Direct Comparison of Quantitative US versus Controlled Attenuation Parameter for Liver Fat Assessment Using MRI Proton Density Fat Fraction as the Reference Standard in Patients Suspected of Having NAFLD

Accurate prediction of disease-risk factors from volumetric medical scans by a deep vision model pre-trained with 2D scans

The application of machine learning to tasks involving volumetric biomedical imaging is constrained by the limited availability of annotated datasets of three-dimensional (3D) scans for model training. Here we report a deep-learning model pre-trained on 2D scans (for which annotated data are relatively abundant) that accurately predicts disease-risk factors from 3D medical-scan modalities. The model, which we named SLIViT (for 'slice integration by vision transformer'), preprocesses a given volumetric scan into 2D images, extracts their feature map and integrates it into a single prediction. We evaluated the model in eight different learning tasks, including classification and regression for six datasets involving four volumetric imaging modalities (computed tomography, magnetic resonance imaging, optical coherence tomography and ultrasound). SLIViT consistently outperformed domain-specific state-of-the-art models and was typically as accurate as clinical specialists who had spent considerable time manually annotating the analysed scans. Automating diagnosis tasks involving volumetric scans may save valuable clinician hours, reduce data acquisition costs and duration, and help expedite medical research and clinical applications.

Accuracy of controlled attenuation parameter for liver steatosis in patients at risk for metabolic dysfunction-associated steatotic liver disease using magnetic resonance imaging: a systematic review and meta-analysis

Background

The controlled attenuation parameter (CAP) enables the noninvasive assessment of liver steatosis. We performed a systematic review and meta-analysis to evaluate the diagnostic accuracy of CAP for identifying liver steatosis in patients at risk for metabolic dysfunction-associated steatotic liver disease (MASLD), using magnetic resonance imaging proton density fat fraction (MRI-PDFF) as the reference standard.

Methods

We searched Medline, Embase, Cochrane Library and gray literature sources up to March 2024. We defined MASLD as MRI-PDFF ≥5%. We also assessed the accuracy of CAP for identifying patients with MRI-PDFF ≥10%. We calculated pooled sensitivity and specificity estimates using hierarchical random-effects models. We assessed the risk of bias using the Quality Assessment of Diagnostic Accuracy Studies 2 tool, and the certainty in meta-analysis estimates using the Grading of Recommendations Assessment, Development and Evaluation framework.

Results

We included 8 studies with 1116 participants. The prevalence of MASLD ranged from 65.2-93.9%. Pooled sensitivity and specificity of CAP for MRI-PDFF ≥5% were 0.84 (95% confidence interval [CI] 0.79-0.88) and 0.77 (95%CI 0.68-0.84), respectively, with an area under the receiver operating characteristic curve (AUROC) of 0.88. The pooled sensitivity and specificity for MRI-PDFF ≥10% were 0.83 (95%CI 0.80-0.87) and 0.72 (95%CI 0.59-0.82), with an AUROC of 0.85. The certainty in our estimates was low to very low because of the high risk of bias, inconsistency and imprecision.

Conclusions

CAP has acceptable diagnostic accuracy for both MRI-PDFF ≥5% and MRI-PDFF ≥10%. Adequately powered and rigorously conducted diagnostic accuracy studies are warranted to establish the optimal CAP thresholds.

Clinical performance of ultrasonic backscatter parametric and nonparametric statistics in detecting early hepatic steatosis

Diagnosis of early hepatic steatosis would allow timely intervention. B-mode ultrasound imaging was in question for detecting early steatosis, especially with a variety of concomitant parenchymal disease. This study aimed to use the surgical specimen as a reference standard to elucidate the clinical performance of ultrasonic echogenicity and backscatter parametric and nonparametric statistics in real-world scenarios. Ultrasound radio-frequency (RF) signals of right liver lobe and patient data were collected preoperatively. Surgical specimen was then used to histologically determine staging of steatosis. A backscatter nonparametric statistic (h), a known backscatter parametric statistic, i.e., the Nakagami parameter (m), and a quantitative echo intensity (env) were calculated. Among the 236 patients included in the study, 93 were grade 0 (<5% fat) and 143 were with steatosis. All the env, m and h statistics had shown significant discriminatory power of steatosis grades (AUC = 0.643-0.907 with p-value < 0.001). Mann-Whitney U tests, however, revealed that only the backscatter statistics m and h were significantly different between the groups of grades 0 and 1 steatosis. The two-way ANOVA showed a significant confounding effect of the elevated ALT on env (p-value = 0.028), but no effect on m or h. Additionally, the severe fibrosis was found to be a significant covariate for m and h. Ultrasonic signals acquired from different scanners were found linearly comparable.

A comparative study of hepatic steatosis using two different qualitative ultrasound techniques measured based on magnetic resonance imaging-derived proton density fat fraction

AIM

This study aimed to evaluate the diagnostic performance of attenuation measurement (ATT; dual-frequency method) and improved algorithm of ATT (iATT; reference method) for the assessment of hepatic steatosis using magnetic resonance imaging (MRI)-derived proton density fat fraction (PDFF) as the reference standard.

METHODS

We prospectively analyzed 427 patients with chronic liver disease who underwent ATT, iATT, or MRI-derived PDFF. Correlation coefficients were analyzed, and diagnostic values were evaluated by area under the receiver operating characteristic curve (AUROC). The steatosis grade was categorized as S0 (<5.2%), S1 (≥5.2%, <11.3%), S2 (≥11.3%, <17.1%), and S3 (≥17.1%) according to MRI-derived PDFF values.

RESULTS

The median ATT and iATT values were 0.61 dB/cm/MHz (interquartile range 0.55-0.67 dB/cm/MHz) and 0.66 dB/cm/MHz (interquartile range 0.57-0.77 dB/cm/MHz). ATT and iATT values increased significantly as the steatosis grade increased in the order S0, S1, S2, and S3 (p < 0.001). The correlation coefficients between ATT or iATT values and MRI-derived PDFF values were 0.533 (95% confidence interval [CI] 0.477-0.610) and 0.803 (95% CI 0.766-0.834), with a significant difference between them (p < 0.001). For the detection of hepatic steatosis of ≥S1, ≥S2, and ≥S3, iATT yielded AUROCs of 0.926 (95% CI 0.901-0.951), 0.913 (95% CI 0.885-0.941), and 0.902 (95% CI 0.869-0.935), with significantly higher AUROC values than for ATT (p < 0.001, p < 0.001, p = 0.001).

CONCLUSION

iATT showed excellent diagnostic performance for hepatic steatosis, and was strongly correlated with MRI-derived PDFF, with AUROCs of ≥0.900.

Quantitative US fat fraction for noninvasive assessment of hepatic steatosis in suspected metabolic-associated fatty liver disease

OBJECTIVES

To evaluate the agreement between quantitative ultrasound system fat fraction (USFF) and proton magnetic resonance spectroscopy (1H-MRS) and the diagnostic value of USFF in assessing metabolic-associated fatty liver disease (MAFLD).

METHODS

The participants with or suspected of MAFLD were prospectively recruited and underwent 1H-MRS, USFF, and controlled attenuation parameter (CAP) measurements. The correlation between USFF and 1H-MRS was assessed using Pearson correlation coefficients. The USFF diagnostic performance for different grades of steatosis was evaluated using receiver operating characteristic curve analysis (ROC) and was compared with CAP, visual hepatic steatosis grade (VHSG).

RESULTS

A total of 113 participants (mean age 44.79 years ± 13.56 (SD); 71 males) were enrolled, of whom 98 (86.73%) had hepatic steatosis (1H-MRS ≥ 5.56%). USFF showed a good correlation (Pearson r = 0.76) with 1H-MRS and showed a linear relationship, which was superior to the correlation between CAP and 1H-MRS (Pearson r = 0.61). The USFF provided high diagnostic performance for different grades of hepatic steatosis, with ROC from 0.84 to 0.98, and the diagnostic performance was better than that of the CAP and the VHSG. The cut-off values of the USFF were different for various grades of steatosis, and the cut-off values for S1, S2, and S3 were 12.01%, 19.98%, and 22.22%, respectively.

CONCLUSIONS

There was a good correlation between USFF and 1H-MRS. Meanwhile, USFF had good diagnostic performance for hepatic steatosis and was superior to CAP and VHSG. USFF represents a superior method for noninvasive quantitative assessment of MAFLD.

CRITICAL RELEVANCE STATEMENT

Quantitative ultrasound system fat fraction (USFF) accurately assesses liver fat content and has a good correlation with magnetic resonance spectroscopy (1H-MRS) for the assessment of metabolic-associated fatty liver disease (MAFLD), as well as for providing an accurate quantitative assessment of hepatic steatosis.

KEY POINTS

Current diagnostic and monitoring modalities for metabolic-associated fatty liver disease have limitations. USFF correlated well with 1H-MRS and was superior to the CAP. USFF has good diagnostic performance for steatosis, superior to CAP and VHSG.

Interoperator reproducibility of quantitative ultrasound analysis of hepatic steatosis in participants with suspected MASLD: A prospective study

OBJECTIVES

To evaluate the reproducibility of tissue attenuation imaging (TAI) and tissue scatter distribution imaging (TSI) measurements in adults with suspected metabolic dysfunction-associated steatotic liver disease (MASLD) between radiologists with varying experience.

MATERIALS AND METHODS

Participants with suspected MASLD were prospectively recruited. TAI and TSI were performed for each participant by two radiologists with different levels of experience. Interoperability reliability was assessed on the basis of Bland-Altman analysis and intraclass correlation coefficients (ICCs). The study determined and compared the diagnostic performance of TAI and TSI with clinical prediction models using proton magnetic resonance spectroscopy (1H-MRS) as a reference.

RESULTS

A total of 180 participants (women, n = 56; men, n = 124, mean age, 46.98 ± 14.92 years; mean BMI, 25.81 ± 4.47) were enrolled from August 2022 to September 2022. Bland-Altman plots showed only slight deviation in the TAI and TSI results of the two radiologists; there was good interoperator reproducibility for TAI (ICC = 0.92) and TSI (ICC = 0.86). Senior and junior radiologists performed examinations labeled as TAI-1 and TSI-1, and TAI-2 and TSI-2, respectively. The areas under the curves (AUCs) of TAI-1, TAI-2, TSI-1, and TAI-2 for the detection of ≥5 % hepatic steatosis were 0.90, 0.96, 0.91 and 0.96, respectively. According to ROC analysis, the diagnostic performance of both radiologists for TAI and TSI was statistically similar and superior to that of the clinical prediction model.

CONCLUSIONS

TAI and TSI have good reproducibility between radiologists with different levels of experience. Meanwhile, both TAI and TSI demonstrated good diagnostic performance for hepatic steatosis (≥5%), surpassing that of clinical prediction models.

The Calculation and Evaluation of an Ultrasound-Estimated Fat Fraction in Non-Alcoholic Fatty Liver Disease and Metabolic-Associated Fatty Liver Disease

We aimed to develop a non-linear regression model that could predict the fat fraction of the liver (UEFF), similar to magnetic resonance imaging proton density fat fraction (MRI-PDFF), based on quantitative ultrasound (QUS) parameters. We measured and retrospectively collected the ultrasound attenuation coefficient (AC), backscatter-distribution coefficient (BSC-D), and liver stiffness (LS) using shear wave elastography (SWE) in 90 patients with clinically suspected non-alcoholic fatty liver disease (NAFLD), and 51 patients with clinically suspected metabolic-associated fatty liver disease (MAFLD). The MRI-PDFF was also measured in all patients within a month of the ultrasound scan. In the linear regression analysis, only AC and BSC-D showed a significant association with MRI-PDFF. Therefore, we developed prediction models using non-linear least squares analysis to estimate MRI-PDFF based on the AC and BSC-D parameters. We fitted the models on the NAFLD dataset and evaluated their performance in three-fold cross-validation repeated five times. We decided to use the model based on both parameters to calculate UEFF. The correlation between UEFF and MRI-PDFF was strong in NAFLD and very strong in MAFLD. According to a receiver operating characteristics (ROC) analysis, UEFF could differentiate between <5% vs. ≥5% and <10% vs. ≥10% MRI-PDFF steatosis with excellent, 0.97 and 0.91 area under the curve (AUC), accuracy in the NAFLD and with AUCs of 0.99 and 0.96 in the MAFLD groups. In conclusion, UEFF calculated from QUS parameters is an accurate method to quantify liver fat fraction and to diagnose ≥5% and ≥10% steatosis in both NAFLD and MAFLD. Therefore, UEFF can be an ideal non-invasive screening tool for patients with NAFLD and MAFLD risk factors.

Steatosis Quantification on Ultrasound Images by a Deep Learning Algorithm on Patients Undergoing Weight Changes

INTRODUCTION

A deep learning algorithm to quantify steatosis from ultrasound images may change a subjective diagnosis to objective quantification. We evaluate this algorithm in patients with weight changes.

MATERIALS AND METHODS

Patients (N = 101) who experienced weight changes ≥ 5% were selected for the study, using serial ultrasound studies retrospectively collected from 2013 to 2021. After applying our exclusion criteria, 74 patients from 239 studies were included. We classified images into four scanning views and applied the algorithm. Mean values from 3-5 images in each group were used for the results and correlated against weight changes.

RESULTS

Images from the left lobe (G1) in 45 patients, right intercostal view (G2) in 67 patients, and subcostal view (G4) in 46 patients were collected. In a head-to-head comparison, G1 versus G2 or G2 versus G4 views showed identical steatosis scores (R2 > 0.86, p < 0.001). The body weight and steatosis scores were significantly correlated (R2 = 0.62, p < 0.001). Significant differences in steatosis scores between the highest and lowest body weight timepoints were found (p < 0.001). Men showed a higher liver steatosis/BMI ratio than women (p = 0.026).

CONCLUSIONS

The best scanning conditions are 3-5 images from the right intercostal view. The algorithm objectively quantified liver steatosis, which correlated with body weight changes and gender.

Development and validation of multivariable quantitative ultrasound for diagnosing hepatic steatosis

This study developed and validated multivariable quantitative ultrasound (QUS) model for diagnosing hepatic steatosis. Retrospective secondary analysis of prospectively collected QUS data was performed. Participants underwent QUS examinations and magnetic resonance imaging proton density fat fraction (MRI-PDFF; reference standard). A multivariable regression model for estimating hepatic fat fraction was determined using two QUS parameters from one tertiary hospital (development set). Correlation between QUS-derived estimated fat fraction(USFF) and MRI-PDFF and diagnostic performance of USFF for hepatic steatosis (MRI-PDFF ≥ 5%) were assessed, and validated in an independent data set from the other health screening center(validation set). Development set included 173 participants with suspected NAFLD with 126 (72.8%) having hepatic steatosis; and validation set included 452 health screening participants with 237 (52.4%) having hepatic steatosis. USFF was correlated with MRI-PDFF (Pearson r = 0.799 and 0.824; development and validation set). The model demonstrated high diagnostic performance, with areas under the receiver operating characteristic curves of 0.943 and 0.924 for development and validation set, respectively. Using cutoff of 6.0% from development set, USFF showed sensitivity, specificity, positive predictive value, and negative predictive value of 87.8%, 78.6%, 81.9%, and 85.4% for diagnosing hepatic steatosis in validation set. In conclusion, multivariable QUS parameters-derived estimated fat fraction showed high diagnostic performance for detecting hepatic steatosis.

Non-Invasive Imaging Methods to Evaluate Non-Alcoholic Fatty Liver Disease with Fat Quantification: A Review

Hepatic steatosis without specific causes (e.g., viral infection, alcohol abuse, etc.) is called non-alcoholic fatty liver disease (NAFLD), which ranges from non-alcoholic fatty liver (NAFL) to non-alcoholic steatohepatitis (NASH), fibrosis, and NASH-related cirrhosis. Despite the usefulness of the standard grading system, liver biopsy has several limitations. In addition, patient acceptability and intra- and inter-observer reproducibility are also concerns. Due to the prevalence of NAFLD and limitations of liver biopsies, non-invasive imaging methods such as ultrasonography (US), computed tomography (CT), and magnetic resonance imaging (MRI) that can reliably diagnose hepatic steatosis have developed rapidly. US is widely available and radiation-free but cannot examine the entire liver. CT is readily available and helpful for detection and risk classification, significantly when analyzed using artificial intelligence; however, it exposes users to radiation. Although expensive and time-consuming, MRI can measure liver fat percentage with magnetic resonance imaging proton density fat fraction (MRI-PDFF). Specifically, chemical shift-encoded (CSE)-MRI is the best imaging indicator for early liver fat detection. The purpose of this review is to provide an overview of each imaging modality with an emphasis on the recent progress and current status of liver fat quantification.

Two-dimensional Convolutional Neural Network Using Quantitative US for Noninvasive Assessment of Hepatic Steatosis in NAFLD

Background Quantitative US (QUS) using radiofrequency data analysis has been recently introduced for noninvasive evaluation of hepatic steatosis. Deep learning algorithms may improve the diagnostic performance of QUS for hepatic steatosis. Purpose To evaluate a two-dimensional (2D) convolutional neural network (CNN) algorithm using QUS parametric maps and B-mode images for diagnosis of hepatic steatosis, with the MRI-derived proton density fat fraction (PDFF) as the reference standard, in patients with nonalcoholic fatty liver disease (NAFLD). Materials and Methods: Consecutive adult participants with suspected NAFLD were prospectively enrolled at a single academic medical center from July 2020 to June 2021. Using radiofrequency data analysis, two QUS parameters (tissue attenuation imaging [TAI] and tissue scatter-distribution imaging [TSI]) were measured. On B-mode images, hepatic steatosis was graded using visual scoring (none, mild, moderate, or severe). Using B-mode images and two QUS parametric maps (TAI and TSI) as input data, the algorithm estimated the US fat fraction (USFF) as a percentage. The correlation between the USFF and MRI PDFF was evaluated using the Pearson correlation coefficient. The diagnostic performance of the USFF for hepatic steatosis (MRI PDFF ≥5%) was evaluated using receiver operating characteristic curve analysis and compared with that of TAI, TSI, and visual scoring. Results Overall, 173 participants (mean age, 51 years ± 14 [SD]; 96 men) were included, with 126 (73%) having hepatic steatosis (MRI PDFF ≥5%). USFF correlated strongly with MRI PDFF (Pearson r = 0.86, 95% CI: 0.82, 0.90; P < .001). For diagnosing hepatic steatosis (MRI PDFF ≥5%), the USFF yielded an area under the receiver operating characteristic curve of 0.97 (95% CI: 0.93, 0.99), higher than those of TAI, TSI, and visual scoring (P = .015, .006, and < .001, respectively), with a sensitivity of 90% (95% CI: 84, 95 [114 of 126]) and a specificity of 91% (95% CI: 80, 98 [43 of 47]) at a cutoff value of 5.7%. Conclusion A deep learning algorithm using quantitative US parametric maps and B-mode images accurately estimated the hepatic fat fraction and diagnosed hepatic steatosis in participants with nonalcoholic fatty liver disease. ClinicalTrials.gov registration nos. NCT04462562, NCT04180631 © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Sidhu and Fang in this issue.

Extracting Quantitative Ultrasonic Parameters from the Backscatter Coefficient

The ultrasonic backscatter coefficient (BSC) is a fundamental quantitative ultrasound (QUS) parameter that contains rich information about the underlying tissue. Deriving parameters from the BSC is essential for fully utilizing the information contained in BSC for tissue characterization. In this chapter, we review two primary approaches for extracting parameters from the BSC versus frequency curve: the model-based approach and the model-free approach, focusing on the model-based approach, where a scattering model is fit to the observed BSC to yield model parameters. For this approach, we will attempt to unite commonly used models under a coherent theoretical framework. We will focus on the underlying assumptions and conditions for various BSC models. Computer code is provided to facilitate the use of some of the models. The strengths and weaknesses of various models are also discussed.