Purpose To evaluate the association of multiparametric and multiregional magnetic resonance (MR) imaging features with key molecular characteristics in patients with newly diagnosed glioblastoma. ...Materials and Methods Retrospective data evaluation was approved by the local ethics committee, and the requirement to obtain informed consent was waived. Preoperative MR imaging features were correlated with key molecular characteristics within a single-institution cohort of 152 patients with newly diagnosed glioblastoma. Preoperative MR imaging features (n = 31) included multiparametric (anatomic and diffusion-, perfusion-, and susceptibility-weighted images) and multiregional (contrast-enhancing regions and hyperintense regions at nonenhanced fluid-attenuated inversion recovery imaging) information with histogram quantification of tumor volumes, volume ratios, apparent diffusion coefficients, cerebral blood flow, cerebral blood volume, and intratumoral susceptibility signals. Molecular characteristics determined included global DNA methylation subgroups (eg, mesenchymal, RTK I "PGFRA," RTK II "classic"), MGMT promoter methylation status, and hallmark copy number variations (EGFR, PDGFRA, MDM4, and CDK4 amplification; PTEN, CDKN2A, NF1, and RB1 loss). Univariate analyses (voxel-lesion symptom mapping for tumor location, Wilcoxon test for all other MR imaging features) and machine learning models were applied to study the strength of association and discriminative value of MR imaging features for predicting underlying molecular characteristics. Results There was no tumor location predilection for any of the assessed molecular parameters (permutation-adjusted P > .05). Univariate imaging parameter associations were noted for EGFR amplification and CDKN2A loss, with both demonstrating increased Gaussian-normalized relative cerebral blood volume and Gaussian-normalized relative cerebral blood flow values (area under the receiver operating characteristics curve: 63%-69%, false discovery rate-adjusted P < .05). Subjecting all MR imaging features to machine learning-based classification enabled prediction of EGFR amplification status and the RTK II glioblastoma subgroup with a moderate, yet significantly greater, accuracy (63% for EGFR P < .01, 61% for RTK II P = .01) than prediction by chance; prediction accuracy for all other molecular parameters was not significant. Conclusion The authors found associations between established MR imaging features and molecular characteristics, although not of sufficient strength to enable generation of machine learning classification models for reliable and clinically meaningful prediction of molecular characteristics in patients with glioblastoma.
RSNA, 2016 Online supplemental material is available for this article.
Purpose To evaluate a radiomics model of Breast Imaging Reporting and Data System (BI-RADS) 4 and 5 breast lesions extracted from breast-tissue-optimized kurtosis magnetic resonance (MR) imaging for ...lesion characterization by using a sensitivity threshold similar to that of biopsy. Materials and Methods This institutional study included 222 women at two independent study sites (site 1: training set of 95 patients; mean age ± standard deviation, 58.6 years ± 6.6; 61 malignant and 34 benign lesions; site 2: independent test set of 127 patients; mean age, 58.2 years ± 6.8; 61 malignant and 66 benign lesions). All women presented with a finding suspicious for cancer at x-ray mammography (BI-RADS 4 or 5) and an indication for biopsy. Before biopsy, diffusion-weighted MR imaging (b values, 0-1500 sec/mm
) was performed by using 1.5-T imagers from different MR imaging vendors. Lesions were segmented and voxel-based kurtosis fitting adapted to account for fat signal contamination was performed. A radiomics feature model was developed by using a random forest regressor. The fixed model was tested on an independent test set. Conventional interpretations of MR imaging were also assessed for comparison. Results The radiomics feature model reduced false-positive results from 66 to 20 (specificity 70.0% 46 of 66) at the predefined sensitivity of greater than 98.0% 60 of 61 in the independent test set, with BI-RADS 4a and 4b lesions benefiting from the analysis (specificity 74.0%, 37 of 50; 60.0% nine of 15) and BI-RADS 5 lesions showing no added benefit. The model significantly improved specificity compared with the median apparent diffusion coefficient (P < .001) and apparent kurtosis coefficient (P = .02) alone. Conventional reading of dynamic contrast material-enhanced MR imaging provided sensitivity of 91.8% (56 of 61) and a specificity of 74.2% (49 of 66). Accounting for fat signal intensity during fitting significantly improved the area under the curve of the model (P = .001). Conclusion A radiomics model based on kurtosis diffusion-weighted imaging performed by using MR imaging machines from different vendors allowed for reliable differentiation between malignant and benign breast lesions in both a training and an independent test data set.
RSNA, 2018 Online supplemental material is available for this article.
Prostate cancer is the most frequently diagnosed cancer in males and the second leading cause of cancer-related death in men. Assessment of prostate cancer can be divided into detection, ...localization, and staging; accurate assessment is a prerequisite for optimal clinical management and therapy selection. Magnetic resonance (MR) imaging has been shown to be of particular help in localization and staging of prostate cancer. Traditional prostate MR imaging has been based on morphologic imaging with standard T1-weighted and T2-weighted sequences, which has limited accuracy. Recent advances include additional functional and physiologic MR imaging techniques (diffusion-weighted imaging, MR spectroscopy, and perfusion imaging), which allow extension of the obtainable information beyond anatomic assessment. Multiparametric MR imaging provides the highest accuracy in diagnosis and staging of prostate cancer. In addition, improvements in MR imaging hardware and software (3-T vs 1.5-T imaging) continue to improve spatial and temporal resolution and the signal-to-noise ratio of MR imaging examinations. Another recent advancement in the field is MR imaging guidance for targeted prostate biopsy, which is an alternative to the current standard of transrectal ultrasonography-guided systematic biopsy.
Purpose
To assess radiomics as a tool to determine how well lesions found suspicious on breast cancer screening X‐ray mammography can be categorized into malignant and benign with unenhanced magnetic ...resonance (MR) mammography with diffusion‐weighted imaging and T2‐weighted sequences.
Materials and Methods
From an asymptomatic screening cohort, 50 women with mammographically suspicious findings were examined with contrast‐enhanced breast MRI (ceMRI) at 1.5T. Out of this protocol an unenhanced, abbreviated diffusion‐weighted imaging protocol (ueMRI) including T2‐weighted, (T2w), diffusion‐weighted imaging (DWI), and DWI with background suppression (DWIBS) sequences and corresponding apparent diffusion coefficient (ADC) maps were extracted. From ueMRI‐derived radiomic features, three Lasso‐supervised machine‐learning classifiers were constructed and compared with the clinical performance of a highly experienced radiologist: 1) univariate mean ADC model, 2) unconstrained radiomic model, 3) constrained radiomic model with mandatory inclusion of mean ADC.
Results
The unconstrained and constrained radiomic classifiers consisted of 11 parameters each and achieved differentiation of malignant from benign lesions with a .632 + bootstrap receiver operating characteristics (ROC) area under the curve (AUC) of 84.2%/85.1%, compared to 77.4% for mean ADC and 95.9%/95.9% for the experienced radiologist using ceMRI/ueMRI.
Conclusion
In this pilot study we identified two ueMRI radiomics classifiers that performed well in the differentiation of malignant from benign lesions and achieved higher performance than the mean ADC parameter alone. Classification was lower than the almost perfect performance of a highly experienced breast radiologist. The potential of radiomics to provide a training‐independent diagnostic decision tool is indicated. A performance reaching the human expert would be highly desirable and based on our results is considered possible when the concept is extended in larger cohorts with further development and validation of the technique.
Level of Evidence: 1
Technical Efficacy: Stage 2
J. MAGN. RESON. IMAGING 2017;46:604–616
Objectives
To simulate clinical deployment, evaluate performance, and establish quality assurance of a deep learning algorithm (U-Net) for detection, localization, and segmentation of clinically ...significant prostate cancer (sPC), ISUP grade group ≥ 2, using bi-parametric MRI.
Methods
In 2017, 284 consecutive men in active surveillance, biopsy-naïve or pre-biopsied, received targeted and extended systematic MRI/transrectal US-fusion biopsy, after examination on a single MRI scanner (3 T). A prospective adjustment scheme was evaluated comparing the performance of the Prostate Imaging Reporting and Data System (PI-RADS) and U-Net using sensitivity, specificity, predictive values, and the Dice coefficient.
Results
In the 259 eligible men (median 64 IQR 61–72 years), PI-RADS had a sensitivity of 98% 106/108/84% 91/108 with a specificity of 17% 25/151/58% 88/151, for thresholds at ≥ 3/≥ 4 respectively. U-Net using dynamic threshold adjustment had a sensitivity of 99% 107/108/83% 90/108 (
p
> 0.99/> 0.99) with a specificity of 24% 36/151/55% 83/151 (
p
> 0.99/> 0.99) for probability thresholds d3 and d4 emulating PI-RADS ≥ 3 and ≥ 4 decisions respectively, not statistically different from PI-RADS. Co-occurrence of a radiological PI-RADS ≥ 4 examination and U-Net ≥ d3 assessment significantly improved the positive predictive value from 59 to 63% (
p
= 0.03), on a per-patient basis.
Conclusions
U-Net has similar performance to PI-RADS in simulated continued clinical use. Regular quality assurance should be implemented to ensure desired performance.
Key Points
• U-Net maintained similar diagnostic performance compared to radiological assessment of PI-RADS ≥ 4 when applied in a simulated clinical deployment.
• Application of our proposed prospective dynamic calibration method successfully adjusted U-Net performance within acceptable limits of the PI-RADS reference over time, while not being limited to PI-RADS as a reference.
• Simultaneous detection by U-Net and radiological assessment significantly improved the positive predictive value on a per-patient and per-lesion basis, while the negative predictive value remained unchanged.
Abstract Purpose To assess the interobserver agreement in 50 patients with hepatocellular carcinoma (HCC) before and 1 month after intra-arterial therapy (IAT) using two semi-automated methods and a ...manual approach for the following functional, volumetric and morphologic parameters: (1) apparent diffusion coefficient (ADC), (2) arterial phase enhancement (AE), (3) portal venous phase enhancement (VE), (4) tumor volume, and assessment according to (5) the Response Evaluation Criteria in Solid Tumors (RECIST), and (6) the European Association for the Study of the Liver (EASL). Materials and methods This HIPAA-compliant retrospective study had institutional review board approval. The requirement for patient informed consent was waived. Tumor ADC, AE, VE, volume, RECIST, and EASL in 50 index lesions was measured by three observers. Interobserver reproducibility was evaluated using intraclass correlation coefficients (ICC). P < 0.05 was considered to indicate a significant difference. Results Semi-automated volumetric measurements of functional parameters (ADC, AE, and VE) before and after IAT as well as change in tumor ADC, AE, or VE had better interobserver agreement (ICC = 0.830–0.974) compared with manual ROI-based axial measurements (ICC = 0.157–0.799). Semi-automated measurements of tumor volume and size in the axial plane before and after IAT had better interobserver agreement (ICC = 0.854–0.996) compared with manual size measurements (ICC = 0.543–0.596), and interobserver agreement for change in tumor RECIST size was also higher using semi-automated measurements (ICC = 0.655) compared with manual measurements (ICC = 0.169). EASL measurements of tumor enhancement in the axial plane before and after IAT ((ICC = 0.758–0.809), and changes in EASL after IAT (ICC = 0.653) had good interobserver agreement. Conclusion Semi-automated measurements of functional changes assessed by ADC and VE based on whole-lesion segmentation demonstrated better reproducibility than ROI-based axial measurements, or RECIST or EASL measurements.
Chordomas are rare bone tumors with few therapeutic options. Here we show, using whole-exome and genome sequencing within a precision oncology program, that advanced chordomas (n = 11) may be ...characterized by genomic patterns indicative of defective homologous recombination (HR) DNA repair and alterations affecting HR-related genes, including, for example, deletions and pathogenic germline variants of BRCA2, NBN, and CHEK2. A mutational signature associated with HR deficiency was significantly enriched in 72.7% of samples and co-occurred with genomic instability. The poly(ADP-ribose) polymerase (PARP) inhibitor olaparib, which is preferentially toxic to HR-incompetent cells, led to prolonged clinical benefit in a patient with refractory chordoma, and whole-genome analysis at progression revealed a PARP1 p.T910A mutation predicted to disrupt the autoinhibitory PARP1 helical domain. These findings uncover a therapeutic opportunity in chordoma that warrants further exploration, and provide insight into the mechanisms underlying PARP inhibitor resistance.
Purpose
To evaluate whether quantitative susceptibility (QSM) may be used as an alternative to computed tomography (CT) to detect calcification in prostate cancer patients.
Materials and Methods
...Susceptibility map calculation was performed using 3D gradient echo magnetic resonance imaging (MRI) data from 26 patients measured at 3T who previously received a planning CT of the prostate. Phase images were unwrapped using Laplacian‐based phase unwrapping, the background field was removed with the V‐SHARP method, and susceptibility maps were calculated with the iLSQR method. Two blinded readers were asked to identify peri‐ and intraprostatic calcifications.
Results
Average mean and minimum susceptibility values (referenced to iliopsoas muscle) of calcifications were −0.249 ± 0.179 ppm and –0.551 ± 0.323 ppm, and average mean and maximum intensities in CT images were 319 ± 164 HU and 679 ± 392 HU. Twenty‐one and 17 out of 22 prostatic calcifications were identified using susceptibility maps and magnitude images, respectively, as well as more than half of periprostatic phleboliths depicted by CT. Calcifications in the prostate and its periphery were quantitatively differentiable from noncalcified prostate tissue in CT (mean values for calcifications / for noncalcified tissue: 71 to 649 / –1 to 83 HU) and in QSM (mean values for calcifications / for noncalcified tissue: –0.641 to 0.063 / –0.046 to 0.181 ppm). Moreover, there was a significant correlation between susceptibility values and CT image intensities for calcifications (P < 0.004).
Conclusion
Prostatic calcifications could be well identified with QSM. Susceptibility maps can be easily obtained from clinical prostate MR protocols that include a 3D gradient echo sequence, rendering it a promising technique for detection and quantification of intraprostatic calcifications.
Level of Evidence: 1
J. Magn. Reson. Imaging 2017;45:889–898.