MR imaging of diffuse liver diseases

[Reduction of Motion Artifacts in Liver MRI Using Deep Learning with High-pass Filtering]

PURPOSE

To investigate whether deep learning with high-pass filtering can be used to effectively reduce motion artifacts in magnetic resonance (MR) images of the liver.

METHODS

The subjects were 69 patients who underwent liver MR examination at our hospital. Simulated motion artifact images (SMAIs) were created from non-artifact images (NAIs) and used for deep learning. Structural similarity index measure (SSIM) and contrast ratio (CR) were used to verify the effect of reducing motion artifacts in motion artifact reduction image (MARI) output from the obtained deep learning model. In the visual assessment, reduction of motion artifacts and image sharpness were evaluated between motion artifact images (MAIs) and MARIs.

RESULTS

The SSIM values were 0.882 on the MARIs and 0.869 on the SMAIs. There was no statistically significant difference in CR between NAIs and MARIs. The visual assessment showed that MARIs had reduced motion artifacts and improved sharpness compared to MAIs.

CONCLUSION

The learning model in this study is indicated to be reduced motion artifacts without decreasing the sharpness of liver MR images.

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.

Accelerated single-shot T2-weighted fat-suppressed (FS) MRI of the liver with deep learning-based image reconstruction: qualitative and quantitative comparison of image quality with conventional T2-weighted FS sequence

OBJECTIVE

To compare the image quality of an accelerated single-shot T2-weighted fat-suppressed (FS) MRI of the liver with deep learning-based image reconstruction (DL HASTE-FS) with conventional T2-weighted FS sequence (conventional T2 FS) at 1.5 T.

METHODS

One hundred consecutive patients who underwent clinical MRI of the liver at 1.5 T including the conventional T2-weighted fat-suppressed sequence (T2 FS) and accelerated single-shot T2-weighted MRI of the liver with deep learning-based image reconstruction (DL HASTE-FS) were included. Images were reviewed independently by three blinded observers who used a 5-point confidence scale for multiple measures regarding the artifacts and image quality. Descriptive statistics and McNemar's test were used to compare image quality scores and percentage of lesions detected by each sequence, respectively. Intra-class correlation coefficient (ICC) was used to assess consistency in reader scores.

RESULTS

Acquisition time for DL HASTE-FS was 51.23 +/ 10.1 s, significantly (p < 0.001) shorter than conventional T2-FS (178.9 ± 85.3 s). DL HASTE-FS received significantly higher scores than conventional T2-FS for strength and homogeneity of fat suppression; sharpness of liver margin; sharpness of intra-hepatic vessel margin; in-plane and through-plane respiratory motion; other ghosting artefacts; liver-fat contrast; and overall image quality (all, p < 0.0001). DL HASTE-FS also received higher scores for lesion conspicuity and sharpness of lesion margin (all, p < .001), without significant difference for liver lesion contrast (p > 0.05).

CONCLUSIONS

Accelerated single-shot T2-weighted MRI of the liver with deep learning-based image reconstruction showed superior image quality compared to the conventional T2-weighted fat-suppressed sequence despite a 4-fold reduction in acquisition time.

KEY POINTS

• Conventional fat-suppressed T2-weighted sequence (conventional T2 FS) can take unacceptably long to acquire and is the most commonly repeated sequence in liver MRI due to motion. • DL HASTE-FS demonstrated superior image quality, improved respiratory motion and other ghosting artefacts, and increased lesion conspicuity with comparable liver-to-lesion contrast compared to conventional T2FS sequence. • DL HASTE- FS has the potential to replace conventional T2 FS sequence in routine clinical MRI of the liver, reducing the scan time, and improving the image quality.

Automated image quality evaluation of T2 -weighted liver MRI utilizing deep learning architecture

PURPOSE

To develop and test a deep learning approach named Convolutional Neural Network (CNN) for automated screening of T₂ -weighted (T₂ WI) liver acquisitions for nondiagnostic images, and compare this automated approach to evaluation by two radiologists.

MATERIALS AND METHODS

We evaluated 522 liver magnetic resonance imaging (MRI) exams performed at 1.5T and 3T at our institution between November 2014 and May 2016 for CNN training and validation. The CNN consisted of an input layer, convolutional layer, fully connected layer, and output layer. 351 T₂ WI were anonymized for training. Each case was annotated with a label of being diagnostic or nondiagnostic for detecting lesions and assessing liver morphology. Another independently collected 171 cases were sequestered for a blind test. These 171 T₂ WI were assessed independently by two radiologists and annotated as being diagnostic or nondiagnostic. These 171 T₂ WI were presented to the CNN algorithm and image quality (IQ) output of the algorithm was compared to that of two radiologists.

RESULTS

There was concordance in IQ label between Reader 1 and CNN in 79% of cases and between Reader 2 and CNN in 73%. The sensitivity and the specificity of the CNN algorithm in identifying nondiagnostic IQ was 67% and 81% with respect to Reader 1 and 47% and 80% with respect to Reader 2. The negative predictive value of the algorithm for identifying nondiagnostic IQ was 94% and 86% (relative to Readers 1 and 2).

CONCLUSION

We demonstrate a CNN algorithm that yields a high negative predictive value when screening for nondiagnostic T₂ WI of the liver.

LEVEL OF EVIDENCE

2 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018;47:723-728.

Liver lesion detection and characterization: role of diffusion-weighted imaging

Diffusion-weighted imaging (DWI) plays an emerging role for the assessment of focal and diffuse liver diseases. This growing interest is due to that fact that DWI is a noncontrast technique with inherent high contrast resolution, with promising results for detection and characterization of focal liver lesions. Recent advances in diffusion image quality have also added interest to this technique in the abdomen. The purpose of this review is to describe the current clinical roles of DWI for the detection and characterization of focal liver lesions, and to review pitfalls, limitations, and future directions of DWI for assessment of focal liver disease.

Diffusion and perfusion imaging of the liver

MRI of the liver is an important tool for the detection and characterization of focal liver lesions, for assessment of tumor response to treatment, and for the evaluation of diffuse liver disease. With recent advances in technology, functional MRI methods such as diffusion-weighted (DW) and perfusion-weighted (PW)-MRI are increasingly used in the abdomen with promising results, particularly in the evaluation of diffuse and focal liver diseases. In this review, we will discuss background, technical considerations, acquisition, applications, limitations and future applications of DW-MRI and PW-MRI applied in evaluation of diffuse and focal liver diseases.

Gadolinium-enhanced imaging of liver tumors and manifestations of hepatitis: pharmacodynamic and technical considerations

The ability for contrast-enhanced magnetic resonance imaging to provide significant diagnostic impact to focal and diffuse liver diseases requires knowledge, analysis, and technical optimization of the imaging techniques. Our review outlines the technical requirements needed to perform reproducible contrast-enhanced liver imaging and describes the important imaging features for assessing liver disease with conventional and alternate gadolinium-based contrast media. We present an experimental review of timing and quantification methods in dynamic contrast-enhanced liver imaging, with results of analysis showing perfusion and uptake curves in a series of patients and healthy subjects. An evidence-based methodology for reproducible arterial-phase imaging is detailed for performing a real-time bolus-tracking method. Additional diagnostic imaging features manifest at later imaging phases, in which the kinetic behavior of the contrast media serves to further specify focal lesions, while revealing detailed information of diffuse liver disease, particularly hepatic fibrosis. We review the utility of alternate gadolinium-based contrast media that undergo hepatocyte uptake, for applications related to liver tumor imaging. We also introduce results showing the potential for using alternate hepatocyte uptake agents to detect and quantify liver changes related to acute and chronic hepatitides.

Clinical applications of magnetic resonance imaging in the diagnosis of hepatic diseases: present status

Magnetic resonance imaging (MRI) is one of the most important imaging modalities commonly used for the diagnosis of various human diseases. With the advance in MRI technique, MRI has been widely used to diagnose abdominal diseases, including hepatic diseases. Here, we will review the indications, contraindications and techniques of MRI as well as its diagnostic advantages and disadvantages for hepatic diseases.

Quantitative and qualitative comparison of 1.5 and 3.0 Tesla MRI in patients with chronic liver diseases

PURPOSE

To compare the quantitative and qualitative image quality intra-individually, at 1.5 and 3.0 Tesla (T) in patients with chronic liver diseases.

MATERIALS AND METHODS

The study group included 24 consecutive patients (17 males, 7 females; mean age +/- standard deviation 56.5 +/- 11.5) who had chronic liver diseases and underwent abdominal MRI for the liver evaluation at both 1.5 and 3.0 T within a 4-month period. All MRI studies were retrospectively evaluated quantitatively and qualitatively. Quantitative analysis was performed by measuring signal to noise ratio (SNR) on various abdominal organs. Qualitative analysis was performed by two reviewers to assess image quality, artifacts, and imaging findings of chronic liver diseases. Quantitative and qualitative analyses findings were compared between 1.5 and 3.0 T using the paired Student t-test and Wilcoxon signed rank test, respectively.

RESULTS

The statistically significant increase in SNRs in various abdominal tissues ranged from 1.3- to 3.5-fold at 3.0 T compared to 1.5 T. Three-dimensional gradient echo (3D-GE) sequences demonstrated significantly higher image quality at 3.0 T (P < 0.01), whereas precontrast spoiled gradient echo (SGE) sequences demonstrated significantly higher image quality at 1.5 T (P < 0.01). T2-weighted sequences did not show any significant difference in image quality between 1.5 and 3.0 T (P > 0.05).

CONCLUSION

The SNRs of various abdominal tissues demonstrated significant increases at 3.0 T. The image quality of 3D-GE sequences was higher at 3.0 T, whereas the image quality of precontrast SGE sequences was higher at 1.5T.

Imaging of liver disease: comparison between quadruple-phase multidetector computed tomography and magnetic resonance imaging

BACKGROUND AND AIM

To compare quadruple-phase multidetector computed tomography (MDCT) and magnetic resonance imaging (MRI) for the assessment of focal and diffuse liver disease.

METHODS

Quadruple-phase contrast-enhanced MDCT and MRI of 37 consecutive patients were retrospectively reviewed by two readers (R1 and R2). In patients with focal liver lesions, the gold standard was histopathology (n = 17) and/or long-term (>6 months) follow-up imaging (n = 27) or transarterial chemoembolization (n = 1). Diffuse liver disease was confirmed by histopathology in all patients, when present.

RESULTS

Both readers identified 60 focal liver lesions on MDCT and 56 focal liver lesions on MRI. Gold standard diagnoses revealed 48 focal liver lesions in 25 patients. Diagnosis of malignant liver lesions revealed a sensitivity of 88% (R1) and 91% (R2) for MRI; 63% (R1) and 66% (R2) for MDCT; and a specificity of 75% (R1) and 79% (R2) for MRI; 50% (R1) and 64% (R2) for MDCT. MRI was superior to MDCT for the diagnosis of malignant focal liver lesions, when the mean areas under the alternative free-response receiver operating characteristic curves (A(Z)) were compared (MRI = 0.93 vs CT = 0.69), (P < 0.00001). Thirty-three patients had histopathologically confirmed diffuse liver disease. Overall diagnosis of diffuse liver disease revealed a sensitivity of 88% (R1) and 92% (R2) for MRI; 75% (R1) and 74% (R2) for MDCT; and a specificity of 100% for both modalities by both readers.

CONCLUSIONS

MRI is superior for the assessment of malignant focal liver lesions and diffuse liver disease compared to quadruple-phase MDCT, and can be considered as primary diagnostic imaging modality for liver imaging.

Quantitative ultrasound, magnetic resonance imaging, and histologic image analysis of hepatic iron accumulation in pigeons (Columbia livia)

Iron overload was induced by iron dextran i.v. in clinically healthy adult pigeons, Columbia livia, (n = 8). Hemosiderosis was induced in all treated birds. Two control pigeons received no iron injections. Pigeons did not show clinical signs of iron overload during the 6-wk study. Ultrasound examination of the liver in the pigeons receiving iron dextran was performed on days 0, 13, 28, and 42. No ultrasound images were collected on the control pigeons. Magnetic resonance imaging was performed on days 0, 13, 28, and 42 on all study pigeons and imaging sequences were collected in three different imaging formats: T1, T2, and gradient-recalled echo (GRE). Surgical liver biopsies were performed on pigeons receiving iron dextran on days 2, 16, and 45 (at necropsy). A single liver sample was collected at necropsy from the control birds. Histologic examination, quantitative image analysis, and tissue iron analysis by thin-layer chromatography were performed on each liver sample and compared to the imaging studies. Although hemosiderosis was confirmed histologically in each experimental pigeon, no significant change in pixel intensity of the ultrasound images was seen at any point in the study. Signal intensity, in all magnetic resonance imaging formats, significantly decreased in a linear fashion as the accumulation of iron increased.