Radiomics-based analysis of dynamic contrast-enhanced magnetic resonance image: A prediction nomogram for lymphovascular invasion in breast cancer.

OBJECTIVE: To develop and validate a nomogram for quantitively predicting lymphovascular invasion (LVI) of breast cancer (BC) based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) radiomics and morphological features. METHODS: We retrospectively divided 238 patients with BC into training and validation cohorts. Radiomic features from DCE-MRI were subdivided into A1 and A2, representing the first and second post-contrast images respectively. We utilized the minimal redundancy maximal relevance filter to extract radiomic features, then we employed the least absolute shrinkage and selection operator regression to screen these features and calculate individualized radiomics score (Rad score). Through the application of multivariate logistic regression, we built a prediction nomogram that integrated DCE-MRI radiomics and MR morphological features (MR-MF). The diagnostic capabilities were evaluated by comparing C-indices and calibration curves. RESULTS: The diagnostic efficiency of the A1/A2 radiomics model surpassed that of the A1 and A2 alone. Furthermore, we incorporated the MR-MF (diffusion-weighted imaging rim sign, peritumoral edema) and optimized Radiomics into a hybrid nomogram. The C-indices for the training and validation cohorts were 0.868 (95% CI: 0.839-0.898) and 0.847 (95% CI: 0.787-0.907), respectively, indicating a good level of discrimination. Moreover, the calibration plots demonstrated excellent agreement in the training and validation cohorts, confirming the effectiveness of the calibration. CONCLUSION: This nomogram combined MR-MF and A1/A2 Radiomics has the potential to preoperatively predict LVI in patients with BC.

Predicting disease recurrence in breast cancer patients using machine learning models with clinical and radiomic characteristics: a retrospective study.

BACKGROUND: The goal is to use three different machine learning models to predict the recurrence of breast cancer across a very heterogeneous sample of patients with varying disease kinds and stages. METHODS: A heterogeneous group of patients with varying cancer kinds and stages, including both triple-negative breast cancer (TNBC) and non-triple-negative breast cancer (non-TNBC), was examined. Three distinct models were created using the following five machine learning techniques: Adaptive Boosting (AdaBoost), Random Under-sampling Boosting (RUSBoost), Extreme Gradient Boosting (XGBoost), support vector machines (SVM), and Logistic Regression. The clinical model used both clinical and pathology data in conjunction with the machine learning algorithms. The machine learning algorithms were combined with dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) imaging characteristics in the radiomic model, and the merged model combined the two types of data. Each technique was evaluated using several criteria, including the receiver operating characteristic (ROC) curve, precision, recall, and F1 score. RESULTS: The results suggest that the integration of clinical and radiomic data improves the predictive accuracy in identifying instances of breast cancer recurrence. The XGBoost algorithm is widely recognized as the most effective algorithm in terms of performance. CONCLUSION: The findings presented in this study offer significant contributions to the field of breast cancer research, particularly in relation to the prediction of cancer recurrence. These insights hold great potential for informing future investigations and clinical interventions that seek to enhance the accuracy and effectiveness of recurrence prediction in breast cancer patients.

Comparison of lung lesion assessment using free-breathing dynamic contrast-enhanced 1.5-T MRI with a golden-angle radial stack-of-stars VIBE sequence and CT.

BACKGROUND: Few studies have investigated the feasibility of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) using a free-breathing golden-angle radial stack-of-stars volume-interpolated breath-hold examination (FB radial VIBE) sequence in the lung. PURPOSE: To investigate whether DCE-MRI using the FB radial VIBE sequence can assess morphological and kinetic parameters in patients with pulmonary lesions, with computed tomography (CT) as the reference. MATERIAL AND METHODS: In total, 43 patients (30 men; mean age = 64 years) with one lesion each were prospectively enrolled. Morphological and kinetic features on MRI were calculated. The diagnostic performance of morphological MR features was evaluated using a receiver operating characteristic (ROC) curve. Kinetic features were compared among subgroups based on histopathological subtype, lesion size, and lymph node metastasis. RESULTS: The maximum diameter was not significantly different between CT and MRI (3.66 ± 1.62 cm vs. 3.64 ± 1.72 cm; P = 0.663). Spiculation, lobulation, cavitation or bubble-like areas of low attenuation, and lymph node enlargement had an area under the ROC curve (AUC) >0.9, while pleural indentation yielded an AUC of 0.788. The lung cancer group had significantly lower Ktrans, Ve, and initial AUC values than the other cause inflammation group (0.203, 0.158, and 0.589 vs. 0.597, 0.385, and 1.626; P < 0.05) but significantly higher values than the tuberculosis group (P < 0.05). CONCLUSION: Morphology features derived from FB radial VIBE have high correlations with CT, and kinetic analyses show significant differences between benign and malignant lesions. DCE-MRI with FB radial VIBE could serve as a complementary quantification tool to CT for radiation-free assessments of lung lesions.

Detail-preserving image warping by enforcing smooth image sampling.

Multi-phase dynamic contrast-enhanced magnetic resonance imaging image registration makes a substantial contribution to medical image analysis. However, existing methods (e.g., VoxelMorph, CycleMorph) often encounter the problem of image information misalignment in deformable registration tasks, posing challenges to the practical application. To address this issue, we propose a novel smooth image sampling method to align full organic information to realize detail-preserving image warping. In this paper, we clarify that the phenomenon about image information mismatch is attributed to imbalanced sampling. Then, a sampling frequency map constructed by sampling frequency estimators is utilized to instruct smooth sampling by reducing the spatial gradient and discrepancy between all-ones matrix and sampling frequency map. In addition, our estimator determines the sampling frequency of a grid voxel in the moving image by aggregating the sum of interpolation weights from warped non-grid sampling points in its vicinity and vectorially constructs sampling frequency map through projection and scatteration. We evaluate the effectiveness of our approach through experiments on two in-house datasets. The results showcase that our method preserves nearly complete details with ideal registration accuracy compared with several state-of-the-art registration methods. Additionally, our method exhibits a statistically significant difference in the regularity of the registration field compared to other methods, at a significance level of p < 0.05. Our code will be released at https://github.com/QingRui-Sha/SFM.

Reproducibility and reliability of pancreatic pharmacokinetic parameters derived from dynamic contrast-enhanced magnetic resonance imaging.

BACKGROUND: Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) with an extended Tofts linear (ETL) model for tissue and tumor evaluation has been established, but its effectiveness in evaluating the pancreas remains uncertain. PURPOSE: To understand the pharmacokinetics of normal pancreas and serve as a reference for future studies of pancreatic diseases. MATERIAL AND METHODS: Pancreatic pharmacokinetic parameters of 54 volunteers were calculated using DCE-MRI with the ETL model. First, intra- and inter-observer reliability was assessed through the use of the intra-class correlation coefficient (ICC) and coefficient of variation (CoV). Second, a subgroup analysis of the pancreatic DCE-MRI pharmacokinetic parameters was carried out by dividing the 54 individuals into three groups based on the pancreatic region, three groups based on age, and two groups based on sex. RESULTS: There was excellent agreement and low variability of intra- and inter-observer to pancreatic DCE-MRI pharmacokinetic parameters. The intra- and inter-observer ICCs of Ktrans, kep, ve, and vp were 0.971, 0.952, 0.959, 0.944 and 0.947, 0.911, 0.978, 0.917, respectively. The intra- and inter-observer CoVs of Ktrans, kep, ve, vp were 9.98%, 5.99%, 6.47%, 4.76% and 10.15%, 5.22%, 6.28%, 5.40%, respectively. Only the pancreatic ve of the older group was higher than that of the young and middle-aged groups (P = 0.042, 0.001), and the vp of the pancreatic head was higher than that of the pancreatic body and tail (P = 0.014, 0.043). CONCLUSION: The application of DCE-MRI with an ETL model provides a reliable, robust, and reproducible means of non-invasively quantifying pancreatic pharmacokinetic parameters.

Fibrosarcomatous Transformation in Dermatofibrosarcoma Protuberans of the Male Breast and its Association with Magnetic Resonance Imaging and Immunohistopathologic Features.

BACKGROUND: Dermatofibrosarcoma Protuberans (DFSP) is a rare soft tissue sarcoma, accounting for approximately 1% of all tumors; however, DFSP of the breast is extremely rare. Moreover, DFSP generally has a low malignant potential and is characterized by a high rate of local recurrence along with a small but definite risk of metastasis. The risk of metastasis is higher in fibrosarcomatous transformation in DFSP than in ordinary DFSP. CASE REPORT: We have, herein, reported a case of a 61-year-old male patient with fibrosarcomatous transformation in DFSP. Preoperative Dynamic Contrastenhanced Magnetic Resonance Imaging (DCE-MRI) of the breast revealed an oval-shaped mass with heterogeneous internal enhancement, a large vessel embedded within, and a washout curve pattern on kinetic curve analysis. The mass exhibited a hyperintense signal on Diffusion-weighted Imaging (DWI), with a low apparent diffusion coefficient value. Histologically, the bland spindle tumor cells were arranged in a storiform pattern. Areas with the highest histological grade demonstrated increased cellularity, cytological atypia, and mitotic activity. Immunohistochemically, Ki-67 and p53 were highly expressed. CONCLUSION: Recognizing the risk and accurately diagnosing fibrosarcomatous transformation in male breast DFSP are critical for improving prognosis and establishing appropriate treatment and follow-up plans. This emphasizes the significance of combining immunohistopathological features with DCE-MRI and DWI to assist clinicians in the early and accurate diagnosis of sarcomas arising from male breast DFSP.

Fully automatic quantification for hand synovitis in rheumatoid arthritis using pixel-classification-based segmentation network in DCE-MRI.

PURPOSE: A classification-based segmentation method is proposed to quantify synovium in rheumatoid arthritis (RA) patients using a deep learning (DL) method based on time-intensity curve (TIC) analysis in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). MATERIALS AND METHODS: This retrospective study analyzed a hand MR dataset of 28 RA patients (six males, mean age 53.7 years). A researcher, under expert guidance, used in-house software to delineate regions of interest (ROIs) for hand muscles, bones, and synovitis, generating a dataset with 27,255 pixels with corresponding TICs (muscle: 11,413, bone: 8502, synovitis: 7340). One experienced musculoskeletal radiologist performed ground truth segmentation of enhanced pannus in the joint bounding box on the 10th DCE phase, or around 5 min after contrast injection. Data preprocessing included median filtering for noise reduction, phase-only correlation algorithm for motion correction, and contrast-limited adaptive histogram equalization for improved image contrast and noise suppression. TIC intensity values were normalized using zero-mean normalization. A DL model with dilated causal convolution and SELU activation function was developed for enhanced pannus segmentation, tested using leave-one-out cross-validation. RESULTS: 407 joint bounding boxes were manually segmented, with 129 synovitis masks. On the pixel-based level, the DL model achieved sensitivity of 85%, specificity of 98%, accuracy of 99% and precision of 84% for enhanced pannus segmentation, with a mean Dice score of 0.73. The false-positive rate for predicting cases without synovitis was 0.8%. DL-measured enhanced pannus volume strongly correlated with ground truth at both pixel-based (r = 0.87, p < 0.001) and patient-based levels (r = 0.84, p < 0.001). Bland-Altman analysis showed the mean difference for hand joints at the pixel-based and patient-based levels were -9.46 mm3 and -50.87 mm3, respectively. CONCLUSION: Our DL-based DCE-MRI TIC shape analysis has the potential for automatic segmentation and quantification of enhanced synovium in the hands of RA patients.

Development and Validation of a Nomogram Based on DCE-MRI Radiomics for Predicting Hypoxia-Inducible Factor 1α Expression in Locally Advanced Rectal Cancer.

RATIONALE AND OBJECTIVES: The expression levels of hypoxia-inducible factor 1 alpha (HIF-1α) have been identified as a pivotal marker, correlating with treatment response in patients with locally advanced rectal cancer (LARC). This study aimed to develop and validate a nomogram based on dynamic contrast-enhanced MRI (DCE-MRI) radiomics and clinical features for predicting the expression of HIF-1α in patients with LARC. MATERIALS AND METHODS: A total of 102 patients diagnosed with locally advanced rectal cancer were divided into training (n = 71) and validation (n = 31) cohorts. The expression statuses of HIF-1α were histopathologically classified, categorizing patients into high and low expression groups. The intraclass correlation coefficient (ICC), minimum redundancy maximum relevance (mRMR), and the least absolute shrinkage and selection operator (LASSO) were employed for feature selection to construct a radiomics signature and calculate the radiomics score (Rad-score). Univariate and multivariate analyses of clinical features and Rad-score were applied, and the clinical model and the nomogram were constructed. The predictive performance of the nomogram incorporating clinical features and Rad-score was assessed using Receiver Operating Characteristics (ROC) curves, decision curve analysis (DCA), and calibration curves. RESULTS: Seven radiomics features from DCE-MRI were used to build the radiomics signature. The nomogram incorporating CEA, Ki-67 and Rad-score had the highest AUC values in the training cohort and in the validation cohort (AUC: 0.918 and 0.920). Decision curve analysis showed that the nomogram outperformed the clinical model and radiomics signature in terms of clinical utility. In addition, the calibration curve for the nomogram demonstrated good agreement between prediction and actual observation. CONCLUSION: The nomogram based on DCE-MRI radiomics and clinical features showed favorable predictive efficacy and might be useful for preoperatively discriminating the expression of HIF-1α.

Machine learning models based on quantitative dynamic contrast-enhanced MRI parameters assess the expression levels of CD3+, CD4+, and CD8+ tumor-infiltrating lymphocytes in advanced gastric carcinoma.

Objective: To explore the effectiveness of machine learning classifiers based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in predicting the expression levels of CD3+, CD4+, and CD8+ tumor-infiltrating lymphocytes (TILs) in patients with advanced gastric cancer (AGC). Methods: This study investigated 103 patients with confirmed AGC through DCE-MRI and immunohistochemical staining. Immunohistochemical staining was used to evaluate CD3+, CD4+, and CD8+ T-cell expression. Utilizing Omni Kinetics software, radiomics features (Ktrans, Kep, and Ve) were extracted and underwent selection via variance threshold, SelectKBest, and LASSO methods. Logistic regression (LR), support vector machine (SVM), random forest (RF), and eXtreme Gradient Boosting (XGBoost) are the four classifiers used to build four machine learning (ML) models, and their performance was evaluated using 10-fold cross-validation. The model's performance was evaluated and compared using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. Results: In terms of CD3+, CD4+, and CD8+ T lymphocyte prediction models, the random forest model outperformed the other classifier models in terms of CD4+ and CD8+ T cell prediction, with AUCs of 0.913 and 0.970 on the training set and 0.904 and 0.908 on the validation set, respectively. In terms of CD3+ T cell prediction, the logistic regression model fared the best, with AUCs on the training and validation sets of 0.872 and 0.817, respectively. Conclusion: Machine learning classifiers based on DCE-MRI have the potential to accurately predict CD3+, CD4+, and CD8+ tumor-infiltrating lymphocyte expression levels in patients with AGC.

Correlation of MRI quantitative perfusion parameters with EGFR, VEGF and EGFR gene mutations in non-small cell cancer.

To explore the relationship between quantitative perfusion histogram parameters of dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) with the expression of tumor tissue epidermal growth factor receptor (EGFR), vascular endothelial growth factor (VEGF) and EGFR gene mutations in non-small cell lung cancer (NSCLC). A total of 44 consecutive patients with known NSCLC were recruited from March 2018 to August 2021. Histogram parameters (mean, uniformity, skewness, energy, kurtosis, entropy, percentile) of each (Ktrans, Kep, Ve, Vp, Fp) were obtained by Omni Kinetics software. Immunohistochemistry staining was used in the detection of the expression of VEGF and EGFR protein, and the mutation of EGFR gene was detected by PCR. Corresponding statistical test was performed to compare the parameters and protein expression between squamous cell carcinoma (SCC) and adenocarcinoma (AC), as well as EGFR mutations and wild-type. Correlation analysis was used to evaluate the correlation between parameters with the expression of VEGF and EGFR protein. Fp (skewness, kurtosis, energy) were statistically significant between SCC and AC, and the area under the ROC curve were 0.733, 0.700 and 0.675, respectively. The expression of VEGF in AC was higher than in SCC. Fp (skewness, kurtosis, energy) were negatively correlated with VEGF (r = - 0.527, - 0.428, - 0.342); Ktrans (Q50) was positively correlated with VEGF (r = 0.32); Kep (energy), Ktrans (skewness, kurtosis) were positively correlated with EGFR (r = 0.622, r = 0.375, 0.358), some histogram parameters of Kep, Ktrans (uniformity, entropy) and Ve (kurtosis) were negatively correlated with EGFR (r = - 0.312 to - 0.644). Some perfusion histogram parameters were statistically significant between EGFR mutations and wild-type, they were higher in wild-type than mutated (P < 0.05). Quantitative perfusion histogram parameters of DCE-MRI have a certain value in the differential diagnosis of NSCLC, which have the potential to non-invasively evaluate the expression of cell signaling pathway-related protein.

Deconvolution-Based Pharmacokinetic Analysis to Improve the Prediction of Pathological Information of Breast Cancer.

Pharmacokinetic (PK) parameters, revealing changes in the tumor microenvironment, are related to the pathological information of breast cancer. Tracer kinetic models (e.g., Tofts-Kety model) with a nonlinear least square solver are commonly used to estimate PK parameters. However, the method is sensitive to noise in images. To relieve the effects of noise, a deconvolution (DEC) method, which was validated on synthetic concentration-time series, was proposed to accurately calculate PK parameters from breast dynamic contrast-enhanced magnetic resonance imaging. A time-to-peak-based tumor partitioning method was used to divide the whole tumor into three tumor subregions with different kinetic patterns. Radiomic features were calculated from the tumor subregion and whole tumor-based PK parameter maps. The optimal features determined by the fivefold cross-validation method were used to build random forest classifiers to predict molecular subtypes, Ki-67, and tumor grade. The diagnostic performance evaluated by the area under the receiver operating characteristic curve (AUC) was compared between the subregion and whole tumor-based PK parameters. The results showed that the DEC method obtained more accurate PK parameters than the Tofts method. Moreover, the results showed that the subregion-based Ktrans (best AUCs = 0.8319, 0.7032, 0.7132, 0.7490, 0.8074, and 0.6950) achieved a better diagnostic performance than the whole tumor-based Ktrans (AUCs = 0.8222, 0.6970, 0.6511, 0.7109, 0.7620, and 0.5894) for molecular subtypes, Ki-67, and tumor grade. These findings indicate that DEC-based Ktrans in the subregion has the potential to accurately predict molecular subtypes, Ki-67, and tumor grade.

Microvessel barrier dysfunction in sinonasal inverted papilloma-associated squamous cell carcinoma and its manifestation in dynamic contrast-enhanced MRI.

BACKGROUND: To date, an effective means to preoperatively predict the malignant transformation of sinonasal inverted papilloma (SIP) remains lacking due to similarities in clinical appearance. This study aimed to retrospectively evaluate dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) parameters and microvessel structure in tumors with histologically confirmed SIP and inverted papilloma-associated squamous cell carcinoma (IP-SCC), as well as correlate DCE-MRI findings with angiogenesis biomarkers. METHODS: Absolute quantitative DCE-MRI parameters (Ktrans, Kep, Ve) based on the Tofts model and model-free semi-quantitative indices (Tpeak, WR, MaxSlope) of SIP (n = 22) and IP-SCC (n = 20) were investigated. Regions of interest (ROIs) were oriented according to the tumor subsites in the surgical records. Micro-vessel density (MVD) counts and tight junction protein (claudin-5) expression were evaluated in tumor specimens obtained during surgery. Differences in the above data were compared between the two groups. Correlations between DCE-MRI parameters and angiogenic biomarkers were analyzed. RESULTS: Compared with SIP specimens, IP-SCC specimens were characterized by a significantly higher MVD and a leakier microvessel barrier. The values of Tpeak and Ve were significantly higher for SIP than those for IP-SCC, whereas WR, MaxSlope, and Kep were significantly lower, indicating early enhancement and a faster dispersion model in IP-SCC. MVD was positively correlated with WR and Kep and negatively correlated with Tpeak. Tpeak was slightly positively correlated to claudin-5 expression. CONCLUSION: DCE-MRI can serve as a noninvasive biomarker of angiogenesis in the malignant transformation from SIP to IP-SCC. DCE-MRI may assist in the differentiation of malignancies and treatment selection.

MR Imaging Findings of Uterine Adenomatoid Tumors.

PURPOSE: Adenomatoid tumor is a rare benign genital tract neoplasm of mesothelial origin. Uterine adenomatoid tumors occur in the outer myometrium and may mimic leiomyomas. Because hormonal treatment is not applicable to adenomatoid tumors and laparoscopic enucleation is not easy as myomectomy, it is important to differentiate adenomatoid tumors from leiomyomas for the adequate treatment. The purpose of this study is to evaluate the MRI findings of adenomatoid tumor for the differentiation from leiomyoma. METHODS: MRI findings of surgically proven 10 uterine adenomatoid tumors in 9 women were retrospectively evaluated with correlation to histopathological findings. RESULTS: All 10 tumors appeared as solid myometrial masses and showed heterogeneous signal intensity with admixture of partially ill-defined slight high-intensity areas containing abundant tubular tumor cells and well-defined myoma-like low-intensity areas reflecting smooth muscle hypertrophy on T2WI including 4 lesions with peripheral ring-like high intensity. High-intensity areas on T2WI tended to show high intensity on diffusion-weighted imaging (DWI) with relatively high apparent diffusion coefficient (ADC), suggesting T2 shine-through effect due to abundant tubules. Intra-tumoral hemorrhage revealed on MRI was rare. Early intense contrast-enhanced areas on dynamic contrast-enhanced study were observed dominantly within the high-intensity areas but rarely within the low-intensity areas on T2WI. CONCLUSION: The outer myometrial mass with the admixture of well-defined low- and ill-defined high-intensity areas on T2WI may be suggestive of adenomatoid tumor. Peripheral ring-like high intensity on T2WI and DWI may also be suggestive. Dynamic contrast-enhanced MR study may be helpful for the differentiation from leiomyoma.

Predictive value of triple negative breast cancer based on DCE-MRI multi-phase full-volume ROI clinical radiomics model.

BACKGROUND: Since no studies compared the value of radiomics features of distinct phases of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for predicting triple-negative breast cancer (TNBC). PURPOSE: To identify the optimal phase of DCE-MRI for diagnosing TNBC and, in combination with clinical factors, to develop a clinical-radiomics model to well predict TNBC. MATERIAL AND METHODS: This retrospective study included 158 patients with pathology-confirmed breast cancer, including 38 cases of TNBC. The patients were randomly divided into the training and validation set (7:3). Eight radiomics models were built based on eight DCE-MR phases, and their performances were evaluated using receiver operating characteristic curve (ROC) and DeLong's test. The Radscore derived from the best radiomics model was integrated with independent clinical risk factors to construct a clinical-radiomics predictive model, and evaluate its performance using ROC analysis, calibration, and decision curve analyses. RESULTS: WHO classification, margin, and T2-weighted (T2W) imaging signals were significantly correlated with TNBC and independent risk factors for TNBC (P<0.05). The clinical model yielded areas under the curve (AUCs) of 0.867 and 0.843 in the training and validation sets, respectively. The radiomics model based on DCEphase7 achieved the highest efficacy, with an AUC of 0.818 and 0.777. The AUC of the clinical-radiomics model was 0.936 and 0.886 in the training and validation sets, respectively. The decision curve showed the clinical utility of the clinical-radiomics model. CONCLUSION: The radiomics features of DCE-MRI had the potential to predict TNBC and could improve the performance of clinical risk factors for preoperative personalized prediction of TNBC.

Image quality optimization: dynamic contrast-enhanced MRI of the abdomen at 3T using a continuously acquired radial golden-angle compressed sensing acquisition.

INTRODUCTION: The image quality of continuously acquired free-breathing Dynamic Contrast-Enhanced (DCE) golden-angle radial Magnetic Resonance Imaging (MRI) of abdomen suffers from motion artifacts and motion-related blurring. We propose a scheme by minimizing patients' motion status from breathing as well as optimizing the acquiring parameters to improve image quality and diagnostic performance of DCE-MRI with Golden-Angle Radial Sparse Parallel (GRASP) sequence of abdomen. METHODS: The optimization scheme follows two principles: (1) reduce the impact on images from unpredictable and irregulate motions during examination and (2) adjust the sequence parameters to increase the number of radial views in each partition. For the assessment of image quality, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), the severity of radial artifact, the degree of image sharpness, and a visual scoring of image quality with a 5-point scale were assessed. RESULTS: A total of 64 patients were included in this study before (16 men, 14 women, age: 54.9 ± 17.0) and after (18 men, 16 women, age: 58.6 ± 12.6) the optimization scheme was performed. The results showed that the SNR values of right and left lobe of liver in both plain phase and arterial phase were significantly increased (All P < 0.001) after the GRASP sequence been optimized. Significant improvements in CNR values were observed in the arterial phase (All P < 0.05). The significant differences in scores at each phase for visual scoring of image quality, noise of the right and left lobe of liver, radial artifact, and sharpness indicating that the image quality was significantly improved after the optimization (All P < 0.001). CONCLUSION: Our study demonstrated that the optimized scheme significantly improved the image quality of liver DCE-MRI with GRASP sequence both in plain and arterial phases. The optimized scheme of GRASP sequence could be a superior alternative to conventional approach for the assessment of liver.

Characteristic findings of primary perirenal angiosarcoma on chemical shift subtraction magnetic resonance imaging (CSS-MRI): A case report.

A 70-year-old man with supraglottic carcinoma underwent computed tomography (CT) for staging purposes. A tumor measuring approximately 7 × 10 cm was found incidentally in the left perirenal space. The tumor showed homogeneous high signal intensity on chemical shift subtraction magnetic resonance imaging (CSS-MRI) suggesting the presence of minimal amounts of fat. Five months later, the tumor had grown to approximately 10 × 12 cm with indistinct margins. CSS-MRI showed high signal intensity in the tumor periphery only. The tumor was resected and the pathological diagnosis was angiosarcoma. Angiosarcomas are malignant endothelial vascular neoplasms that are highly invasive to their surroundings. Here we report a case of primary perirenal angiosarcoma that was difficult to differentiate from a dedifferentiated liposarcoma. On CSS-MRI, high signal intensity within a tumor may be a characteristic feature of primary perirenal angiosarcoma.

Quantitative Analysis of Prostate MRI: Correlation between Contrast-Enhanced Magnetic Resonance Fingerprinting and Dynamic Contrast-Enhanced MRI Parameters.

This research aimed to assess the relationship between contrast-enhanced (CE) magnetic resonance fingerprinting (MRF) values and dynamic contrast-enhanced (DCE) MRI parameters including (Ktrans, Kep, Ve, and iAUC). To evaluate the correlation between the MRF-derived values (T1 and T2 values, CE T1 and T2 values, T1 and T2 change) and DCE-MRI parameters and the differences in the parameters between prostate cancer and noncancer lesions in 68 patients, two radiologists independently drew regions-of-interest (ROIs) at the focal prostate lesions. Prostate cancer was identified in 75% (51/68) of patients. The CE T2 value was significantly lower in prostate cancer than in noncancer lesions in the peripheral zone and transition zone. Ktrans, Kep, and iAUC were significantly higher in prostate cancer than noncancer lesions in the peripheral zone (p < 0.05), but not in the transition zone. The CE T1 value was significantly correlated with Ktrans, Ve, and iAUC in prostate cancer, and the CE T2 value was correlated to Ve in noncancer. Some CE MRF values are different between prostate cancer and noncancer tissues and correlate with DCE-MRI parameters. Prostate cancer and noncancer tissues may have different characteristics regarding contrast enhancement.

The value of multiparametric MRI combined with clinical prognostic parameters in predicting the 5-year survival of stage IIIC1 cervical squamous cell carcinoma.

OBJECTIVES: To explore the value of multiparametric magnetic resonance imaging（MRI）in predicting the 5-year progression-free survival (PFS) and overall survival (OS) of cervical squamous cell carcinoma (CSCC) in 2018 FIGO stage IIIC1. METHODS: This retrospective study collected156 patients with CSCC from Dec. 2014 to Jul. 2018. Sixty-one patients underwent radical hysterectomy (RH), and 95 patients underwent concurrent chemoradiotherapy (CCRT). Clinical and MR parameters of primary tumours were analysed. A 1:1 ratio propensity score matching (PSM) was performed for the RH group and CCRT group according to T stage. The Cox proportional hazard model was used to evaluate the associations between imaging or clinical variables and PFS and OS. RESULTS: The 5-year PFS and OS rates were 72.6% and 78.3%, respectively. The analysis results show that the treatment method, ADCmin < 0.604 × 10-3 mm2/s, and Ktrans < 0.699 min-1 correlated with worse PFS, while SCC-Ag > 6.7 ng/L, ADCmin < 0.604 × 10-3 mm2/s, and Ktrans < 0.699 min-1 correlated with worse OS. After PSM, we confirmed that the treatment methods did not affect the long-term survival of patients with stage IIIC1 disease, and a low Ktrans value was an independent poor prognostic factor. CONCLUSION: Functional MRI parameters and SCC-Ag have potential predictive value for the 5-year survival of 2018 FIGOIIIC1 CSCC. There were no significant differences in survival between CCRT and RH + adjuvant therapy for IIIC1 stage CSCC if the T stage was earlier.

Efficient estimation of pharmacokinetic parameters from breast dynamic contrast-enhanced MRI based on a convolutional neural network for predicting molecular subtypes.

Objective. Tracer kinetic models allow for estimating pharmacokinetic (PK) parameters, which are related to pathological characteristics, from breast dynamic contrast-enhanced magnetic resonance imaging. However, existing tracer kinetic models subject to inaccuracy are time-consuming for PK parameters estimation. This study aimed to accurately and efficiently estimate PK parameters for predicting molecular subtypes based on convolutional neural network (CNN).Approach. A CNN integrating global and local features (GL-CNN) was trained using synthetic data where known PK parameters map was used as the ground truth, and subsequently used to directly estimate PK parameters (volume transfer constantKtransand flux rate constantKep) map. The accuracy assessed by the peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and concordance correlation coefficient (CCC) was compared between the GL-CNN and Tofts-based PK parameters in synthetic data. Radiomic features were calculated from the PK parameters map in 208 breast tumors. A random forest classifier was constructed to predict molecular subtypes using a discovery cohort (n= 144). The diagnostic performance evaluated on a validation cohort (n= 64) using the area under the receiver operating characteristic curve (AUC) was compared between the GL-CNN and Tofts-based PK parameters.Main results. The average PSNR (48.8884), SSIM (0.9995), and CCC (0.9995) between the GL-CNN-basedKtransmap and ground truth were significantly higher than those between the Tofts-basedKtransmap and ground truth. The GL-CNN-basedKtransobtained significantly better diagnostic performance (AUCs = 0.7658 and 0.8528) than the Tofts-basedKtransfor luminal B and HER2 tumors. The GL-CNN method accelerated the computation by speed approximately 79 times compared to the Tofts method for the whole breast of all patients.Significance. Our results indicate that the GL-CNN method can be used to accurately and efficiently estimate PK parameters for predicting molecular subtypes.

Comprehensive integrated analysis of MR and DCE-MR radiomics models for prognostic prediction in nasopharyngeal carcinoma.

Although prognostic prediction of nasopharyngeal carcinoma (NPC) remains a pivotal research area, the role of dynamic contrast-enhanced magnetic resonance (DCE-MR) has been less explored. This study aimed to investigate the role of DCR-MR in predicting progression-free survival (PFS) in patients with NPC using magnetic resonance (MR)- and DCE-MR-based radiomic models. A total of 434 patients with two MR scanning sequences were included. The MR- and DCE-MR-based radiomics models were developed based on 289 patients with only MR scanning sequences and 145 patients with four additional pharmacokinetic parameters (volume fraction of extravascular extracellular space (ve), volume fraction of plasma space (vp), volume transfer constant (Ktrans), and reverse reflux rate constant (kep) of DCE-MR. A combined model integrating MR and DCE-MR was constructed. Utilizing methods such as correlation analysis, least absolute shrinkage and selection operator regression, and multivariate Cox proportional hazards regression, we built the radiomics models. Finally, we calculated the net reclassification index and C-index to evaluate and compare the prognostic performance of the radiomics models. Kaplan-Meier survival curve analysis was performed to investigate the model's ability to stratify risk in patients with NPC. The integration of MR and DCE-MR radiomic features significantly enhanced prognostic prediction performance compared to MR- and DCE-MR-based models, evidenced by a test set C-index of 0.808 vs 0.729 and 0.731, respectively. The combined radiomics model improved net reclassification by 22.9%-52.6% and could significantly stratify the risk levels of patients with NPC (p = 0.036). Furthermore, the MR-based radiomic feature maps achieved similar results to the DCE-MR pharmacokinetic parameters in terms of reflecting the underlying angiogenesis information in NPC. Compared to conventional MR-based radiomics models, the combined radiomics model integrating MR and DCE-MR showed promising results in delivering more accurate prognostic predictions and provided more clinical benefits in quantifying and monitoring phenotypic changes associated with NPC prognosis.