Deep Learning Reconstruction for DWIs by EPI and FASE Sequences for Head and Neck Tumors.

BACKGROUND: Diffusion-weighted images (DWI) obtained by echo-planar imaging (EPI) are frequently degraded by susceptibility artifacts. It has been suggested that DWI obtained by fast advanced spin-echo (FASE) or reconstructed with deep learning reconstruction (DLR) could be useful for image quality improvements. The purpose of this investigation using in vitro and in vivo studies was to determine the influence of sequence difference and of DLR for DWI on image quality, apparent diffusion coefficient (ADC) evaluation, and differentiation of malignant from benign head and neck tumors. METHODS: For the in vitro study, a DWI phantom was scanned by FASE and EPI sequences and reconstructed with and without DLR. Each ADC within the phantom for each DWI was then assessed and correlated for each measured ADC and standard value by Spearman's rank correlation analysis. For the in vivo study, DWIs obtained by EPI and FASE sequences were also obtained for head and neck tumor patients. Signal-to-noise ratio (SNR) and ADC were then determined based on ROI measurements, while SNR of tumors and ADC were compared between all DWI data sets by means of Tukey's Honest Significant Difference test. RESULTS: For the in vitro study, all correlations between measured ADC and standard reference were significant and excellent (0.92 ≤ ρ ≤ 0.99, p < 0.0001). For the in vivo study, the SNR of FASE with DLR was significantly higher than that of FASE without DLR (p = 0.02), while ADC values for benign and malignant tumors showed significant differences between each sequence with and without DLR (p < 0.05). CONCLUSION: In comparison with EPI sequence, FASE sequence and DLR can improve image quality and distortion of DWIs without significantly influencing ADC measurements or differentiation capability of malignant from benign head and neck tumors.

Compressed sensing with deep learning reconstruction: Improving capability of gadolinium-EOB-enhanced 3D T1WI.

PURPOSE: The purpose of this study was to determine the utility of compressed sensing (CS) with deep learning reconstruction (DLR) for improving spatial resolution, image quality and focal liver lesion detection on high-resolution contrast-enhanced T1-weighted imaging (HR-CE-T1WI) obtained by CS with DLR as compared with conventional CE-T1WI with parallel imaging (PI). METHODS: Seventy-seven participants with focal liver lesions underwent conventional CE-T1WI with PI and HR-CE-T1WI, surgical resection, transarterial chemoembolization, and radiofrequency ablation, followed by histopathological or >2-year follow-up examinations in our hospital. Signal-to-noise ratios (SNRs) of liver, spleen and kidney were calculated for each patient, after which each SNR was compared by means of paired t-test. To compare focal lesion detection capabilities of the two methods, a 5-point visual scoring system was adopted for a per lesion basis analysis. Jackknife free-response receiver operating characteristic (JAFROC) analysis was then performed, while sensitivity and false positive rates (/data set) for consensus assessment of the two methods were also compared by using McNemar's test or the signed rank test. RESULTS: Each SNR of HR-CE-T1WI was significantly higher than that of conventional CE-T1WI with PI (p < 0.05). Sensitivities for consensus assessment showed that HR-CE-MRI had significantly higher sensitivity than conventional CE-T1WI with PI (p = 0.004). Moreover, there were significantly fewer FP/cases for HR-CE-T1WI than for conventional CE-T1WI with PI (p = 0.04). CONCLUSION: CS with DLR are useful for improving spatial resolution, image quality and focal liver lesion detection capability of Gd-EOB-DTPA enhanced 3D T1WI without any need for longer breath-holding time.

Diffusion weighted imaging with reverse encoding distortion correction: Improvement of image quality and distortion for accurate ADC evaluation in in vitro and in vivo studies.

PURPOSE: The purpose of this in vivo study was to determine the effect of reverse encoding direction (RDC) on apparent diffusion coefficient (ADC) measurements and its efficacy for improving image quality and diagnostic performance for differentiating malignant from benign tumors on head and neck diffusion-weighted imaging (DWI). METHODS: Forty-eight patients with head and neck tumors underwent DWI with and without RDC and pathological examinations. Their tumors were then divided into two groups: malignant (n = 21) and benign (n = 27). To determine the utility of RDC for DWI, the difference in the deformation ratio (DR) between DWI and T2-weighted images of each tumor was determined for each tumor area. To compare ADC measurement accuracy of DWIs with and without RDC for each patient, ADC values for tumors and spinal cord were determined by using ROI measurements. To compare DR and ADC between two methods, Student's t-tests were performed. Then, ADC values were compared between malignant and benign tumors by Student's t-test on each DWI. Finally, sensitivity, specificity and accuracy were compared by means of McNemar's test. RESULTS: DR of DWI with RDC was significantly smaller than that without RDC (p < 0.0001). There were significant differences in ADC between malignant and benign lesions on each DWI (p < 0.05). However, there were no significant difference of diagnostic accuracy between the two DWIs (p > 0.05). CONCLUSION: RDC can improve image quality and distortion of DWI and may have potential for more accurate ADC evaluation and differentiation of malignant from benign head and neck tumors.

Pulmonary MRI with ultra-short TE using single- and dual-echo methods: comparison of capability for quantitative differentiation of non- or minimally invasive adenocarcinomas from other lung cancers with that of standard-dose thin-section CT.

OBJECTIVE: The purpose of this study was thus to compare capabilities for quantitative differentiation of non- and minimally invasive adenocarcinomas from other of pulmonary MRIs with ultra-short TE (UTE) obtained with single- and dual-echo techniques (UTE-MRISingle and UTE-MRIDual) and thin-section CT for stage IA lung cancer patients. METHODS: Ninety pathologically diagnosed stage IA lung cancer patients who underwent thin-section standard-dose CT, UTE-MRISingle, and UTE-MRIDual, surgical treatment and pathological examinations were included in this retrospective study. The largest dimension (Dlong), solid portion (solid Dlong), and consolidation/tumor (C/T) ratio of each nodule were assessed. Two-tailed Student's t-tests were performed to compare all indexes obtained with each method between non- and minimally invasive adenocarcinomas and other lung cancers. Receiver operating characteristic (ROC)-based positive tests were performed to determine all feasible threshold values for distinguishing non- or minimally invasive adenocarcinoma (MIA) from other lung cancers. Sensitivity, specificity, and accuracy were then compared by means of McNemar's test. RESULTS: Each index showed significant differences between the two groups (p < 0.0001). Specificities and accuracies of solid Dlong for UTE-MRIDual2nd echo and CTMediastinal were significantly higher than those of solid Dlong for UTE-MRISingle and UTE-MRIDual1st echo and all C/T ratios except CTMediastinal (p < 0.05). Moreover, the specificities and accuracies of solid Dlong and C/T ratio were significantly higher than those of Dlong for each method (p < 0.05). CONCLUSION: Pulmonary MRI with UTE is considered at least as valuable as thin-section CT for quantitative differentiation of non- and minimally invasive adenocarcinomas from other stage IA lung cancers. CLINICAL RELEVANCE STATEMENT: Pulmonary MRI with UTE's capability for quantitative differentiation of non- and minimally invasive adenocarcinomas from other lung cancers in stage IA lung cancer patients is equal or superior to that of thin-section CT. KEY POINTS: • Correlations were excellent for pathologically examined nodules with the largest dimensions (Dlong) and a solid component (solid Dlong) for all indexes (0.95 ≤ r ≤ 0.99, p < 0.0001). • Pathologically examined Dlong and solid Dlong obtained with all methods showed significant differences between non- and minimally invasive adenocarcinomas and other lung cancers (p < 0.0001). • Solid tumor components are most accurately measured by UTE-MRIDual2nd echo and CTMediastinal, whereas the ground-glass component is imaged by UTE-MRIDual1st echo and CTlung with high accuracy. UTE-MRIDual predicts tumor invasiveness with 100% sensitivity and 87.5% specificity at a C/T threshold of 0.5.

Deep Learning Reconstruction to Improve the Quality of MR Imaging: Evaluating the Best Sequence for T-category Assessment in Non-small Cell Lung Cancer Patients.

PURPOSE: Deep learning reconstruction (DLR) has been recommended as useful for improving image quality. Moreover, compressed sensing (CS) or DLR has been proposed as useful for improving temporal resolution and image quality on MR sequences in different body fields. However, there have been no reports regarding the utility of DLR for image quality and T-factor assessment improvements on T2-weighted imaging (T2WI), short inversion time (TI) inversion recovery (STIR) imaging, and unenhanced- and contrast-enhanced (CE) 3D fast spoiled gradient echo (GRE) imaging with and without CS in comparison with thin-section multidetector-row CT (MDCT) for non-small cell lung cancer (NSCLC) patients. The purpose of this study was to determine the utility of DLR for improving image quality and the appropriate sequence for T-category assessment for NSCLC patients. METHODS: As subjects for this study, 213 pathologically diagnosed NSCLC patients who underwent thin-section MDCT and MR imaging as well as T-factor diagnosis were retrospectively enrolled. SNR of each tumor was calculated and compared by paired t-test for each sequence with and without DLR. T-factor for each patient was assessed with thin-section MDCT and all MR sequences, and the accuracy for T-factor diagnosis was compared among all sequences and thin-section CT by means of McNemar's test. RESULTS: SNRs of T2WI, STIR imaging, unenhanced thin-section Quick 3D imaging, and CE-thin-section Quick 3D imaging with DLR were significantly higher than SNRs of those without DLR (P < 0.05). Diagnostic accuracy of STIR imaging and CE-thick- or thin-section Quick 3D imaging was significantly higher than that of thin-section CT, T2WI, and unenhanced thick- or thin-section Quick 3D imaging (P < 0.05). CONCLUSION: DLR is thus considered useful for image quality improvement on MR imaging. STIR imaging and CE-Quick 3D imaging with or without CS were validated as appropriate MR sequences for T-factor evaluation in NSCLC patients.

Deep learning reconstruction for brain diffusion-weighted imaging: efficacy for image quality improvement, apparent diffusion coefficient assessment, and intravoxel incoherent motion evaluation in in vitro and in vivo studies

PURPOSE: Deep learning reconstruction (DLR) to improve imaging quality has already been introduced, but no studies have evaluated the effect of DLR on diffusion-weighted imaging (DWI) or intravoxel incoherent motion (IVIM) in in vitro or in vivo studies. The purpose of this study was to determine the effect of DLR for magnetic resonance imaging (MRI) in terms of image quality improvement, apparent diffusion coefficient (ADC) assessment, and IVIM index evaluation on DWI through in vitro and in vivo studies. METHODS: For the in vitro study, a phantom recommended by the Quantitative Imaging Biomarkers Alliance was scanned and reconstructed with and without DLR, and 15 patients with brain tumors with normal-appearing gray and white matter examined using IVIM and reconstructed with and without DLR were included in the in vivo study. The ADCs of all phantoms for DWI with and without DLR, as well as the coefficient of variation percentage (CV%), and ADCs and IVIM indexes for each participant, were evaluated based on DWI with and without DLR by means of region-of-interest measurements. For the in vitro study, using the mean ADCs for all phantoms, a t-test was adopted to compare DWI with and without DLR. For the in vivo study, a Wilcoxon signed-rank test was used to compare the CV% between the two types of DWI. In addition, the Wilcoxon signed-rank test was used to compare the ADC, true diffusion coefficient (D), pseudodiffusion coefficient (D*), and percentage of water molecules in micro perfusion within 1 voxel (f) with and without DLR; the limits of agreement of each parameter were determined through a Bland-Altman analysis. RESULTS: The in vitro study identified no significant differences between the ADC values for DWI with and without DLR (P > 0.05), and the CV% was significantly different for DWI with and without DLR (P < 0.05) when b values ≥250 s/mm2 were used. The in vivo study revealed that D* and f with and without DLR were significantly different (P < 0.001). The limits of agreement of the ADC, D, and D* values for DWI with and without DLR were determined as 0.00 ± 0.51 × 10-3, 0.00 ± 0.06 × 10-3, and 1.13 ± 4.04 × 10-3 mm2/s, respectively. The limits of agreement of the f values for DWI with and without DLR were determined as -0.01 ± 0.07. CONCLUSION: Deep learning reconstruction for MRI has the potential to significantly improve DWI quality at higher b values. It has some effect on D* and f values in the IVIM index evaluation, but ADC and D values are less affected by DLR.

Three-Dimensional Gradient-Echo-Based Amide Proton Transfer-Weighted Imaging of Brain Tumors: Comparison With Two-Dimensional Spin-Echo-Based Amide Proton Transfer-Weighted Imaging.

OBJECTIVE: Although amide proton transfer-weighted (APTw) imaging is reported by 2-dimensional (2D) spin-echo-based sequencing, 3-dimensional (3D) APTw imaging can be obtained by gradient-echo-based sequencing. The purpose of this study was to compare the efficacy of APTw imaging between 2D and 3D imaging in patients with various brain tumors. METHODS: A total of 49 patients who had undergone 53 examinations [5 low-grade gliomas (LGG), 16 high-grade gliomas (HGG), 6 malignant lymphomas, 4 metastases, and 22 meningiomas] underwent APTw imaging using 2D and 3D sequences. The magnetization transfer ratio asymmetry (MTR asym ) was assessed by means of region of interest measurements. Pearson correlation was performed to determine the relationship between MTR asym for the 2 methods, and Student's t test to compare MTR asym for LGG and HGG. The diagnostic accuracy to differentiate HGG from LGG of the 2 methods was compared by means of the McNemar test. RESULTS: Three-dimensional APTw imaging showed a significant correlation with 2D APTw imaging ( r = 0.79, P < 0.0001). The limits of agreement between the 2 methods were -0.021 ± 1.42%. The MTR asym of HGG (2D: 1.97 ± 0.96, 3D: 2.11 ± 0.95) was significantly higher than those of LGG (2D: 0.46 ± 0.89%, P < 0.01; 3D: 0.15 ± 1.09%, P < 0.001). The diagnostic performance of the 2 methods to differentiate HGG from LGG was not significantly different ( P = 1). CONCLUSIONS: The potential capability of 3D APTw imaging is equal to or greater than that of 2D APTw imaging and is considered at least as valuable in patients with brain tumors.

Reverse encoding distortion correction for diffusion-weighted MRI: Efficacy for improving image quality and ADC evaluation for differentiating malignant from benign areas in suspected prostatic cancer patients.

PURPOSE: The purpose of this study was to determine the influenceof reverse encoding distortion correction (RDC) on ADC measurement and its efficacy for improving image quality and diagnostic performance for differentiating malignant from benign prostatic areas on prostatic DWI. METHODS: Forty suspected prostatic cancer patients underwent DWI with or without RDC (i.e. RDC DWI or DWI) using a 3 T MR system as well as pathological examinations. The pathological examination results indicated 86 areas were malignant while 86 out of 394 areas were computationally selected as benign. SNR for benign areas and muscle and ADCs for malignant and benign areas were determined by ROI measurements on each DWI. Moreover, overall image quality was assessed with a 5-point visual scoring system on each DWI. Paired t-test or Wilcoxon's signed rank test was performed to compare SNR and overall image quality for DWIs. ROC analysis was then used to compare the diagnostic performance, and sensitivity (SE), specificity (SP) and accuracy (AC) of ADC were compared between two DWI by means of McNemar's test. RESULTS: SNR and overall image quality of RDC DWI showed significant improvements when compared with those of DWI (p < 0.05). Areas under the curve (AUC), SP and AC of DWI RDC DWI (AUC: 0.85, SP: 72.1%, AC: 79.1%) were significantly better than those of DWI (AUC: 0.79, p = 0.008; SP: 64%, p = 0.02; AC: 74.4%, p = 0.008). CONCLUSION: RDC technique has the potential to improve image quality and ability to differentiate malignant from benign prostatic areas on DWIs of suspected prostatic cancer patients.

Chemical Exchange Saturation Transfer MRI: Capability for Predicting Therapeutic Effect of Chemoradiotherapy on Non-Small Cell Lung Cancer Patients.

BACKGROUND: Amide proton transfer (APT) weighted chemical exchange saturation transfer CEST (APTw/CEST) magnetic resonance imaging (MRI) has been suggested as having the potential for assessing the therapeutic effect of brain tumors or rectal cancer. Moreover, diffusion-weighted imaging (DWI) and positron emission tomography fused with computed tomography by means of 2-[fluorine-18]-fluoro-2-deoxy-D-glucose (FDG-PET/CT) have been suggested as useful in same setting. PURPOSE: To compare the capability of APTw/CEST imaging, DWI, and FDG-PET/CT for predicting therapeutic effect of chemoradiotherapy (CRT) on stage III non-small cell lung cancer (NSCLC) patients. STUDY TYPE: Prospective. POPULATION: Eighty-four consecutive patients with Stage III NSCLC, 45 men (age range, 62-75 years; mean age, 71 years) and 39 women (age range, 57-75 years; mean age, 70 years). All patients were then divided into two groups (Response Evaluation Criteria in Solid Tumors [RECIST] responders, consisting of the complete response and partial response groups, and RECIST non-responders, consisting of the stable disease and progressive disease groups). FIELD STRENGTH/SEQUENCE: 3 T, echo planar imaging or fast advanced spin-echo (FASE) sequences for DWI and 2D half Fourier FASE sequences with magnetization transfer pulses for CEST imaging. ASSESSMENT: Magnetization transfer ratio asymmetry (MTRasym ) at 3.5 ppm, apparent diffusion coefficient (ADC), and maximum standard uptake value (SUVmax, ) on PET/CT were assessed by means of region of interest (ROI) measurements at primary tumor. STATISTICAL TESTS: Kaplan-Meier method followed by log-rank test and Cox proportional hazards regression analysis with multivariate analysis. A P value <0.05 was considered statistically significant. RESULTS: Progression-free survival (PFS) and overall survival (OS) had significant difference between two groups. MTRasym at 3.5 ppm (hazard ratio [HR] = 0.70) and SUVmax (HR = 1.41) were identified as significant predictors for PFS. Tumor staging (HR = 0.57) was also significant predictors for OS. DATA CONCLUSION: APTw/CEST imaging showed potential performance as DWI and FDG-PET/CT for predicting the therapeutic effect of CRT on stage III NSCLC patients. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY: Stage 1.

Three-Dimensional Amide Proton Transfer-Weighted Imaging for Differentiating between Glioblastoma, IDH-Wildtype and Primary Central Nervous System Lymphoma.

Distinguishing primary central nervous system lymphoma (PCNSL) from glioblastoma, isocitrate dehydrogenase (IDH)-wildtype is sometimes hard. Because the role of operation on them varies, accurate preoperative diagnosis is crucial. In this study, we evaluated whether a specific kind of chemical exchange saturation transfer imaging, i.e., amide proton transfer-weighted (APTw) imaging, was useful to distinguish PCNSL from glioblastoma, IDH-wildtype. A total of 14 PCNSL and 27 glioblastoma, IDH-wildtype cases were evaluated. There was no significant difference in the mean APTw signal values between the two groups. However, the percentile values from the 1st percentile to the 20th percentile APTw signals and the width1-100 APTw signals significantly differed. The highest area under the curve was 0.796, which was obtained from the width1-100 APTw signal values. The sensitivity and specificity values were 64.3% and 88.9%, respectively. APTw imaging was useful to distinguish PCNSL from glioblastoma, IDH-wildtype. To avoid unnecessary aggressive surgical resection, APTw imaging is recommended for cases in which PCNSL is one of the differential diagnoses.

Computed DWI MRI Results in Superior Capability for N-Stage Assessment of Non-Small Cell Lung Cancer Than That of Actual DWI, STIR Imaging, and FDG-PET/CT.

BACKGROUND: Computed diffusion-weighted imaging (cDWI) is a mathematical computation technique that generates DWIs for any b-value by using actual DWI (aDWI) data with at least two different b-values and may improve differentiation of metastatic from nonmetastatic lymph nodes. PURPOSE: To determine the appropriate b-value for cDWI to achieve a better diagnostic capability for lymph node staging (N-staging) in non-small cell lung cancer (NSCLC) patients compared to aDWI, short inversion time (TI) inversion recovery (STIR) imaging, or positron emission tomography with 2-[fluorine-18] fluoro-2-deoxy-d-glucose combined with computed tomography (FDG-PET/CT). STUDY TYPE: Prospective. SUBJECTS: A total of 245 (127 males and 118 females; mean age 72 years) consecutive histopathologically confirmed NSCLC patients. FIELD STRENGTH/SEQUENCE: A 3 T, half-Fourier single-shot turbo spin-echo sequence, electrocardiogram (ECG)-triggered STIR fast advanced spin-echo (FASE) sequence with black blood and STIR acquisition and DWI obtained by FASE with b-values of 0 and 1000 sec/mm2 . ASSESSMENT: From aDWIs with b-values of 0 and 1000 (aDWI1000 ) sec/mm2 , cDWI using 400 (cDWI400 ), 600 (cDWI600 ), 800 (cDWI800 ), and 2000 (cDWI2000 ) sec/mm2 were generated. Then, 114 metastatic and 114 nonmetastatic nodes (mediastinal and hilar lymph nodes) were selected and evaluated with a contrast ratio (CR) for each cDWI and aDWI, apparent diffusion coefficient (ADC), lymph node-to-muscle ratio (LMR) on STIR, and maximum standard uptake value (SUVmax ). STATISTICAL TESTS: Receiver operating characteristic curve (ROC) analysis, Youden index, and McNemar's test. RESULTS: Area under the curve (AUC) of CR600 was significantly larger than the CR400 , CR800 , CR2000 , aCR1000 , and SUVmax . Comparison of N-staging accuracy showed that CR600 was significantly higher than CR400 , CR2000 , ADC, aCR1000 , and SUVmax , although there were no significant differences with CR800 (P = 0.99) and LMR (P = 0.99). DATA CONCLUSION: cDWI with b-value at 600 sec/mm2 may have potential to improve N-staging accuracy as compared with aDWI, STIR, and PET/CT. EVIDENCE LEVEL: 2 TECHNICAL EFFICACY: Stage 2.

MR imaging for shoulder diseases: Effect of compressed sensing and deep learning reconstruction on examination time and imaging quality compared with that of parallel imaging.

PURPOSE: To compare capabilities of compressed sensing (CS) with and without deep learning reconstruction (DLR) with those of conventional parallel imaging (PI) with and without DLR for improving examination time and image quality of shoulder MRI for patients with various shoulder diseases. METHODS AND MATERIALS: Thirty consecutive patients with suspected shoulder diseases underwent MRI at a 3 T MR system using PI and CS. All MR data was reconstructed with and without DLR. For quantitative image quality evaluation, ROI measurements were used to determine signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR). For qualitative image quality assessment, two radiologists evaluated overall image quality, artifacts and diagnostic confidence level using a 5-point scoring system, and consensus of the two readers determined each final value. Tukey's HSD test was used to compare examination times to establish the capability of the two techniques for reducing examination time. All indexes for all methods were then compared by means of Tukey's HSD test or Wilcoxon's signed rank test. RESULTS: CS with and without DLR showed significantly shorter examination times than PI with and without DLR (p < 0.05). SNR and CNR of CS or PI with DLR were significantly higher than of those without DLR (p < 0.05). Use of DLR significantly improved overall image quality and artifact incidence of CS and PI (p < 0.05). CONCLUSION: Examination time with CS is shorter than with PI without deterioration of image quality of shoulder MRI. Moreover, DLR is useful for both CS and PI for improvement of image quality on shoulder MRI.

Comparison of utility of deep learning reconstruction on 3D MRCPs obtained with three different k-space data acquisitions in patients with IPMN.

OBJECTIVE: To compare the utility of deep learning reconstruction (DLR) for improving acquisition time, image quality, and intraductal papillary mucinous neoplasm (IPMN) evaluation for 3D MRCP obtained with parallel imaging (PI), multiple k-space data acquisition for each repetition time (TR) technique (Fast 3D mode multiple: Fast 3Dm) and compressed sensing (CS) with PI. MATERIALS AND METHODS: A total of 32 IPMN patients who had undergone 3D MRCPs obtained with PI, Fast 3Dm, and CS with PI and reconstructed with and without DLR were retrospectively included in this study. Acquisition time, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) obtained with all protocols were compared using Tukey's HSD test. Results of endoscopic ultrasound, ERCP, surgery, or pathological examination were determined as standard reference, and distribution classifications were compared among all 3D MRCP protocols by McNemar's test. RESULTS: Acquisition times of Fast 3Dm and CS with PI with and without DLR were significantly shorter than those of PI with and without DLR (p < 0.05). Each MRCP sequence with DLR showed significantly higher SNRs and CNRs than those without DLR (p < 0.05). IPMN distribution accuracy of PI with and without DLR and Fast 3Dm with DLR was significantly higher than that of Fast 3Dm without DLR and CS with PI without DLR (p < 0.05). CONCLUSION: DLR is useful for improving image quality and IPMN evaluation capability on 3D MRCP obtained with PI, Fast 3Dm, or CS with PI. Moreover, Fast 3Dm and CS with PI may play as substitution to PI for MRCP in patients with IPMN. KEY POINTS: • Mean examination times of multiple k-space data acquisitions for each TR and compressed sensing with parallel imaging were significantly shorter than that of parallel imaging (p < 0.0001). • When comparing image quality of 3D MRCPs with and without deep learning reconstruction, deep learning reconstruction significantly improved signal-to-noise ratio and contrast-to-noise ratio (p < 0.05). • IPMN distribution accuracies of parallel imaging with and without deep learning reconstruction (with vs. without: 88.0% vs. 88.0%) and multiple k-space data acquisitions for each TR with deep learning reconstruction (86.0%) were significantly higher than those of others (p < 0.05).

Deep Learning Reconstruction of Diffusion-weighted MRI Improves Image Quality for Prostatic Imaging.

Background Deep learning reconstruction (DLR) may improve image quality. However, its impact on diffusion-weighted imaging (DWI) of the prostate has yet to be assessed. Purpose To determine whether DLR can improve image quality of diffusion-weighted MRI at b values ranging from 1000 sec/mm2 to 5000 sec/mm2 in patients with prostate cancer. Materials and Methods In this retrospective study, images of the prostate obtained at DWI with a b value of 0 sec/mm2, DWI with a b value of 1000 sec/mm2 (DWI1000), DWI with a b value of 3000 sec/mm2 (DWI3000), and DWI with a b value of 5000 sec/mm2 (DWI5000) from consecutive patients with biopsy-proven cancer from January to June 2020 were reconstructed with and without DLR. Image quality was assessed using signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) from region-of-interest analysis and qualitatively assessed using a five-point visual scoring system (1 [very poor] to 5 [excellent]) for each high-b-value DWI sequence with and without DLR. The SNR, CNR, and visual score for DWI with and without DLR were compared with the paired t test and the Wilcoxon signed rank test with Bonferroni correction, respectively. Apparent diffusion coefficients (ADCs) from DWI with and without DLR were also compared with the paired t test with Bonferroni correction. Results A total of 60 patients (mean age, 67 years; age range, 49-79 years) were analyzed. DWI with DLR showed significantly higher SNRs and CNRs than DWI without DLR (P < .001); for example, with DWI1000 the mean SNR was 38.7 ± 0.6 versus 17.8 ± 0.6, respectively (P < .001), and the mean CNR was 18.4 ± 5.6 versus 7.4 ± 5.6, respectively (P < .001). DWI with DLR also demonstrated higher qualitative image quality than DWI without DLR (mean score: 4.8 ± 0.4 vs 4.0 ± 0.7, respectively, with DWI1000 [P = .001], 3.8 ± 0.7 vs 3.0 ± 0.8 with DWI3000 [P = .002], and 3.1 ± 0.8 vs 2.0 ± 0.9 with DWI5000 [P < .001]). ADCs derived with and without DLR did not differ substantially (P > .99). Conclusion Deep learning reconstruction improves the image quality of diffusion-weighted MRI scans of prostate cancer with no impact on apparent diffusion coefficient quantitation with a 3.0-T MRI system. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Turkbey in this issue.

Efficacy of Ultrashort Echo Time Pulmonary MRI for Lung Nodule Detection and Lung-RADS Classification.

Background Pulmonary MRI with ultrashort echo time (UTE) has been compared with chest CT for nodule detection and classification. However, direct comparisons of these methods' capabilities for Lung CT Screening Reporting and Data System (Lung-RADS) evaluation remain lacking. Purpose To compare the capabilities of pulmonary MRI with UTE with those of standard- or low-dose thin-section CT for Lung-RADS classification. Materials and Methods In this prospective study, standard- and low-dose chest CT (270 mA and 60 mA, respectively) and MRI with UTE were used to examine consecutive participants enrolled between January 2017 and December 2020 who met American College of Radiology Appropriateness Criteria for lung cancer screening with low-dose CT. Probability of nodule presence was assessed for all methods with a five-point visual scoring system by two board-certified radiologists. All nodules were then evaluated in terms of their Lung-RADS classification using each method. To compare nodule detection capability of the three methods, consensus for performances was rated by using jackknife free-response receiver operating characteristic analysis, and sensitivity was compared by means of the McNemar test. In addition, weighted κ statistics were used to determine the agreement between Lung-RADS classification obtained with each method and the reference standard generated from standard-dose CT evaluated by two radiologists who were not included in the image analysis session. Results A total of 205 participants (mean age: 64 years ± 7 [standard deviation], 106 men) with 1073 nodules were enrolled. Figure of merit (FOM) (P < .001) had significant differences among three modalities (standard-dose CT: FOM = 0.91, low-dose CT: FOM = 0.89, pulmonary MRI with UTE: FOM = 0.94), with no evidence of false-positive findings in participants with all modalities (P > .05). Agreements for Lung-RADS classification between all modalities and the reference standard were almost perfect (standard-dose CT: κ = 0.82, P < .001; low-dose CT: κ = 0.82, P < .001; pulmonary MRI with UTE: κ = 0.82, P < .001). Conclusion In a lung cancer screening population, ultrashort echo time pulmonary MRI was comparable to standard- or low-dose CT for Lung CT Screening Reporting and Data System classification. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Wielpütz in this issue.

Comparison of Interobserver Agreement and Diagnostic Accuracy for IASLC/ITMIG Thymic Epithelial Tumor Staging Among Co-registered FDG-PET/MRI, Whole-body MRI, Integrated FDG-PET/CT, and Conventional Imaging Examination with and without Contrast Media Administrations.

RATIONALE AND OBJECTIVES: The purpose of this study was to compare the interobserver agreements and diagnostic accuracies for IASLC/ITMIG (International Association for the Study of Lung Cancer/International Thymic Malignancies Interest Group) thymic epithelial tumor staging of co-registered fluorodeoxyglucose positron emission tomography/magnetic resonance imaging (FDG-PET/MRI), MRI, integrated fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT), and conventional imaging examination. MATERIALS AND METHODS: Prospective whole-body MRI including diffusion-weighted imaging, integrated PET/CTs, conventional imaging examinations, pathological examinations, and surgical reports, as well as follow-up examinations, were performed for 64 consecutive patients with thymic epithelial tumor. All FDG-PET/MRIs were co-registered PET data with MRI. TNM staging was evaluated by two radiologists on the basis of the IASLC/ITMIG thymic epithelial tumor staging system. Kappa statistics were determined for evaluations of agreements of all factors between each of the methods and final diagnosis. Finally, the diagnostic accuracy of each factor and of determination of the clinical stage was statistically compared to each other using McNemar test. RESULTS: Agreements for all factors between each method and final diagnosis were assessed as fair, moderate, substantial, or almost perfect (0.28 ≤ kappa value ≤ 0.80; P < .0001). Diagnostic accuracy for N factor of PET/MRI (93.8% [60/64]) and MRI (93.8% [60/64]) was significantly higher than that of conventional imaging examination (81.3% [52/64] vs PET/MRI and MRI; P = .008). In addition, diagnostic accuracy for staging of PET/MRI (84.4% [54/64]) and MRI (84.4 [54/64]) was significantly higher than that of conventional imaging examination (71.9% [46/64] vs PET/MRI and MRI; P = .008). CONCLUSIONS: Whole-body PET/MRI, MRI, and PET/CT have better interobserver agreements and accuracies than conventional imaging examination for the new IASLC/ITMIG thymic epithelial tumor staging.

Small Cell Lung Cancer Staging: Prospective Comparison of Conventional Staging Tests, FDG PET/CT, Whole-Body MRI, and Coregistered FDG PET/MRI.

BACKGROUND. Whole-body MRI and FDG PET/MRI have shown encouraging results for staging of thoracic malignancy but are poorly studied for staging of small cell lung cancer (SCLC). OBJECTIVE. The purpose of our study was to compare the performance of conventional staging tests, FDG PET/CT, whole-body MRI, and FDG PET/MRI for staging of SCLC. METHODS. This prospective study included 98 patients (64 men, 34 women; median age, 74 years) with SCLC who underwent conventional staging tests (brain MRI; neck, chest, and abdominopelvic CT; and bone scintigraphy), FDG PET/CT, and whole-body MRI within 2 weeks before treatment; coregistered FDG PET/MRI was generated. Two nuclear medicine physicians independently reviewed conventional tests and FDG PET/CT examinations in separate sessions, and two chest radiologists independently reviewed whole-body MRI and FDG PET/MRI examinations in separate sessions. Readers assessed T, N, and M categories; TNM stage; and Veterans Administration Lung Cancer Study Group (VALSG) stage. Reader pairs subsequently reached consensus. Stages determined clinically during tumor board sessions served as the reference standard. RESULTS. Accuracy for T category was higher (p < .05) for whole-body MRI (94.9%) and FDG PET/MRI (94.9%) than for FDG PET/CT (85.7%). Accuracy for N category was higher (p < .05) for whole-body MRI (84.7%), FDG PET/MRI (83.7%), and FDG PET/CT (81.6%) than for conventional staging tests (75.5%). Accuracy for M category was higher (p < .05) for whole-body MRI (94.9%), FDG PET/MRI (94.9%), and FDG PET/CT (94.9%) than for conventional staging tests (84.7%). Accuracy for TNM stage was higher (p < .05) for whole-body MRI (88.8%) and FDG PET/MRI (86.7%) than for FDG PET/CT (77.6%) and conventional staging tests (72.4%). Accuracy for VALSG stage was higher (p < .05) for whole-body MRI (95.9%), FDG PET/MRI (95.9%), and FDG PET/CT (98.0%) than for conventional staging tests (82.7%). Interobserver agreement, expressed as kappa coefficients, ranged from 0.81 to 0.94 across imaging tests and staging endpoints. CONCLUSION. FDG PET/CT, whole-body MRI, and coregistered FDG PET/MRI outperformed conventional tests for various staging endpoints in patients with SCLC. Whole-body MRI and FDG PET/MRI outperformed FDG PET/CT for T category and thus TNM stage, indicating the utility of MRI for assessing extent of local invasion in SCLC. CLINICAL IMPACT. Incorporation of either MRI approach may improve initial staging evaluation in SCLC.

The Spatial Distribution of Water Components with Similar T2 May Provide Insight into Pathways for Large Molecule Transportation in the Brain.

PURPOSE: Although there is no lymphatic system in the central nervous system (CNS), there seems to be a mechanism to remove macro molecules from the brain. Cerebrospinal fluid (CSF) and interstitial fluid (ISF) are thought to be parts of this pathway, but the details are not known. In this study, MR signal of the extracellular water was decomposed into components with distinct T2's, to obtain some information about distribution of waste material in the brain. METHODS: Images were acquired using a Curr, Purcell, Meiboom, Gill (CPMG) imaging sequence. In order to reduce T1 contamination and the signal oscillation, hard pulses were used as refocusing pulses. The signal was then decomposed into many T2 components using non-negative least squares (NNLS) in pixel-by-pixel basis. Finally, a color map was generated by assigning different color for each T2 component, then adding them together. RESULTS: From the multi-echo images, it was possible to decompose the decaying signal into separate T2 components. By adjusting the color table to create the color map, it is possible to visualize the extracellular water distribution, as well as their T2 values. Several observation points include: (1) CSF inside ventricles has very long T2 (~2 s), and seems to be relatively homogeneous, (2) subarachnoid CSF also have long T2, but there are short T2 component at the brain surface, at the surface of dura, at the blood vessels in the subarachnoid space, etc., (3) in the brain parenchyma, short T2 components (longer than intracellular component but shorter than CSF) exists along the white matter, in the choroid plexus, etc. These can be considered as distribution of macromolecules (waste materials) in the brain. CONCLUSION: From T2 component analysis it is possible to obtain some insight into pathways for the transport of large molecules in the CNS, where no lymphatic system is present.

Compressed sensing and deep learning reconstruction for women's pelvic MRI denoising: Utility for improving image quality and examination time in routine clinical practice.

PURPOSE: To demonstrate the utility of compressed sensing with parallel imaging (Compressed SPEEDER) and AiCE compared with that of conventional parallel imaging (SPEEDER) for shortening examination time and improving image quality of women's pelvic MRI. METHOD: Thirty consecutive patients with women's pelvic diseases (mean age 50 years) underwent T2-weighted imaging using Compressed SPEEDER as well as conventional SPEEDER reconstructed with and without AiCE. The examination times were recorded, and signal-to-noise ratio (SNR) was calculated for every patient. Moreover, overall image quality was assessed using a 5-point scoring system, and final scores for all patients were determined by consensus of two readers. Mean examination time, SNR and overall image quality were compared among the four data sets by Wilcoxon signed-rank test. RESULTS: Examination times for Compressed SPEEDER with and without AiCE were significantly shorter than those for conventional SPEEDER with and without AiCE (with AiCE: p < 0.0001, without AiCE: p < 0.0001). SNR of Compressed SPEEDER and of SPEEDER with AiCE was significantly superior to that of Compressed SPEEDER without AiCE (vs. Compressed SPEEDER, p = 0.01; vs. SPEEDER, p = 0.009). Overall image quality of Compressed SPEEDER with AiCE and of SPEEDER with and without AiCE was significantly higher than that of Compressed SPEEDER without AiCE (vs. Compressed SPEEDER with AiCE, p < 0.0001; vs. SPEEDER with AiCE, p < 0.0001; SPEEDER without AiCE, p = 0.0003). CONCLUSION: Image quality and shorten examination time for T2-weighted imaging in women's pelvic MRI can be significantly improved by using Compressed SPEEDER with AiCE in comparison with conventional SPEEDER, although other sequences were not tested.

3D Oxygen-Enhanced MRI at 3T MR System: Comparison With Thin-Section CT of Quantitative Capability for Pulmonary Functional Loss Assessment and Clinical Stage Classification of COPD in Smokers.

BACKGROUND: Oxygen (O2 )-enhanced MRI is mainly performed by a 2D sequence using 1.5T MR systems but trying to be obtained by a 3D sequence using a 3T MR system. PURPOSE: To compare the capability of 3D O2 -enhanced MRI and that of thin-section computed tomography (CT) for pulmonary functional loss assessment and clinical stage classification of chronic obstructive pulmonary disease (COPD) in smokers. STUDY TYPE: Prospective study. POPULATION: Fifty six smokers were included. FIELD STRENGTH/ SEQUENCE: 3T, 3D O2 -enhanced MRIs were performed with a 3D T1 -weighted fast field echo pulse sequence using the multiple flip angles. ASSESSMENTS: Smokers were classified into four stages ("Without COPD," "Mild COPD," "Moderate COPD," "Severe or very severe COPD"). Maps of regional changes in T1 values were generated from O2 -enhanced MR data. Regions of interest (ROIs) were then placed over the lung on all slices and averaged to determine mean T1 value change (ΔT1 ). Quantitative CT used the percentage of low attenuation areas within the entire lung (LAA%). STATISTICAL TESTS: ΔT1 and LAA% were correlated with pulmonary functional parameters, and compared for four stages using Tukey's Honestly Significant Difference test. Discrimination analyses were performed and McNemar's test was used for a comparison of the accuracy of the indexes. RESULTS: There were significantly higher correlations between ΔT1 and pulmonary functional parameters (-0.83 ≤ r ≤ -0.71, P < 0.05) than between LAA% and the same pulmonary functional parameters (-0.76 ≤ r ≤ -0.69, P < 0.05). ΔT1 and LAA% of the "Mild COPD" and "Moderate COPD" groups were significantly different from those of the "Severe or Very Severe COPD" group (P < 0.05). Discriminatory accuracy of ΔT1 (62.5%) and ΔT1 with LAA% (67.9%) was significantly greater than that of LAA% (48.2%, P < 0.05). DATA CONCLUSION: Compared with thin-section CT, 3D O2 -enhanced MRI has a similar capability for pulmonary functional assessment but better potential for clinical stage classification in smokers. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY STAGE: 1.