was performed on 32 cases from 27 patients with histologically confirmed laryngeal SCC. The diagnostic performance of time to peak (TTP), peak intensity (PI), wash-in slope (WIS), area under the curve (AUC), and their quantitative differences (∆TTP, ∆PI, ∆WIS, and ∆AUC) to discriminate between the invaded and the non-invaded non-ossified thyroid cartilage was determined using ROC analysis. A logistic regression analysis was employed to identify significant predictors. Results: In an ROC analysis, of all TIC parameters analyzed separately, ∆TTP showed the greatest diagnostic performance (AUC: 0.85). A ∆TTP cut-off of ≤ 8.9 s differentiated between the invaded and the non-invaded non-ossified thyroid cartilage with a sensitivity of 100%, specificity of 76.9%, and accuracy of 81.3%. A combination of ∆TTP and PI increased the AUC to 0.93, specificity to 100%, and accuracy to 96.8%, but reduced the sensitivity to 83.3%. Meanwhile, the visual assessment of enhancement on CEUS to detect cartilage invasion had 83.3% sensitivity and 84.6% specificity. In a univariate logistic regression, only ∆TTP was a significant predictor of non-ossified thyroid cartilage invasion (OR: 0.80; 95% CI: 0.64–1.00). For every second increase in ∆TTP, the probability of thyroid cartilage invasion decreased by 20%. Conclusions: CEUS TIC parameters, particularly a combination of ∆TTP and PI, showed high diagnostic performance in the detection of non-ossified thyroid cartilage invasion in laryngeal SCC. Full article

Background: According to the World Health Organization (WHO), approximately 5% of children and 2.5% of adults suffer from attention deficit hyperactivity disorder (ADHD). This disorder can have significant negative consequences on people’s lives, particularly children. In recent years, methods based on artificial intelligence and neuroimaging techniques, such as MRI, have made significant progress, paving the way for development of more reliable diagnostic tools. In this proof of concept study, our aim was to investigate the potential utility of neuroimaging data and clinical information in combination with a deep learning-based analytical approach, more precisely, a novel feature extraction technique for the diagnosis of ADHD with high accuracy. Methods: Leveraging the ADHD200 dataset, which encompasses demographic information and anatomical MRI scans collected from a diverse ADHD population, our study focused on developing modern deep learning-based diagnostic models. The data preprocessing employed a pre-trained Visual Geometry Group16 (VGG16) network to extract two-dimensional (2D) feature maps from three-dimensional (3D) anatomical MRI data to reduce computational complexity and enhance diagnostic power. The inclusion of personal attributes, such as age, gender, intelligence quotient, and handedness, strengthens the diagnostic models. Four deep-learning architectures—convolutional neural network 2D (CNN2D), CNN1D, long short-term memory (LSTM), and gated recurrent units (GRU)—were employed for analysis of the MRI data, with and without the inclusion of clinical characteristics. Results: A 10-fold cross-validation test revealed that the LSTM model, which incorporated both MRI data and personal attributes, had the best diagnostic performance among all tested models in the diagnosis of ADHD with an accuracy of 0.86 and area under the receiver operating characteristic (ROC) curve (AUC) score of 0.90. Conclusions: Our findings demonstrate that the proposed approach of extracting 2D features from 3D MRI images and integrating these features with clinical characteristics may be useful in the diagnosis of ADHD with high accuracy. Full article

Objectives: Extensive-stage small-cell lung cancer (SCLC) has a poor prognosis, but recently, the combination of immunotherapy and chemotherapy has improved treatment outcomes in some patients, and treatment plans may vary depending on the individual’s general condition and tumor response. In addition, intrathoracic tumor control remains a major challenge for this disease. In the current study, we aim to share our clinical experience and demonstrate that consolidative high-dose thoracic radiotherapy effectively reduces intrathoracic tumor recurrence while maintaining acceptable treatment-related toxicities. Materials and Methods: The medical records of 81 SCLC patients treated at Korea University Guro Hospital from January 2019 to December 2023 were reviewed retrospectively. Among them, 22 patients with extensive-stage SCLC who had a favorable tumor response after systemic therapy, including those with oligo-progressive disease limited to the thoracic region and suitable for curative local therapy, received consolidative radiotherapy. A total dose of 52.5 Gy in 25 fractions was administered over 5 weeks to all patients with extensive-stage SCLC. Results and Conclusions: The median follow-up time was 22 months (range: 8–59 months), with the median follow-up period after completing consolidative radiotherapy being 13 months (range: 4–35 months). In-field local recurrence occurred in only one patient after consolidative thoracic radiotherapy. Most importantly, 10 patients with oligo-progressive disease at the thoracic site, at the time of tumor response, remained stable without further intrathoracic in-field recurrence. Additionally, no severe cases of radiation pneumonitis or esophagitis were observed. Based on our institution’s experience, consolidative high-dose thoracic radiotherapy is well-tolerated and associated with fewer intrathoracic recurrences, leading to improved long-term survival in carefully selected patients with extensive-stage SCLC. Given these findings, we believe consolidative radiotherapy should be considered more proactively in clinical practice. Furthermore, these results may help guide the design of future clinical trials. Full article

Background. Statins appear to be useful in patients with acute ischemic stroke. Our aim was to evaluate the association between premorbid statin treatment and CT perfusion characteristics of acute ischemic stroke. Methods. A retrospective analysis of patients with acute stroke secondary to occlusion of large vessels in the anterior circulation was performed to assess collateral flow, ischemic core volume, and ischemic penumbra using CT angiography and CT perfusion maps. Fisher’s exact test was used to compare baseline characteristics of patients in the two groups. The Wilcoxon rank-sum test for independent groups was used to compare all variables obtained for the two different groups with and without statin use. Results. We identified 61 patients, including 29 treated with statins and 32 not treated with statins before stroke onset matched by age, gender, and vascular risk factors except for hypercholesterolemia. The statin group showed lower National Institutes of health Stroke Scale scores at onset (14 ± 6.1 vs. 16 ± 4.5; p = 0.04) and lower volumes of brain tissue characterized by impaired cerebral blood flow (CBF), cerebral blood volume (CBV), and ${\mathrm{T}}_{\mathrm{m}\mathrm{a}\mathrm{x}}^{9.5-25\mathrm{s}}$; otherwise, no statistically significant difference was found in the volume of the ${\mathrm{T}}_{\mathrm{m}\mathrm{a}\mathrm{x}}^{16-25\mathrm{s}}$ between the two groups. Conclusions. Premorbid statin treatment is associated with a favorable imaging condition of acute ischemic stroke in terms of ischemic core and ischemic penumbra volume. Full article

Objectives: Shoulder injury related to vaccine administration (SIRVA), previously observed with influenza vaccines, has gained clinical significance with widespread COVID-19 vaccination. However, few studies correlate vaccine types and demographic factors with the MRI findings of SIRVA. This study aimed to evaluate MRI findings of SIRVA following COVID-19 vaccination and assess associations with vaccine type and patient characteristics. Methods: A retrospective cohort study was conducted on 35 patients with new-onset shoulder complaints within six weeks of COVID-19 vaccination between May 2021 and May 2022. MRI findings suggestive of SIRVA were reviewed, including subacromial bursitis, rotator cuff tears, and adhesive capsulitis. Demographic data, vaccine type, clinical symptoms, and treatments were collected. Follow-up interviews (1–30 September 2024) assessed symptom persistence and vaccine hesitancy. Descriptive statistics and Chi-square tests were used to explore associations. Results: Of the 35 patients (mean age 53.6 ± 9.0 years; 54.3% female), subacromial bursitis was the most common MRI finding (89.5%), followed by tendonitis (47.4%) and adhesive capsulitis (36.8%). Tendonitis correlated with older age (p = 0.024) and AstraZeneca vaccination (p = 0.033). Subacromial bursitis was linked to female sex (p = 0.013) and higher BMI (p = 0.023). Adhesive capsulitis was associated with receiving the Sinopharm vaccine (p = 0.029). Persistent symptoms (22.9%) were more common in younger patients, women, and those with right-sided injections. Conclusions: SIRVA following COVID-19 vaccination showed different MRI patterns associated with female sex, higher BMI, and vaccine type. Awareness of these patterns may expedite recognition of COVID-19-associated SIRVA in routine practice. Full article

Medical image segmentation is a critical application of computer vision in the analysis of medical images. Its primary objective is to isolate regions of interest in medical images from the background, thereby assisting clinicians in accurately identifying lesions, their sizes, locations, and their relationships with surrounding tissues. However, compared to natural images, medical images present unique challenges, such as low resolution, poor contrast, inconsistency, and scattered target regions. Furthermore, the accuracy and stability of segmentation results are subject to more stringent requirements. In recent years, with the widespread application of Convolutional Neural Networks (CNNs) in computer vision, deep learning-based methods for medical image segmentation have become a focal point of research. This paper categorizes, reviews, and summarizes the current representative methods and research status in the field of medical image segmentation. A comparative analysis of relevant experiments is presented, along with an introduction to commonly used public datasets, performance evaluation metrics, and loss functions in medical image segmentation. Finally, potential future research directions and development trends in this field are predicted and analyzed. Full article

Objectives: To evaluate the medical image segmentation performance of monochromatic images in various energy levels. Methods: The low-density module (25 mm in diameter, 6 Hounsfield Unit (HU) in density difference from background) from the ACR464 phantom was scanned at both 10 mGy and 5 mGy dose levels. Virtual monoenergetic images (VMIs) at different energy levels of 40, 50, 60, 68, 74, and 100 keV were generated. The images at 10 mGy reconstructed with 50% adaptive statistical iterative reconstruction veo (ASIR-V50%) were used to train an image segmentation model based on U-Net. The evaluation set used 5 mGy VMIs reconstructed with various reconstruction algorithms: FBP, ASIR-V50%, ASIR-V100%, deep learning image reconstruction (DLIR) with low (DLIR-L), medium (DLIR-M), and high (DLIR-H) strength levels. U-Net was employed as a tool to compare algorithm performance. Image noise and segmentation metrics, such as the DICE coefficient, intersection over union (IOU), sensitivity, and Hausdorff distance, were calculated to assess both image quality and segmentation performance. Results: DLIR-M and DLIR-H consistently achieved lower image noise and better segmentation performance, with the highest results observed at 60 keV, and DLIR-H had the lowest image noise across all energy levels. The performance metrics, including IOU, DICE, and sensitivity, were ranked in descending order with energy levels of 60 keV, 68 keV, 50 keV, 74 keV, 40 keV, and 100 keV. Specifically, at 60 keV, the average IOU values for each reconstruction method were 0.60 for FBP, 0.67 for ASIR-V50%, 0.68 for ASIR-V100%, 0.72 for DLIR-L, 0.75 for DLIR-M, and 0.75 for DLIR-H. The average DICE values were 0.75, 0.80, 0.82, 0.83, 0.85, and 0.86. The sensitivity values were 0.93, 0.91, 0.96, 0.95, 0.98, and 0.98. Conclusions: For low-density, non-enhancing objects under a low dose, the 60 keV VMIs performed better in automatic segmentation. DLIR-M and DLIR-H algorithms delivered the best results, whereas DLIR-H provided the lowest image noise and highest sensitivity. Full article

Background/Objectives: Developing a treatment strategy that effectively prolongs the lives of people with brain tumors requires an accurate diagnosis of the condition. Therefore, improving the preoperative classification of meningiomas is a priority. Machine learning (ML) has made great strides thanks to the development of convolutional neural networks (CNNs) and computer-aided tumor detection systems. The deep convolutional layers automatically extract important and dependable information from the input space, in contrast to more traditional neural network layers. One recent and promising advancement in this field is ML. Still, there is a dearth of studies being carried out in this area. Methods: Therefore, starting with the analysis of magnetic resonance images, we have suggested in this research work a tried-and-tested and methodical strategy for real-time meningioma diagnosis by image segmentation using a very deep transfer learning CNN model or DNN model (VGG-16) with CUDA. Since the VGGNet CNN model has a greater level of accuracy than other deep CNN models like AlexNet, GoogleNet, etc., we have chosen to employ it. The VGG network that we have constructed with very small convolutional filters consists of 13 convolutional layers and 3 fully connected layers. Our VGGNet model takes in an sMRI FLAIR image input. The VGG’s convolutional layers leverage a minimal receptive field, i.e., 3 × 3, the smallest possible size that still captures up/down and left/right. Moreover, there are also 1 × 1 convolution filters acting as a linear transformation of the input. This is followed by a ReLU unit. The convolution stride is fixed at 1 pixel to keep the spatial resolution preserved after convolution. All the hidden layers in our VGG network also use ReLU. A dataset consisting of 264 3D FLAIR sMRI image segments from three different classes (meningioma, tuberculoma, and normal) was employed. The number of epochs in the Sequential Model was set to 10. The Keras layers that we used were Dense, Dropout, Flatten, Batch Normalization, and ReLU. Results: According to the simulation findings, our suggested model successfully classified all of the data in the dataset used, with a 99.0% overall accuracy. The performance metrics of the implemented model and confusion matrix for tumor classification indicate the model’s high accuracy in brain tumor classification. Conclusions: The good outcomes demonstrate the possibility of our suggested method as a useful diagnostic tool, promoting better understanding, a prognostic tool for clinical outcomes, and an efficient brain tumor treatment planning tool. It was demonstrated that several performance metrics we computed using the confusion matrix of the previously used model were very good. Consequently, we think that the approach we have suggested is an important way to identify brain tumors. Full article

Background: This study aims to systematically evaluate the findings from computed tomography (CT) examinations conducted at least three months post-diagnosis of COVID-19 in patients diagnosed between 2020 and 2024. Objective: To determine the frequency and characteristics of CT findings in the post-COVID-19 period, analyze long-term effects on lung parenchyma, and contribute to the development of clinical follow-up and treatment strategies based on the collected data. Materials and Methods: Ethical approval was obtained for this retrospective study, and individual consent was waived. A total of 76 patients were included in the study, aged 18 and older, diagnosed with COVID-19 between March 2020 and November 2024, who underwent follow-up chest CT scans at 3–6 months, 6–12 months, and/or 12 months post-diagnosis. CT images were obtained in the supine position without contrast and evaluated by two experienced radiologists using a CT severity score (CT-SS) system, which quantifies lung involvement. Statistical analyses were performed using IBM SPSS 23.0, with significance set at p < 0.05. Results: The results indicated a mean CT-SS of 10.58 ± 0.659. Significant associations were found between age, CT scores, and the necessity for intensive care or mechanical ventilation. The most common CT findings included ground-glass opacities, reticular patterns, and traction bronchiectasis, particularly increasing with age and over time. Conclusion: This study emphasizes the persistent alterations in lung parenchyma following COVID-19, highlighting the importance of continuous monitoring and tailored treatment strategies for affected patients to improve long-term outcomes. Full article

Background: Deep learning (DL) reconstruction techniques have shown promise in reducing MRI acquisition times while maintaining image quality. However, the impact of different acceleration factors on diagnostic accuracy in shoulder MRI remains unexplored in clinical practice. Purpose: The purpose of this study was to evaluate the diagnostic accuracy of 2-fold and 4-fold DL-accelerated shoulder MRI protocols compared to standard protocols in clinical practice. Materials and Methods: In this prospective single-center study, 88 consecutive patients (49 males, 39 females; mean age, 51 years) underwent shoulder MRI examinations using standard, 2-fold (DL2), and 4-fold (DL4) accelerated protocols between June 2023 and January 2024. Four independent radiologists (experience range: 4–25 years) evaluated the presence of bone marrow edema (BME), rotator cuff tears, and labral lesions. The sensitivity, specificity, and interobserver agreement were calculated. Diagnostic confidence was assessed using a 4-point scale. The impact of reader experience was analyzed by stratifying the radiologists into ≤10 and >10 years of experience. Results: Both accelerated protocols demonstrated high diagnostic accuracy. For BME detection, DL2 and DL4 achieved 100% sensitivity and specificity. In rotator cuff evaluation, DL2 showed a sensitivity of 98–100% and specificity of 99–100%, while DL4 maintained a sensitivity of 95–98% and specificity of 99–100%. Labral tear detection showed perfect sensitivity (100%) with DL2 and slightly lower sensitivity (89–100%) with DL4. Interobserver agreement was excellent across the protocols (Kendall’s W = 0.92–0.98). Reader experience did not significantly impact diagnostic performance. The area under the ROC curve was 0.94 for DL2 and 0.90 for DL4 (p = 0.32). Clinical Implications: The implementation of DL-accelerated protocols, particularly DL2, could improve workflow efficiency by reducing acquisition times by 50% while maintaining diagnostic reliability. This could increase patient throughput and accessibility to MRI examinations without compromising diagnostic quality. Conclusions: DL-accelerated shoulder MRI protocols demonstrate high diagnostic accuracy, with DL2 showing performance nearly identical to that of the standard protocol. While DL4 maintains acceptable diagnostic accuracy, it shows a slight sensitivity reduction for subtle pathologies, particularly among less experienced readers. The DL2 protocol represents an optimal balance between acquisition time reduction and diagnostic confidence. Full article

Background/Objectives: Reviewing the entire history of imaging exams of a single patient’s records is an essential step in clinical practice, but it is time and resource consuming, with potential negative effects on workflow and on the quality of medical decisions. The main objective of this study was to evaluate the applicability of three open-source large language models (LLMs) for the automatic generation of concise summaries of patient’s imaging records. Secondary objectives were to assess correlations among the LLMs and to evaluate the length reduction provided by each model. Methods: Three state-of-the-art open-source large language models were selected: Llama 3.2 11B, Mistral 7B, and Falcon 7B. Each model was given a set of radiology reports. The summaries produced by the models were evaluated by two experienced radiologists and one experienced clinical physician using standardized metrics. Results: A variable number of radiological reports (n = 12–56) from four patients were selected and evaluated. The summaries generated by the three LLM showed a good level of accuracy compared with the information contained in the original reports, with positive ratings on both clinical relevance and ease of reference. According to the experts’ evaluations, the use of the summaries generated by LLMs could help to reduce the time spent on reviewing the previous imaging examinations performed, preserving the quality of clinical data. Conclusions: Our results suggest that LLMs are able to generate summaries of the imaging history of patients, and these summaries could improve radiology workflow making it easier to manage large volumes of reports. Full article

Background: When plotting data points corresponding to the contrast-agent-induced change in transverse relaxation rate from a dynamic gradient-echo (GRE) magnetic resonance imaging (MRI) study versus a corresponding spin-echo study, a loop or vortex curve rather than a reversible line is formed. The vortex curve area is likely to reflect vessel architecture and oxygenation level. In this study, the vortex effect seen when using only GRE-based estimates, i.e., contrast-agent concentration based on GRE transverse relaxation rate and contrast-agent concentration based on quantitative susceptibility mapping (QSM), was investigated. Methods: Twenty healthy volunteers were examined using 3 T MRI. Magnitude and phase dynamic contrast-enhanced MRI (DSC-MRI) data were obtained using GRE echo-planar imaging. Vortex curves for grey-matter (GM) regions and for arterial input function (AIF) data were constructed by plotting concentration based on GRE transverse relaxation rate versus concentration based on QSM. Vortex parameters (vortex area and normalised vortex width) were compared with QSM-based whole-brain OEF estimates obtained using 3D GRE. Results: An obvious vortex effect was observed, and both GM vortex parameters showed a moderate and significant correlation with OEF (r = −0.51, p = 0.02). The vortex parameters for AIF data showed no significant correlation with OEF. Conclusions: GRE-based GM vortex parameters correlated significantly with whole-brain OEF. In agreement with expectations, the corresponding AIF data, representing high fractions of arterial blood, showed no significant correlation. Novel parameters, based solely on standard GRE protocols, are of relevance to investigate, considering that GRE-based DSC-MRI is very common in brain tumour applications. Full article

Background: The radiological evaluation of the carotid sinus (CS) anatomy and its morphometry is essentially important for various surgical procedures involving the carotid bifurcation and the CS itself. Despite its tremendous clinical significance, studies dealing with the CS anatomy are seldom reported. Hence, the present study aimed to evaluate the frequencies of the CS positional variants and their morphometry and correlate them with age and body mass index (BMI). Methods: In this retrospective cross-sectional study, a total of 754 disease-free carotid arteries were examined using computed tomography angiography scans to determine the CS positional variations (such as types I to III) and its morphometry, including the CS diameter and length. Additionally, the association between these parameters and factors such as sex, age, and body mass index were explored using appropriate statistical tests. The inter-rater agreement of the collected dataset was evaluated using Cohen’s Kappa. Results: The CS type I was observed in 87.67% of the cases, and type II and type III were observed at lower frequencies with 9.02% and 3.32%, respectively. There were statistically significant (p < 0.001) differences observed in the mean diameter and length of the sinus between the sex and the type I CS variations. However, there was no significant and strong correlation between the age and BMI factors with sinus length and sinus diameter. The kappa values for inter-rater agreement ranged from 0.77 to 0.99 for all parameters. Conclusions: In type I, the CS length and carotid vessel’s diameter is significantly different between the sexes. However, age and BMI do not affect the CS anatomy in radiologically disease-free carotid arteries. Knowledge of the CS variant anatomy is clinically significant as it influences the patients’ surgical and physiological outcomes. Full article

Background: Rectal cancer represents a major cause of mortality in the United States. Management strategies are highly individualized, depending on patient-specific factors and tumor characteristics. The therapeutic landscape is rapidly evolving, with notable advancements in response rates to both radiotherapy and chemotherapy. For locally advanced rectal cancer (LARC, defined as up to T3–4 N+), the standard of care involves total mesorectal excision (TME) following neoadjuvant chemoradiotherapy (nCRT). Magnetic resonance imaging (MRI) has emerged as the gold standard for local tumor staging and is increasingly pivotal in post-treatment restaging. Aim: In our study, we proposed an MRI-based radiomic model to identify characteristic features of peritumoral mesorectal fat in two patient groups: good responders and poor responders to neoadjuvant therapy. The aim was to assess the potential presence of predictive factors for favorable or unfavorable responses to neoadjuvant chemoradiotherapy, thereby optimizing treatment management and improving personalized clinical decision-making. Methods: We conducted a retrospective analysis of adult patients with LARC who underwent pre- and post-nCRT MRI scans. Patients were classified as good responders (Group 0) or poor responders (Group 1) based on MRI findings, including tumor volume reduction, signal intensity changes on T2-weighted and diffusion-weighted imaging (DWI), and alterations in the circumferential resection margin (CRM) and extramural vascular invasion (EMVI) status. Classification criteria were based on the established literature to ensure consistency. Key clinical and imaging parameters, such as age, TNM stage, CRM involvement, and EMVI presence, were recorded. A radiomic model was developed using the LASSO algorithm for feature selection and regularization from 107 extracted radiomic features. Results: We included 44 patients (26 males and 18 females) who, following nCRT, were categorized into Group 0 (28 patients) and Group 1 (16 patients). The pre-treatment MRI analysis identified significant features (out of 107) for each sequence based on the Mann–Whitney test and t-test. The LASSO algorithm selected three features (shape_Sphericity, shape_Maximum2DDiameterSlice, and glcm_Imc2) for the construction of the radiomic logistic regression model, and ROC curves were subsequently generated for each model (AUC: 0.76). Conclusions: We developed an MRI-based radiomic model capable of differentiating and predicting between two groups of rectal cancer patients: responders and non-responders to neoadjuvant chemoradiotherapy (nCRT). This model has the potential to identify, at an early stage, lesions with a high likelihood of requiring surgery and those that could potentially be managed with medical treatment alone. Full article

Objectives: This study aims to assess the potential role of the ultrasound (US) monochromatic Superb Microvascular Index (mSMI) to predict malignancy of solid focal lesions, correlating the vascular index (VI) with bioptic histological results. Methods: In this single-center retrospective analysis, patients undergoing percutaneous US-guided biopsy of solid lesions were considered. Biopsy indication was given by a multidisciplinary team evaluation based on clinical radiological data. Exclusion criteria were: unfeasible SMI evaluations due to poor respiratory compliance, locations not appreciable with the SMI, previous antiangiogenetic chemo/immunotherapies, and inconclusive histological reports. The mSMI examination was conducted in order to visualize extremely low-velocity flows with a high resolution and high frame rate; the VI was semi-automatically calculated. All bioptic procedures were performed under sole US guidance using 16G or 18G needles, immediately after mSMI assessment. Results: Forty-four patients were included (mean age: 64 years; 27 males, 17 females). Liver (15/43), kidneys (9/43), and lymph nodes (6/43) were the most frequent targets. At histopathological analysis, 7 lesions were benign and 37 malignant, metastasis being the most represented. The VI calculated in malignant lesions was statistically higher compared to benign lesions (35.45% and 11% in malignant and benign, respectively; p-value 0.013). A threshold VI value of 15.4% was identified to differentiate malignant lesions. The overall diagnostic accuracy of the VI with the mSMI was 0.878, demonstrating a high level of diagnostic accuracy. Conclusions: In this study, the mSMI analysis of solid focal lesions undergoing percutaneous biopsy significantly correlated with histological findings in terms of malignant/benign predictive value, reflecting histological vascular changes in malignant lesions. Full article

Purpose: To evaluate the variability of oncogenic risk related to radiation exposure in patients frequently exposed to ionizing radiation for diagnostic purposes, specifically ICU patients, according to different risk models, including the BEIR VII, ICRP 103, and US EPA models. Methods: This was an IRB-approved observational retrospective study. A total of 71 patients (58 male, 13 female; median age, 66 years; interquartile range [IQR], 65–71 years) admitted to the ICU who underwent X-ray examinations between 1 October 2021 and 28 February 2023 were included. For each patient, the cumulative effective dose during a single hospital admission was calculated. Lifetime attributable risk (LAR) was estimated based on the BEIR VII, ICRP 103, and US EPA risk models to calculate additional oncogenic risk related to radiation exposure. The Friedman test for repeated-measures analysis of variance was used to compare risk values between different models. The intraclass correlation coefficient (ICC) was used to assess the consistency of risk values between different models. Results: Different organ, leukemia, and all-cancer risk values estimated according to different oncogenic risk models were significantly different, but the intraclass correlation coefficient revealed a good (>0.75) or even excellent (>0.9) agreement between different risk models. The ICRP 103 model estimated a lower all-cancer (median 69.05 [IQR 30.35–195.37]) and leukemia risk (8.22 [3.02–27.93]) compared to the US EPA (all-cancer: 139.68 [50.51–416.16]; leukemia: 23.34 [3.47–64.37]) and BEIR VII (all-cancer: 162.08 [70.6–371.40]; leukemia: 24.66 [12.9–58.8]) models. Conclusions: Cancer risk values were significantly different between risk models, though inter-model agreement in the consistency of risk values was found to be good, or even excellent. Full article

Introduction: An efficient pipeline for rosette trajectory magnetic resonance imaging reconstruction is proposed, combining the inverse Fourier transform with a vision transformer (ViT) network enhanced with a convolutional layer. This method addresses the challenges of reconstructing high-quality images from non-Cartesian data by leveraging the ViT’s ability to handle complex spatial dependencies without extensive preprocessing. Materials and Methods: The inverse fast Fourier transform provides a robust initial approximation, which is refined by the ViT network to produce high-fidelity images. Results and Discussion: This approach outperforms established deep learning techniques for normalized root mean squared error, peak signal-to-noise ratio, and entropy-based image quality scores; offers better runtime performance; and remains competitive with respect to other metrics. Full article

Background: The updated definition of a TIA emphasizes tissue characteristics rather than symptom duration, defining a TIA as a transient neurological episode without ischemic lesions in brain imaging, including in DWI. If imaging reveals a lesion, even in patients with transient symptoms, the event is reclassified as a minor ischemic stroke. Objective: This retrospective observational study aimed to determine the prevalence of ischemic lesions in DWI in patients with a TIA diagnosis. Results: Adults aged 18–90 years, diagnosed with a TIA by a neurologist and who underwent MRI-DWI at CMT hospital within the first week after symptom onset (May 2023–July 2024), were included. Ethical approval was obtained. Descriptive statistics summarized patient demographics, clinical features, Fazekas scale grades, and imaging findings. Conclusions: Among the 26 patients clinically diagnosed with TIAs, 7 (26.9%) exhibited ischemic lesions in DWI, reclassifying these cases as minor ischemic strokes under the updated definition. The prevalence of ischemic lesions was notably higher in patients with comorbidities such as hypertension and diabetes. These findings highlight the importance of early MRI-DWI to accurately distinguish TIAs from minor ischemic strokes. Routine urgent DWI within the first week of TIA symptoms enhances diagnosis and risk stratification and can guide targeted stroke prevention strategies, particularly when combined with the ABCD2 score. Full article

Cardiac computed tomography (CT) has rapidly advanced, becoming an invaluable tool for diagnosing and prognosticating various cardiovascular diseases. While echocardiography and cardiac magnetic resonance imaging (CMR) remain the gold standards for myocardial assessment, modern CT technologies offer enhanced spatial resolution, making it an essential tool in clinical practice. Cardiac CT has expanded beyond coronary artery disease evaluation, now playing a key role in assessing cardiomyopathies and structural heart diseases. Innovations like photon-counting CT enable precise estimation of myocardial extracellular volume, facilitating the detection of infiltrative disorders and myocardial fibrosis. Additionally, CT-based myocardial strain analysis allows for the classification of impaired myocardial contractility, while quantifying cardiac volumes and function remains crucial in cardiomyopathy evaluation. This review explores the emerging role of cardiac CT in cardiomyopathy phenotyping, emphasizing recent technological advancements. Full article