Radiologia Medica最新文献

Purpose: Assessment of the stability of intracranial aneurysms is important in the clinic but remains challenging. The aim of this study was to construct a deep learning model (DLM) to identify unstable aneurysms on computed tomography angiography (CTA) images.

Methods: The clinical data of 1041 patients with 1227 aneurysms were retrospectively analyzed from August 2011 to May 2021. Patients with aneurysms were divided into unstable (ruptured, evolving and symptomatic aneurysms) and stable (fortuitous, nonevolving and asymptomatic aneurysms) groups and randomly divided into training (833 patients with 991 aneurysms) and internal validation (208 patients with 236 aneurysms) sets. One hundred and ninety-seven patients with 229 aneurysms from another hospital were included in the external validation set. Six models based on a convolutional neural network (CNN) or logistic regression were constructed on the basis of clinical, morphological and deep learning (DL) features. The area under the curve (AUC), accuracy, sensitivity and specificity were calculated to evaluate the discriminating ability of the models.

Results: The AUCs of Models A (clinical), B (morphological) and C (DL features from the CTA image) in the external validation set were 0.5706, 0.9665 and 0.8453, respectively. The AUCs of Model D (clinical and DL features), Model E (clinical and morphological features) and Model F (clinical, morphological and DL features) in the external validation set were 0.8395, 0.9597 and 0.9696, respectively.

Conclusions: The CNN-based DLM, which integrates clinical, morphological and DL features, outperforms other models in predicting IA stability. The DLM has the potential to assess IA stability and support clinical decision-making.

Purpose: To investigate the clinical feasibility of a CT collateral map compared with an MRA collateral map, focusing on collateral perfusion (CP) estimation and baseline lesion assessment in acute ischemic stroke (AIS).

Materials and methods: This retrospective analysis used selected data from a prospectively collected database. We generated CT collateral maps derived from CT perfusion, encompassing images of arterial, capillary, early venous (CMEV), late venous, and delay phases. Three raters assessed CP scores from MRA and CT collateral maps and CMEV lesion volumes. Lesion volumes of baseline diffusion-weighted imaging (bDWI) and cerebral blood flow rate (CBF) < 30% were automatically measured by the software. The agreement between MRA and CT collateral maps in CP estimation and the correlation between lesion volumes with a CBF < 30% and the CMEV for bDWI lesion volumes were analyzed.

Results: One-hundred ten patients (mean age ± standard deviation, 71 ± 14; 60 women) with AIS due to steno-occlusion of the internal carotid and/or middle cerebral arteries were included. The agreement between the MRA and CT collateral maps in CP grading was excellent (weighted κ = 0.93; 95% CI, 0.90-0.97). The concordance correlation coefficients (CCCs) of the CBF < 30% and CMEV for bDWI lesion volumes were 0.76 (95% CI, 0.60-0.91) and 0.97 (0.95-0.98), respectively.

Conclusion: The clinical feasibility of the CT collateral map is demonstrated by its significant correlation with the MRA collateral map in CP estimation and baseline lesion assessment.

Purpose: The aim of this study was to assess the feasibility of quantifying morphologic changes in tumors during immunotherapy, as a reflection of response or survival.

Methods and materials: A retrospective single-center analysis was performed in patients with unresectable liver cancer previously enrolled in clinical trials combining immunotherapy (tremelimumab ± durvalumab) and locoregional treatment (either ablation or transarterial chemoembolization). Conventional response (RECIST 1.1) was assessed at 6-month follow-up. For morphologic assessment, the largest target lesion was manually segmented on axial slices in two dimensions using contrast-enhanced CT. Solidity and circularity of tumors were calculated at baseline, 3-month follow-up, and at 6-months follow-up. Survival analysis was performed.

Results: From the 68 patients enrolled in clinical trials, 28 did not have target lesions separate from lesions treated by locoregional therapies, and 3 had no follow-up imaging. Thirty-seven patients (9 with biliary cancer and 28 with hepatocellular carcinoma) were included. Shape features and shape variation were not correlated with RECIST 1.1 status at 6-month follow-up. However, patients with low solidity tumors at 6-month follow-up showed poorer prognosis compared with patients with high solidity tumors at 6-month follow-up (p = 0.01). Solidity variation analysis confirmed that a decrease of tumor solidity at 6-month follow-up was associated with poorer prognosis (p = 0.01). No association was found between shape features at baseline or shape features at 3-month follow-up with overall survival.

Conclusion: Evolution and variation of tumor morphology during treatment may reflect or correlate with outcomes and contribute toward adapted response criteria.

Objective: To evaluate whether breast parenchymal features of the contralateral breast on preoperative MRI are associated with primary breast cancer characteristics and disease-free survival (DFS) in women with invasive breast cancer.

Materials and methods: Women with newly diagnosed invasive breast cancer who underwent preoperative breast MRI followed by surgery were retrospectively evaluated. Background parenchymal enhancement (BPE) on dynamic contrast-enhanced MRI and background diffusion signal (BDS) on diffusion-weighted MRI of the contralateral breast were qualitatively assessed using a four-category scale: minimal, mild, moderate, or marked. Primary breast cancer characteristics were compared based on the degree of BPE or BDS. Cox proportional hazards models were used to evaluate the association between MRI parenchymal features and DFS after adjusting for clinicopathologic features.

Results: A total of 515 women (mean age, 54 years) were included. Of whom, 46 (8.9%) patients who developed disease recurrence at a median follow-up of 60 months were observed. A high level (moderate/marked) of BPE or BDS was associated with younger age (≤ 45) and premenopausal status (all P < 0.05) compared to a low level (minimal/mild), but it was not associated with primary cancer characteristics such as tumor stage, grade, or subtype. Multivariable Cox proportional hazards analysis demonstrated that larger tumor size (> 2 cm) (hazard ratio [HR], 3.877; P < . 001), triple-negative subtype (HR, 2.440; P = .013), and axillary node metastasis (HR, 1.823; P = .049) were associated with worse DFS. No associations were observed between background parenchymal features and disease outcomes.

Conclusions: MRI parenchymal features, including BPE and BDS, of the contralateral breast showed no associations with primary breast cancer characteristics or DFS in women with invasive breast cancer.

Background: Prostate cancer ranks as the second most common cancer in men worldwide. Dose escalation to the tumor and/or the prostate improves biochemical recurrence-free survival. However, interobserver variability in lesion contouring poses a significant limitation to such therapeutic approaches. Therefore, a study of factors influencing this variability is necessary.

Materials and methods: Three independent readers delineated the index prostate lesion (DIL) using T2w and ADC sequences for each patient. Clinical data were retrospectively collected for all participants. Logistic regression analysis was employed to examine the correlation between clinical features and a mean DICE coefficient > 0.7. Features exhibiting a p value < 0.05 on univariate analysis were subjected to multivariate analysis.

Results: The study comprised 68 patients, with a median DICE coefficient of 0.69 (95% CI 0.65-0.71), wherein 42.6% (29/68) attained a mean DICE > 0.7. Univariate analysis identified the PI-QUAL score, maximum diameter of DIL, and mean DIL volume as significant (p < 0.05) predictors. In multivariate analysis, only the PI-QUAL score (p = 0.008) remained statistically associated with a DICE coefficient > 0.7.

Conclusion: The PI-QUAL score emerges as the primary predictive factor for minimizing inter-reader variability in intraprostatic dominant lesion segmentation. These findings underscore the importance of considering PI-QUAL scores when devising focal treatment plans. Adoption of a multi-reader approach involving diverse medical specialists (radiologists, radiotherapists, urologists) is advocated, particularly for MRIs with low PI-QUAL scores.

Clinical relevance statement: Radiotherapy is a major treatment for patients with localized prostate cancer. Dose escalation to the tumor leads to improved cancer control. Precise delineation of the dominant intraprostatic lesion (DIL) remains a limitation to focal treatments. Features influencing inter-reader variability were never evaluated. In this study, we identified that the PI-QUAL score was the sole predictor of the inter-reader delineation variability of the DIL.

Β-thalassemia major (TM) patients underwent regular transfusions to prevent complications of chronic anemia. However, these regular transfusions result in progressive iron accumulation in vital organs, including the heart. Myocardial iron overload can lead to cardiac dysfunction and ultimately to heart failure. Diagnosis of cardiac dysfunction in β-TM patients is usually made through clinical examination, electrocardiogram, and echocardiography. Cardiac magnetic resonance (CMR), through the measurement of T2* relaxation time, represents the diagnostic modality of choice for assessing myocardial iron overload and guiding the iron chelation therapy. Despite a tailored chelation therapy reducing myocardial iron overload, heart failure remains the leading cause of morbidity and mortality even in well-treated β-TM patients. Advances in CMR, including myocardial strain, parametric mapping (T1, T2, and extracellular volume), and late gadolinium enhancement (LGE) measurements, have expanded its role in the diagnosis, prognosis, and follow-up of these patients. This review seeks to offer a thorough overview of the potential uses of CMR in β-TM, extending beyond the established role of T2* measurement in guiding chelation therapy. It delves into the emerging applications of new CMR imaging biomarkers that could improve the overall management of β-TM patients.

Purpose: To assess the diagnostic performance and inter-observer agreement of a PREoperative sarcoma scoring based on ultrasound (PRESS-US) in differentiating uterine leiomyosarcoma (uLMS) from leiomyoma (LM).

Methods: We conducted a retrospective evaluation of patients who underwent surgery and received standardized ultrasound examinations due to the presence of uterine myoma-like masses. Histological diagnosis was used as the reference standard. The masses were analyzed using morphological uterus sonographic assessment criteria, and the diagnostic accuracy of PRESS-US was evaluated using ROC curve analysis. Kappa (κ) statistics were used to assess the inter-observer agreement between a less experienced and an expert radiologist.

Results: Among the 646 patients, 632 (97.8%) were diagnosed with LM, and 14 (2.2%) had uLMS. The malignancy rates for low-risk and high-risk patients were 0.35% and 13.48%, respectively. The optimal PRESS-US cutoff was 17.5, resulting in an AUC of 89.7% (95% CI, 0.79-1.00), with a sensitivity of 85.7% and a specificity of 87.8%. The inter-observer agreement between a less experienced and an expert radiologist was excellent (κ = 0.811, P < 0.001).

Conclusions: PRESS-US provides effective risk stratification for uLMS for radiologists with different levels of experience, with high reliability. Subgrouping high-risk patients helps in better risk stratification.

Objective: This study, derived from the phase 3 SAFE trial (ClinicalTrials.gov identifier: NCT2236806), explores subclinical cardiac damage in breast cancer patients receiving anthracycline-based chemotherapy and left-sided breast radiation therapy (RT).

Materials and methods: Eligible patients were randomized to a cardioprotective pharmacological therapy (bisoprolol, ramipril, or both) or placebo, with cardiac surveillance at multiple time-point using standard and 3-dimensional echocardiography. Dosimetric parameters were analysed, including mean heart dose (MHD) and various metrics for heart substructures, employing advanced contouring techniques and auto-contouring software.

Results: In the analysis of left-sided breast RT patients, the study encompassed 39 out of 46 irradiated individuals, focusing on GLS and 3D-LVEF outcomes with ≥ 10% worsening, defined as subclinical heart damage. Distinct RT schedules were used, with placebo exhibiting the highest ≥ 10% worsening (36.4%). In terms of treatment arms, bisoprolol exhibited 11.1% worsening, while ramipril 16.7% and bisoprolol + ramipril 25%. For patients with no subclinical damage, the mean MHD was 1.5 Gy; for patients with subclinical heart damage, the mean MHD was 1.6 Gy (p = 0.94). Dosimetric parameters related to heart and heart substructures (left anterior descending artery, right and left atrium, right and left ventricle) showed no statistically significant differences between patients with and without subclinical damage.

Conclusion: Our results emphasize the crucial role of cardioprotective measures in mitigating adverse effects, highlighting RT as having negligible influence on cardiac performance. An extended follow-up assessment of the whole series is warranted to determine whether a subclinical effect could significantly influence clinical outcomes and cardiac events.

Large language models like ChatGPT, with their growing accessibility, are attracting increasing interest within the artificial intelligence medical field, particularly in the analysis of radiology reports. These present a valuable opportunity to explore the potential clinical applications of large language models, given their huge capabilities in processing and understanding written language. Early research indicates that ChatGPT could offer benefits in radiology reporting. ChatGPT can assist but not replace radiologists in achieving diagnoses, generating structured reports, extracting data, identifying errors or incidental findings, and can also serve as a support in creating patient-friendly reports. However, ChatGPT also has intrinsic limitations, such as hallucinations, stochasticity, biases, deficiencies in complex clinical scenarios, data privacy and legal concerns. To fully utilize the potential of ChatGPT in radiology reporting, careful integration planning and rigorous validation of their outputs are crucial, especially for tasks requiring abstract reasoning or nuanced medical context. Radiologists' expertise in medical imaging and data analysis positions them exceptionally well to lead the responsible integration and utilization of ChatGPT within the field of radiology. This article offers a topical overview of the potential strengths and limitations of ChatGPT in radiological reporting.

Cardiovascular magnetic resonance (CMR) is gaining ground in guiding electrophysiology (EP)-based ablation procedures of typical atrial flutter and atrial fibrillation, allowing for the avoidance of radiation exposure for patients and operators and reducing the risk of occupational illnesses. CMR allows comprehensive assessment of cardiac anatomy and provides tissue characterization by identifying pathological substrates, such as myocardial scars and edema, identified with the implementation of late gadolinium enhancement and T2-weighted short-tau inversion recovery sequences. Intraprocedural imaging is useful for real-time catheter tracking during the ablation procedure while simultaneously providing visualization of cardiac anatomy. Additionally, CMR facilitates the evaluation of the ablation procedure accuracy by acquiring edema-sensitive sequences, thereby aiding in preventing early complications. This report serves as a primer for radiologists and illustrates the value of CMR in planning and performing the ablation procedure, as well as its role in post-procedural imaging.