European Journal of Radiology最新文献

Purpose

This study aimed to determine if initial MRI findings could predict a pathological complete response (pCR) following neoadjuvant systemic therapy (NST) in HER2-positive breast cancers.

Methods

The study retrospectively included 111 patients (Center 1, training set) and 71 patients (Center 2, validation set) with HER2-positive cancer who underwent NST. Initial clinicopathological data and MRI findings were recorded. Continuous variables were analyzed using the Mann–Whitney and Student’s t-tests, while categorical variables were analyzed using the χ² or Fisher’s exact test. Univariate analysis was conducted to determine the associations between these variables and pathological complete response (pCR), defined as the absence of invasive malignant cells in the breast and lymph nodes. Interobserver reproducibility was assessed for associated non-mass enhancement (NME) parameter by analyzing 50 MR studies (intraclass correlation coefficient).

Results

pCR was achieved in 67 patients, 51 (46 %) from Center 1 and 16 (23%) from Center 2 (p = 0.003), with significant differences between Centers 1 and 2 in tumor-infiltrating lymphocyte levels and lymphovascular invasion (p < 0.001). The initial presence of suspicious associated NME was the only significant parameter predictive of pCR (p < 0.001 for Center 1 and 0.04 for Center 2). The inter-observer reproducibility for this MRI feature was good, with an intraclass correlation coefficient of 0.872 (95 % CI: 0.73–1.00).

Conclusion

The presence of suspicious associated NME in HER2-positive cancers on the initial MRI study was predictive of achieving pCR after NST. This significant preliminary finding warrants confirmation through prospective multicenter studies.

Purpose

Different imaging tools, including digital breast tomosynthesis (DBT), are frequently used for evaluating tumor response during neoadjuvant chemotherapy (NACT). This study aimed to explore whether using artificial intelligence (AI) for serial DBT acquisitions during NACT for breast cancer can predict pathological complete response (pCR) after completion of NACT.

Methods

A total of 149 women (mean age 53 years, pCR rate 22 %) with breast cancer treated with NACT at Skane University Hospital, Sweden, between 2014 and 2019, were prospectively included in this observational cohort study (ClinicalTrials.gov: NCT02306096). DBT images from both the cancer and contralateral healthy breasts acquired at three time points: pre-NACT, after two cycles of NACT, and after the completion of six cycles of NACT (pre-surgery) were analyzed. The deep learning AI system used to predict pCR consisted of a backbone 3D ResNet and an attention and prediction module. The GradCAM method was used to obtain insights into the model decision basis through a quantitative analysis of the importance maps on the validation set. Moreover, specific model choices were motivated by ablation studies.

Results

The AI model reached an AUC of 0.83 (95% CI: 0.63–1.00) (test set). The spatial correlation of importance maps for input volumes from the same patient but at different time points was high, possibly indicating that the model focuses on the same areas during decision-making.

Conclusions

We demonstrate a high discriminative performance of our algorithm for predicting pCR/non-pCR. Availability of larger datasets would permit more comprehensive training of the models and more rigorous evaluation of their prediction performance for future patients.

Purpose

To investigate the value of microstructural characteristics derived from time-dependent diffusion MRI in distinguishing high-grade serous ovarian cancer (HGSOC) from serous borderline ovarian tumor (SBOT) and the associations of immunohistochemical markers with microstructural features.

Methods

Totally 34 HGSOC and 12 SBOT cases who received preoperative pelvic MRI were retrospectively included in this study. Two radiologists delineated the tumors to obtain the regions of interest (ROIs). Time-dependent diffusion MRI signals were fitted by the IMPULSED (imaging microstructural parameters using limited spectrally edited diffusion) model, to extract microstructural parameters, including fraction of the intracellular component (f_in), cell diameter (d), cellularity and extracellular diffusivity (D_ex). Apparent diffusion coefficient (ADC) values were obtained from standard diffusion-weighted imaging (DWI). The parameters of HGSOCs and SBOTs were compared, and the diagnostic performance was evaluated. The associations of microstructural indexes with immunopathological parameters were assessed, including Ki-67, P53, Pax-8, ER and PR.

Results

In this study, f_in, cellularity, D_ex and ADC had good diagnostic performance levels in differentiating HGSOC from SBOT, with AUCs of 0.936, 0.909, 0.902 and 0.914, respectively. There were no significant differences in diagnostic performance among these parameters. Spearman analysis revealed in the HGSOC group, cellularity had a significant positive correlation with P53 expression (P = 0.028, r = 0.389) and D_ex had a significant positive correlation with Pax-8 expression (P = 0.018, r = 0.415). ICC showed excellent agreement for all parameters.

Conclusion

Time-dependent diffusion MRI had value in evaluating the microstructures of HGSOC and SBOT and could discriminate between these tumors.

Objective

To demonstrate the value of using 50 keV virtual monochromatic images with deep learning image reconstruction (DLIR) in low-dose dual-energy CT enterography (CTE).

Methods

In this prospective study, 114 participants (62 % M; 41.9 ± 16 years) underwent dual-energy CTE. The early-enteric phase was performed using standard-dose (noise index (NI): 8) and images were reconstructed at 70 keV and 50 keV with 40 % strength ASIR-V (ASIR-V40%). The late-enteric phase used low-dose (NI: 12) and images were reconstructed at 50 keV with ASIR-V40%, and DLIR at medium (DLIR-M) and high strength (DLIR-H). Image standard deviation (SD), signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), edge-rise-slope (ERS) were computed. The quantitative comb sign score was calculated for the 27 patients with Crohn’s disease. The subjective noise, image contrast, display of rectus artery were scored using a 5-point scale by two radiologists blindly.

Results

Effective dose was reduced by 50 % (P < 0.001) in the late-enteric phase to 3.26 mSv. The lower-dose 50 keV-DLIR-H images (SD:17.7 ± 0.5HU) had similar image noise (P = 0.97) as the standard-dose 70 keV-ASIR-V40% images (SD:17.7 ± 0.73HU), but with higher (P < 0.001) SNR, CNR, ERS and quantitative comb sign score (5.7 ± 0.17, 1.8 ± 0.12, 156.04 ± 5.21 and 5.05 ± 0.73, respectively). Furthermore, the lower-dose 50 keV-DLIR-H images obtained the highest score in the rectus artery visibility (4.27 ± 0.6).

Conclusions

The 50 keV images in dual-energy CTE with DLIR provides high-quality images, with a 50 % reduction in radiation dose. Images with high contrast and density resolutions significantly enhance the diagnostic confidence of Crohn’s disease and are essential for the clinical development of individualized treatment plans.

Purpose

This study aimed to automatically segment knee computed tomography (CT) images of tibial plateau fractures using a three-dimensional (3D) U-net-based method, accurately construct 3D maps of tibial plateau fractures, and examine their usefulness for Schatzker classification in clinical practice.

Methods

We retrospectively enrolled 234 cases with tibial plateau fractures from our hospital in this study. The four constituent bones of the knee were manually annotated using ITK-SNAP software. Finally, image features were extracted using deep learning. The usefulness of the results for Schatzker classification was examined by an orthopaedic and a radiology resident.

Results

On average, our model required < 40 s to process a 3D CT scan of the knee. The average Dice coefficient for all four knee bones was higher than 0.950, and highly accurate 3D maps of the tibia were produced. With the aid of the results of our model, the accuracy, sensitivity, and specificity of the Schatzker classification of both residents improved.

Conclusions

The proposed method can rapidly and accurately segment knee CT images of tibial plateau fractures and assist residents with Schatzker classification, which can help improve diagnostic efficiency and reduce the workload of junior doctors in clinical practice.

Purpose

In planning transcatheter aortic valve replacement (TAVR), retrospective cardiac spiral-CT is recommended to measure aortic annulus with subsequent CT-angiography (CTA) to evaluate access routes. Photon-counting detector (PCD)-CT enables to assess the aortic annulus in desired cardiac phases, using prospective ECG-gated high-pitch CTA. The aim of this study was to evaluate the measurement accuracy of aortic annulus using prospective ECG-gated high-pitch CTA against retrospective spiral-CT reference.

Method

Thirty patients underwent cardiac spiral-CT and prospective ECG-gated (30% R-R on aortic valve level) high-pitch CTA. Using propensity score matching, another 30 patients were identified whose CTA was performed using high-pitch mode without ECG-synchronization. Two investigators measured annular diameter, perimeter, and area on cardiac spiral-CT and high-pitch CTA.

Results

The aortic valve was imaged in systole in 90 % of prospective ECG-gated CTA cases but only 50 % of non-ECG-gated CTA cases (p = 0.002). There was a strong correlation (r ≥ 0.94) without significant differences (p ≥ 0.09) between cardiac spiral-CT and prospective ECG-gated high-pitch CTA for all annulus measurements. In contrast, significant differences were found in annular short-axis diameter and area between cardiac spiral-CT and non-ECG-gated high-pitch CTA (p ≤ 0.03). Furthermore, prospective ECG-gated high-pitch CTA showed significantly reduced radiation exposure compared with cardiac spiral-CT (CTDI 4.52 vs. 24.10 mGy; p < 0.001).

Conclusion

PCD-CT-based prospective ECG-gated high-pitch scans with targeted systolic acquisition at the level of the aortic valve can simultaneously visualize TAVR access routes and accurately measure systolic annulus size. This approach could aid in optimizing protocols to achieve lower radiation doses in the growing population of younger, low-risk TAVR patients.

Objectives

To evaluate a novel calcium-only imaging technique (VCa) with subtracted bone marrow in osteoporosis in dual-layer CT (DLCT) compared to conventional CT images (CI) and dual-energy X-ray absorptiometry (DXA).

Material and methods

Images of a multi-energy CT phantom with calcium inserts, quantitative CT calibration phantom, and of 55 patients (mean age: 64.6 ± 11.5 years) were acquired on a DLCT to evaluate bone mineral density (BMD). CI, calcium-suppressed images, and VCa were calculated. For investigating the association of VCa and CI with DXA a subsample of 30 patients (<90 days between DXA and CT) was used. Multiple regression analysis was performed to identify further factors improving the prediction of DXA BMD.

Results

The calcium concentrations of the CT phantom inserts were significantly associated with CT numbers from VCa (R² = 0.94) and from CI (R² = 0.89–0.92). VCa showed significantly higher CT numbers than CI in the phantom (p ≤ 0.001) and clinical setting (p < 0.001). CT numbers from VCa were significantly associated with CI (R² = 0.95, p < 0.001) and with DXA (R² = 0.31, p = 0.007), whereas no significant association between DXA and CI was found. Prediction of DXA BMD based on CT numbers derived from VCa yielded R² = 0.76 in multiple regression analysis. ROC for the differentiation of normal from pathologic BMD in VCa yielded an AUC of 0.7, and a cut-off value of 126HU (sensitivity: 0.90; specificity: 0.47).

Conclusion

VCa images showed better agreement with DXA and known calcium concentrations than CI, and could be used to estimate BMD. A VCa cut-off of 126 HU could be used to identify abnormal bone mineral density.

Purpose

The aim of this study was to develop and validate radiomics signatures based on MRI for preoperative prediction of Ki-67 proliferative index (PI) expression in primary central nervous system lymphoma (PCNSL).

Methods

A total of 341 patients with PCNSL were retrospectively analyzed, including 286 patients in one center as the training set and 55 patients in another two centers as the external validation set. Radiomics features were extracted and selected from preoperative contrast-enhanced T1-weighted images, fluid attenuation inversion recovery to build radiomics signatures according to the Ki-67 PI. The predictive performances of the radiomics model were evaluated using four classifiers including random forest, K-Nearest Neighbors, Neural Network and Decision Tree. A combined model was built by incorporating radiomics signature, clinical variables and MRI radiological characteristics using multivariate logistic regression analysis, and a nomogram was established to predict the expression of Ki-67 individually. The predictive performances of the models were evaluated using area under receiver operating characteristic curve (AUC) and decision curve analysis (DCA).

Results

Radiomics signatures were independent predictors of the expression level of Ki-67 (OR: 2.523, P < 0.001). RF radiomics models had the highest accuracy (0.934 in the training set and 0.811 in the external validation set) and F1 Score (0.920 in the training set and 0.836 in the external validation set). The clinic-radiologic-radiomics nomogram showed better predictive performance with AUCs of 0.877(95 % CI: 0.837–0.918) in the training set and 0.866(95 % CI: 0.774–0.957) in the external validation set. The calibration curve and DCA demonstrated goodness-of-fit and improved benefits in clinical practice of the nomogram.

Conclusions

Nomograms integrating MRI-based radiomics and clinical-radiological characteristics could effectively predict Ki-67 PI in primary PCNSL.

Introduction

The non-perfused volume divided by total fibroid load (NPV/TFL) is a predictive outcome parameter for MRI-guided high-intensity focused ultrasound (MR-HIFU) treatments of uterine fibroids, which is related to long-term symptom relief. In current clinical practice, the MR-HIFU outcome parameters are typically determined by visual inspection, so an automated computer-aided method could facilitate objective outcome quantification. The objective of this study was to develop and evaluate a deep learning-based segmentation algorithm for volume measurements of the uterus, uterine fibroids, and NPVs in MRI in order to automatically quantify the NPV/TFL.

Materials and Methods

A segmentation pipeline was developed and evaluated using expert manual segmentations of MRI scans of 115 uterine fibroid patients, screened for and/or undergoing MR-HIFU treatment. The pipeline contained three separate neural networks, one per target structure. The first step in the pipeline was uterus segmentation from contrast-enhanced (CE)-T1w scans. This segmentation was subsequently used to remove non-uterus background tissue for NPV and fibroid segmentation. In the following step, NPVs were segmented from uterus-only CE-T1w scans. Finally, fibroids were segmented from uterus-only T2w scans. The segmentations were used to calculate the volume for each structure. Reliability and agreement between manual and automatic segmentations, volumes, and NPV/TFLs were assessed.

Results

For treatment scans, the Dice similarity coefficients (DSC) between the manually and automatically obtained segmentations were 0.90 (uterus), 0.84 (NPV) and 0.74 (fibroid). Intraclass correlation coefficients (ICC) were 1.00 [0.99, 1.00] (uterus), 0.99 [0.98, 1.00] (NPV) and 0.98 [0.95, 0.99] (fibroid) between manually and automatically derived volumes. For manually and automatically derived NPV/TFLs, the mean difference was 5% [-41%, 51%] (ICC: 0.66 [0.32, 0.85]).

Conclusion

The algorithm presented in this study automatically calculates uterus volume, fibroid load, and NPVs, which could lead to more objective outcome quantification after MR-HIFU treatments of uterine fibroids in comparison to visual inspection. When robustness has been ascertained in a future study, this tool may eventually be employed in clinical practice to automatically measure the NPV/TFL after MR-HIFU procedures of uterine fibroids.

Objectives

This study aims to assess the effectiveness of super-resolution deep-learning-based reconstruction (SR-DLR), which leverages k-space data, on the image quality of lumbar spine magnetic resonance (MR) bone imaging using a 3D multi-echo in-phase sequence.

Materials and methods

In this retrospective study, 29 patients who underwent lumbar spine MRI, including an MR bone imaging sequence between January and April 2023, were analyzed. Images were reconstructed with and without SR-DLR (Matrix sizes: 960 × 960 and 320 × 320, respectively). The signal-to-noise ratio (SNR) of the vertebral body and spinal canal and the contrast and contrast-to-noise ratio (CNR) between the vertebral body and spinal canal were quantitatively evaluated. Furthermore, the slope at half-peak points of the profile curve drawn across the posterior border of the vertebral body was calculated. Two radiologists independently assessed image noise, contrast, artifacts, sharpness, and overall image quality of both image types using a 4-point scale. Interobserver agreement was evaluated using weighted kappa coefficients, and quantitative and qualitative scores were compared via the Wilcoxon signed-rank test.

Results

SNRs of the vertebral body and spinal canal were notably improved in images with SR-DLR (p < 0.001). Contrast and CNR were significantly enhanced with SR-DLR compared to those without SR-DLR (p = 0.023 and p = 0.022, respectively). The slope of the profile curve at half-peak points across the posterior border of the vertebral body and spinal canal was markedly higher with SR-DLR (p < 0.001). Qualitative scores (noise: p < 0.001, contrast: p < 0.001, artifact p = 0.042, sharpness: p < 0.001, overall image quality: p < 0.001) were superior in images with SR-DLR compared to those without. Kappa analysis indicated moderate to good agreement (noise: κ = 0.56, contrast: κ = 0.51, artifact: κ = 0.46, sharpness: κ = 0.76, overall image quality: κ = 0.44).

Conclusion

SR-DLR, which is based on k-space data, has the potential to enhance the image quality of lumbar spine MR bone imaging utilizing a 3D gradient echo in-phase sequence.

Clinical relevance statement: The application of SR-DLR can lead to improvements in lumbar spine MR bone imaging quality.