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

Ultrasound detection of nonalcoholic steatohepatitis using convolutional neural networks with dual-branch global-local feature fusion architecture

Nonalcoholic steatohepatitis (NASH) is a contributing factor to liver cancer, with ultrasound B-mode imaging as the first-line diagnostic tool. This study applied deep learning to ultrasound B-scan images for NASH detection and introduced an ultrasound-specific data augmentation (USDA) technique with a dual-branch global-local feature fusion architecture (DG-LFFA) to improve model performance and adaptability across imaging conditions. A total of 137 participants were included. Ultrasound images underwent data augmentation (rotation and USDA) for training and testing convolutional neural networks-AlexNet, Inception V3, VGG16, VGG19, ResNet50, and DenseNet201. Gradient-weighted class activation mapping (Grad-CAM) analyzed model attention patterns, guiding the selection of the optimal backbone for DG-LFFA implementation. The models achieved testing accuracies of 0.81-0.83 with rotation-based data augmentation. Grad-CAM analysis showed that ResNet50 and DenseNet201 exhibited stronger liver attention. When USDA simulated datasets from different imaging conditions, DG-LFFA (based on ResNet50 and DenseNet201) improved accuracy (0.79 to 0.84 and 0.78 to 0.83), recall (0.72 to 0.81 and 0.70 to 0.78), and F1 score (0.80 to 0.84 for both models). In conclusion, deep architectures (ResNet50 and DenseNet201) enable focused analysis of liver regions for NASH detection. Under USDA-simulated imaging variations, the proposed DG-LFFA framework further improves diagnostic performance.

Quantitative Ultrasound Parameters as Predictors of Chemotherapy Toxicity in Lymphoma: A Novel Approach to Assessing Muscle Mass and Quality Based on Ultrasound Radiofrequency Signals

OBJECTIVES

The aim of this study was to use quantitative ultrasound (QUS) parameters to assess the muscle mass and quality in patients with lymphoma. Additionally, the study aimed to investigate the relationship between these QUS parameters and post-chemotherapy myelosuppression.

METHODS

The study cohort comprised 202 patients diagnosed with lymphoma (105 males, 97 females; mean age 57.0 ± 14.9 years). The skeletal muscle index (SMI) and mean skeletal muscle density (SMD) were measured on CT and used as the gold standards to evaluate low skeletal muscle mass and quality. The muscle thickness (MT) of the forearm flexor and extensor muscles was measured in both the relaxed and contracted states, while the normalized non-linear parameter B/A (MusQBOX.NLP) and normalized mean intensity (MusQBOX.NMI) were extracted from retained ultrasound radiofrequency signals. The correlations between the QUS parameters and grip strength were assessed. Models were constructed using these QUS parameters to predict low SMI and SMD, and to evaluate whether these factors were independently associated with post-chemotherapy myelosuppression.

RESULTS

The MT in both the relaxed and contracted states exhibited the strongest correlations with grip strength, while the MusQBOX.NLP and MusQBOX.NMI were only weakly correlated with grip strength. Models incorporating QUS parameters to predict low SMI and SMD achieved high area under the receiver operating characteristic curve values. The MT, MusQBOX.NLP, and MusQBOX.NMI were independent factors associated with post-chemotherapy myelosuppression.

CONCLUSIONS

QUS parameters show promise in characterizing muscle strength, mass, and quality. They are also independent factors influencing post-chemotherapy myelosuppression.

Quantitative ultrasound analysis for non-invasive assessment of hepatic steatosis in metabolic dysfunction-associated steatotic liver disease

OBJECTIVE

To evaluate the diagnostic performance of novel tissue attenuation imaging (TAI) and tissue scatter distribution imaging (TSI) tools in detecting and grading hepatic steatosis using controlled attenuation parameter (CAP) as reference standard.

METHODS

A total of 185 participants with suspected metabolic dysfunction-associated steatotic liver disease (MASLD) were prospectively enrolled, and all underwent CAP and quantitative ultrasound (QUS) testing. Correlations between CAP, biological data, TAI and TSI were assessed. The influence factors of TAI and TSI as well as the diagnostic performance of TAI and TSI in detecting hepatic steatosis were evaluated.

RESULTS

The QUS parameters (TAI and TSI) showed good intra-observer reliability with ICC of 0.972 and 0.777, respectively. The correlation of CAP with TAI was higher than that of TSI (0.724 vs 0.360, P < 0.05). Multivariate Regression analysis showed that CAP was an important influence factor of TAI and TSI (P < 0.001). The area under the ROC curve (CAP > 250 dB/m) of TAI and TSI tools for detecting hepatic steatosis was 0.876 (95% CI: 0.813-0.923; P < 0.0001) and 0.797(95% CI: 0.724-0.857; P < 0.001), respectively; the sensitivity was 67.18% and 83.21%, the specificity was 95.65% and 69.57%, and the cut-off values were 0.93 dB/cm/MHz and 91.28, respectively. When TAI and TSI were combined, the area under the ROC curve was 0.881, with a sensitivity of 80.92% and a specificity of 82.61%. The Delong test showed that the combined diagnosis of TAI and TSI was equivalent to the use of TAI alone (P > 0.05).

CONCLUSION

TAI and TSI provided good intra-observer reliability, correlated well with CAP, and helped to detect and stage hepatic steatosis.

Diagnostic of fatty liver using radiomics and deep learning models on non-contrast abdominal CT

PURPOSE

This study aims to explore the potential of non-contrast abdominal CT radiomics and deep learning models in accurately diagnosing fatty liver.

MATERIALS AND METHODS

The study retrospectively enrolled 840 individuals who underwent non-contrast abdominal CT and quantitative CT (QCT) examinations at the First Affiliated Hospital of Zhengzhou University from July 2022 to May 2023. Subsequently, these participants were divided into a training set (n = 539) and a testing set (n = 301) in a 9:5 ratio. The liver fat content measured by experienced radiologists using QCT technology served as the reference standard. The liver images from the non-contrast abdominal CT scans were then segmented as regions of interest (ROI) from which radiomics features were extracted. Two-dimensional (2D) and three-dimensional (3D) radiomics models, as well as 2D and 3D deep learning models, were developed, and machine learning models based on clinical data were constructed for the four-category diagnosis of fatty liver. The characteristic curves for each model were plotted, and area under the receiver operating characteristic curve (AUC) were calculated to assess their efficacy in the classification and diagnosis of fatty liver.

RESULTS

A total of 840 participants were included (mean age 49.1 years ± 11.5 years [SD]; 581 males), of whom 610 (73%) had fatty liver. Among the patients with fatty liver, there were 302 with mild fatty liver (CT fat fraction of 5%-14%), 155 with moderate fatty liver (CT fat fraction of 14%-28%), and 153 with severe fatty liver (CT fat fraction >28%). Among all models used for diagnosing fatty liver, the 2D radiomics model based on the random forest algorithm achieved the highest AUC (0.973), while the 2D radiomics model based on the Bagging decision tree algorithm showed the highest sensitivity (0.873), specificity (0.939), accuracy (0.864), precision (0.880), and F1 score (0.876).

CONCLUSION

A systematic comparison was conducted on the performance of 2D and 3D radiomics models, as well as deep learning models, in the diagnosis of four-category fatty liver. This comprehensive model comparison provides a broader perspective for determining the optimal model for liver fat diagnosis. It was found that the 2D radiomics models based on the random forest and Bagging decision tree algorithms show high consistency with the QCT-based classification diagnosis of fatty liver used by experienced radiologists.

Development of a Deep Learning Model for Classification of Hepatic Steatosis from Clinical Standard Ultrasound

OBJECTIVE

Early detection and monitoring of hepatic steatosis can help establish appropriate preventative measures against progression to more advanced disease. We aimed to develop a deep learning (DL) program for classification of hepatic steatosis from standard-of-care grayscale ultrasound (US) images.

METHODS

In this single-center retrospective study, we utilized grayscale US images from January 1, 2010, to October 23, 2022, labeled with magnetic resonance imaging (MRI) proton density fat fraction (MRI-PDFF) to develop a DL multi-instance program for differentiating normal (S0) from steatotic liver (S1/2/3) and normal/mild steatosis (S0/1) from moderate/severe steatosis (S2/3). Diagnostic performances were assessed with area under the receiver operating characteristic curves (AUC), sensitivity, specificity and balanced accuracy with 95% confidence interval (CI).

RESULTS

A total of 403 patients with 403 US exams were included: 171 (42%) were normal (S0: MRI-PDFF <5%), 154 (38%) had mild steatosis (S1: MRI-PDFF 5-17.4%), 29 (7%) had moderate steatosis (S2: MRI-PDFF >17.4%-22.1%) and 49 (12%) had severe steatosis (S3: MRI-PDFF >22.1%). The dataset was split to include 322 patients in train/validation and 81 patients in a holdout test set (kept blind). The S0 versus S1/2/3 model achieved 81.3% (95% CI 72.1-90.5) AUC, 81.1% (70.6-91.6) sensitivity, 71.4% (54.7-88.2) specificity and 76.3% (66.4-86.2) balanced accuracy. The S0/1 versus S2/3 model achieved 95.9% (89-100) AUC, 87.5% (71.3-100) sensitivity, 96.9% (92.7-100) specificity and 92.2% (83.8-100) balanced accuracy. A multi-class model achieved a sensitivity of 71.4% (54.7-88.2) for S0, 67.6% (52.5-82.7) for S1 and 87.5% (71.3-100) for S2/3; specificity for the same model was 81.1% (70.6-91.6) for S0, 77.3% (64.9-89.7) for S1 and 96.9% (92.7-100) for S2/3.

CONCLUSION

Our DL program offered high sensitivity and accuracy in detecting and categorizing hepatic steatosis from standard-of-care ultrasound.

Construction of a Compound Model to Enhance the Accuracy of Hepatic Fat Fraction Estimation with Quantitative Ultrasound

Background: we evaluated regression models based on quantitative ultrasound (QUS) parameters and compared them with a vendor-provided method for calculating the ultrasound fat fraction (USFF) in metabolic dysfunction-associated steatotic liver disease (MASLD). Methods: We measured the attenuation coefficient (AC) and the backscatter-distribution coefficient (BSC-D) and determined the USFF during a liver ultrasound and calculated the magnetic resonance imaging proton-density fat fraction (MRI-PDFF) and steatosis grade (S0-S4) in a combined retrospective-prospective cohort. We trained multiple models using single or various QUS parameters as independent variables to forecast MRI-PDFF. Linear and nonlinear models were trained during five-time repeated three-fold cross-validation in a retrospectively collected dataset of 60 MASLD cases. We calculated the models' Pearson correlation (r) and the intraclass correlation coefficient (ICC) in a prospectively collected test set of 57 MASLD cases. Results: The linear multivariable model (r = 0.602, ICC = 0.529) and USFF (r = 0.576, ICC = 0.54) were more reliable in S0- and S1-grade steatosis than the nonlinear multivariable model (r = 0.492, ICC = 0.461). In S2 and S3 grades, the nonlinear multivariable (r = 0.377, ICC = 0.32) and AC-only (r = 0.375, ICC = 0.313) models' approximated correlation and agreement surpassed that of the multivariable linear model (r = 0.394, ICC = 0.265). We searched a QUS parameter grid to find the optimal thresholds (AC ≥ 0.84 dB/cm/MHz, BSC-D ≥ 105), above which switching from a linear (r = 0.752, ICC = 0.715) to a nonlinear multivariable (r = 0.719, ICC = 0.641) model could improve the overall fit (r = 0.775, ICC = 0.718). Conclusions: The USFF and linear multivariable models are robust in diagnosing low-grade steatosis. Switching to a nonlinear model could enhance the fit to MRI-PDFF in advanced steatosis.

Multiparametric Ultrasound for Chronic Liver Disease

Diffuse liver disease is a substantial world-wide problem. With the combination of conventional ultrasound of the abdomen, fat quantification and elastography, appropriate staging of the patient can be assessed. This information allows for the diagnosis of steatosis and detection of fibrosis as well as prognosis, surveillance, and prioritization for treatment. With the potential for reversibility with appropriate treatment accurate assessment for the stage of chronic liver disease is critical.

ANN uncertainty estimates in assessing fatty liver content from ultrasound data

Background and objective

This article uses three different probabilistic convolutional architectures applied to ultrasound image analysis for grading Fatty Liver Content (FLC) in Metabolic Dysfunction Associated Steatotic Liver Disease (MASLD) patients. Steatosis is a new silent epidemic and its accurate measurement is an impelling clinical need, not only for hepatologists, but also for experts in metabolic and cardiovascular diseases. This paper aims to provide a robust comparison between different uncertainty quantification strategies to identify advantages and drawbacks in a real clinical setting.

Methods

We used a classical Convolutional Neural Network, a Monte Carlo Dropout, and a Bayesian Convolutional Neural Network with the goal of not only comparing the goodness of the predictions, but also to have access to an evaluation of the uncertainty associated with the outputs.

Results

We found that even if the prediction based on a single ultrasound view is reliable (relative RMSE [5.93%-12.04%]), networks based on two ultrasound views outperform them (relative RMSE [5.35%-5.87%]). In addition, the results show that the introduction of a "not confident" category contributes to increase the percentage of correctly predicted cases and to decrease the percentage of mispredicted cases, especially for semi-intrusive methods.

Conclusions

The possibility of having access to information about the confidence with which the network produces its outputs is a great advantage, both from the point of view of physicians who want to use neural networks as computer-aided diagnosis, and for developers who want to limit overfitting and obtain information about dataset problems in terms of out-of-distribution detection.

Post-contrast CT liver attenuation alone is superior to the liver-spleen difference for identifying moderate hepatic steatosis

OBJECTIVE

To assess the diagnostic performance of post-contrast CT for predicting moderate hepatic steatosis in an older adult cohort undergoing a uniform CT protocol, utilizing hepatic and splenic attenuation values.

MATERIALS AND METHODS

A total of 1676 adults (mean age, 68.4 ± 10.2 years; 1045M/631F) underwent a CT urothelial protocol that included unenhanced, portal venous, and 10-min delayed phases through the liver and spleen. Automated hepatosplenic segmentation for attenuation values (in HU) was performed using a validated deep-learning tool. Unenhanced liver attenuation < 40.0 HU, corresponding to > 15% MRI-based proton density fat, served as the reference standard for moderate steatosis.

RESULTS

The prevalence of moderate or severe steatosis was 12.9% (216/1676). The diagnostic performance of portal venous liver HU in predicting moderate hepatic steatosis (AUROC = 0.943) was significantly better than the liver-spleen HU difference (AUROC = 0.814) (p < 0.001). Portal venous phase liver thresholds of 80 and 90 HU had a sensitivity/specificity for moderate steatosis of 85.6%/89.6%, and 94.9%/74.7%, respectively, whereas a liver-spleen difference of -40 HU and -10 HU had a sensitivity/specificity of 43.5%/90.0% and 92.1%/52.5%, respectively. Furthermore, livers with moderate-severe steatosis demonstrated significantly less post-contrast enhancement (mean, 35.7 HU vs 47.3 HU; p < 0.001).

CONCLUSION

Moderate steatosis can be reliably diagnosed on standard portal venous phase CT using liver attenuation values alone. Consideration of splenic attenuation appears to add little value. Moderate steatosis not only has intrinsically lower pre-contrast liver attenuation values (< 40 HU), but also enhances less, typically resulting in post-contrast liver attenuation values of 80 HU or less.

CLINICAL RELEVANCE STATEMENT

Moderate steatosis can be reliably diagnosed on post-contrast CT using liver attenuation values alone. Livers with at least moderate steatosis enhance less than those with mild or no steatosis, which combines with the lower intrinsic attenuation to improve detection.

KEY POINTS

The liver-spleen attenuation difference is frequently utilized in routine practice but appears to have performance limitations. The liver-spleen attenuation difference is less effective than liver attenuation for moderate steatosis. Moderate and severe steatosis can be identified on standard portal venous phase CT using liver attenuation alone.

Assessing Quality of Ultrasound Attenuation Coefficient Results for Liver Fat Quantification

BACKGROUND/OBJECTIVES

Algorithms for quantifying liver fat content based on the ultrasound attenuation coefficient (AC) are currently available; however, little is known about whether their accuracy increases by applying quality criteria such as the interquartile range-to-median ratio (IQR/M) or whether the median or average AC value should be used.

METHODS

AC measurements were performed with the Aplio i800 ultrasound system using the attenuation imaging (ATI) algorithm (Canon Medical Systems, Otawara, Tochigi, Japan). Magnetic resonance imaging proton density fat fraction (MRI-PDFF) was the reference standard. The diagnostic performance of the AC median value of 5 measurements (AC-M) was compared to that of AC average value (AC-A) of 5 or 3 acquisitions and different levels of IQR/M for median values or standard deviation/average (SD/A) for average values were also analyzed. Concordance between AC-5M, AC-5A, and AC3A was evaluated with concordance correlation coefficient (CCC).

RESULTS

A total of 182 individuals (94 females; mean age, 51.2y [SD: 15]) were evaluated. A total of 77 (42.3%) individuals had S0 steatosis (MRI-PDFF < 6%), 75 (41.2%) S1 (MRI-PDFF 6-17%), 10 (5.5%) S2 (MRI-PDFF 17.1-22%), and 20 (11%) S3 (MRI-PDFF ≥ 22.1%). Concordance of AC-5A and AC-3A with AC-5M was excellent (CCC: 0.99 and 0.96, respectively). The correlation with MRI-PDFF was almost perfect. Diagnostic accuracy of AC-5M, AC-5A, and AC3A was not significantly affected by different levels of IQR/M or SD/A.

CONCLUSIONS

The accuracy of AC in quantifying liver fat content was not affected by reducing the number of acquisitions (from five to three), by using the mean instead of the median, or by reducing the IQR/M or SD/A to ≤5%.

Non-invasive diagnosis of pancreatic steatosis with ultrasound images using deep learning network

Objective

This study aimed to verify whether pancreatic steatosis (PS) is an independent risk factor for type 2 diabetes mellitus (T2DM). We also developed and validated a deep learning model for the diagnosis of PS using ultrasonography (US) images based on histological classifications.

Methods

In this retrospective study, we analysed data from 139 patients who underwent US imaging of the pancreas followed by pancreatic resection at our medical institution. Logistic regression analysis was employed to ascertain the independent predictors of T2DM. The diagnostic efficacy of the deep learning model for PS was assessed using receiver operating characteristic curve analysis and compared with traditional visual assessment methodology in US imaging.

Results

The incidence rate of PS in the study cohort was 64.7 %. Logistic regression analysis revealed that age (P = 0.003) and the presence of PS (P = 0.048) were independent factors associated with T2DM. The deep learning model demonstrated robust diagnostic capabilities for PS, with areas under the curve of 0.901 and 0.837, sensitivities of 0.895 and 0.920, specificities of 0.700 and 0.765, accuracies of 0.814 and 0.857, and F1-scores of 0.850 and 0.885 for the training and validation cohorts, respectively. These metrics significantly outperformed those of conventional US imaging (P < 0.001 and P = 0.045, respectively).

Conclusion

The deep learning model significantly enhanced the diagnostic accuracy of conventional ultrasound for PS detection. Its high sensitivity could facilitate widespread screening for PS in large populations, aiding in the early identification of individuals at an elevated risk for T2DM in routine clinical practice.

Quantitative Ultrasound and Ultrasound-Based Elastography for Chronic Liver Disease: Practical Guidance, From the AJR Special Series on Quantitative Imaging

Quantitative ultrasound (QUS) and ultrasound-based elastography techniques are emerging as non-invasive effective methods for assessing chronic liver disease. They are more accurate than B-mode imaging alone and more accessible than MRI as alternatives to liver biopsy. Early detection and monitoring of diffuse liver processes such as steatosis, inflammation, and fibrosis play an important role in guiding patient management. The most widely available and validated techniques are attenuation-based QUS techniques and shear-wave elastography techniques that measure shear-wave speed. Other techniques are supported by a growing body of evidence and are increasingly commercialized. This review explains general physical concepts of QUS and ultrasound-based elastography techniques for evaluating chronic liver disease. The first section describes QUS techniques relying on attenuation, backscatter, and speed of sound. The second section discusses ultrasound-based elastography techniques analyzing shear-wave speed, shear-wave dispersion, and shear-wave attenuation. With an emphasis on clinical implementation, each technique's diagnostic performance along with thresholds for various clinical applications are summarized, to provide guidance on analysis and reporting for radiologists. Measurement methods, advantages, and limitations are also discussed. The third section explores developments in quantitative contrast-enhanced and vascular ultrasound that are relevant to chronic liver disease evaluation.

Artificial intelligence techniques in liver cancer

Hepatocellular Carcinoma (HCC), the most common primary liver cancer, is a significant contributor to worldwide cancer-related deaths. Various medical imaging techniques, including computed tomography, magnetic resonance imaging, and ultrasound, play a crucial role in accurately evaluating HCC and formulating effective treatment plans. Artificial Intelligence (AI) technologies have demonstrated potential in supporting physicians by providing more accurate and consistent medical diagnoses. Recent advancements have led to the development of AI-based multi-modal prediction systems. These systems integrate medical imaging with other modalities, such as electronic health record reports and clinical parameters, to enhance the accuracy of predicting biological characteristics and prognosis, including those associated with HCC. These multi-modal prediction systems pave the way for predicting the response to transarterial chemoembolization and microvascular invasion treatments and can assist clinicians in identifying the optimal patients with HCC who could benefit from interventional therapy. This paper provides an overview of the latest AI-based medical imaging models developed for diagnosing and predicting HCC. It also explores the challenges and potential future directions related to the clinical application of AI techniques.

WFUMB Guidelines/Guidance on Liver Multiparametric Ultrasound. Part 2: Guidance on Liver Fat Quantification

The World Federation for Ultrasound in Medicine and Biology (WFUMB) has promoted the development of this document on multiparametric ultrasound. Part 2 is a guidance on the use of the available tools for the quantification of liver fat content with ultrasound. These are attenuation coefficient, backscatter coefficient, and speed of sound. All of them use the raw data of the ultrasound beam to estimate liver fat content. This guidance has the aim of helping the reader in understanding how they work and interpret the results. Confounding factors are discussed and a standardized protocol for measurement acquisition is suggested to mitigate them. The recommendations were based on published studies and experts' opinion but were not formally graded because the body of evidence remained low at the time of drafting this document.

Survey on Value Elements Provided by Artificial Intelligence and Their Eligibility for Insurance Coverage With an Emphasis on Patient-Centered Outcomes

OBJECTIVE

This study aims to explore the opinions on the insurance coverage of artificial intelligence (AI), as categorized based on the distinct value elements offered by AI, with a specific focus on patient-centered outcomes (PCOs). PCOs are distinguished from traditional clinical outcomes and focus on patient-reported experiences and values such as quality of life, functionality, well-being, physical or emotional status, and convenience.

MATERIALS AND METHODS

We classified the value elements provided by AI into four dimensions: clinical outcomes, economic aspects, organizational aspects, and non-clinical PCOs. The survey comprised three sections: 1) experiences with PCOs in evaluating AI, 2) opinions on the coverage of AI by the National Health Insurance of the Republic of Korea when AI demonstrated benefits across the four value elements, and 3) respondent characteristics. The opinions regarding AI insurance coverage were assessed dichotomously and semi-quantitatively: non-approval (0) vs. approval (on a 1-10 weight scale, with 10 indicating the strongest approval). The survey was conducted from July 4 to 26, 2023, using a web-based method. Responses to PCOs and other value elements were compared.

RESULTS

Among 200 respondents, 44 (22%) were patients/patient representatives, 64 (32%) were industry/developers, 60 (30%) were medical practitioners/doctors, and 32 (16%) were government health personnel. The level of experience with PCOs regarding AI was low, with only 7% (14/200) having direct experience and 10% (20/200) having any experience (either direct or indirect). The approval rate for insurance coverage for PCOs was 74% (148/200), significantly lower than the corresponding rates for other value elements (82.5%-93.5%; P ≤ 0.034). The approval strength was significantly lower for PCOs, with a mean weight ± standard deviation of 5.1 ± 3.5, compared to other value elements (P ≤ 0.036).

CONCLUSION

There is currently limited demand for insurance coverage for AI that demonstrates benefits in terms of non-clinical PCOs.

Inter-platform reproducibility of ultrasound-based fat fraction for evaluating hepatic steatosis in nonalcoholic fatty liver disease

OBJECTIVES

To evaluate the inter-platform reproducibility of ultrasound-based fat fraction examination in nonalcoholic fatty liver disease (NAFLD).

METHODS

Patients suspected of having NAFLD were prospectively enrolled from January 2023. Ultrasound-based fat fraction examinations were performed using two different platforms (ultrasound-derived fat fraction [UDFF] and quantitative ultrasound-derived estimated fat fraction [USFF]) on the same day. The correlation between UDFF and USFF was assessed using Pearson correlation coefficient. Intraclass correlation coefficient (ICC), Bland-Altman analysis with 95% limits of agreement (LOAs), and the coefficient of variation (CV) were used to assess inter-platform reproducibility.

RESULTS

A total of 41 patients (21 men and 20 women; mean age, 53.9 ± 12.6 years) were analyzed. Moderate correlation was observed between UDFF and USFF (Pearson's r = 0.748; 95% confidence interval [CI]: 0.572-0.858). On Bland-Altman analysis, the mean difference between UDFF and USFF values was 1.3% with 95% LOAs ranging from -8.0 to 10.6%. The ICC between UDFF and USFF was 0.842 (95% CI: 0.703-0.916), with a CV of 29.9%.

CONCLUSION

Substantial inter-platform variability was observed among different ultrasound-based fat fraction examinations. Therefore, it is not appropriate to use ultrasound-based fat fraction values obtained from different vendors interchangeably.

CRITICAL RELEVANCE STATEMENT

Considering the substantial inter-platform variability in ultrasound-based fat fraction assessments, caution is imperative when interpreting and comparing fat fraction values obtained from different ultrasound platforms in clinical practice.

KEY POINTS

• Inter-platform reproducibility of ultrasound-based fat fraction examinations is important for its clinical application. • Significant variability across different ultrasound-based fat fraction examinations was observed. • Using ultrasound-based fat fraction values from different vendors interchangeably is not advisable.

Detection of Moderate Hepatic Steatosis on Portal Venous Phase Contrast-Enhanced CT: Evaluation Using an Automated Artificial Intelligence Tool

BACKGROUND. Precontrast CT is an established means of evaluating for hepatic steatosis; postcontrast CT has historically been limited for this purpose. OBJECTIVE. The purpose of this study was to evaluate the diagnostic performance of portal venous phase postcontrast CT in detecting at least moderate hepatic steatosis using liver and spleen attenuation measurements determined by an automated artificial intelligence (AI) tool. METHODS. This retrospective study included 2917 patients (1381 men, 1536 women; mean age, 56.8 years) who underwent a CT examination that included at least two series through the liver. Examinations were obtained from an AI vendor's data lake of data from 24 centers in one U.S. health care network and 29 centers in one Israeli health care network. An automated deep learning tool extracted liver and spleen attenuation measurements. The reference for at least moderate steatosis was precontrast liver attenuation of less than 40 HU (i.e., estimated liver fat > 15%). A radiologist manually reviewed examinations with outlier AI results to confirm portal venous timing and identify issues impacting attenuation measurements. RESULTS. After outlier review, analysis included 2777 patients with portal venous phase images. Prevalence of at least moderate steatosis was 13.9% (387/2777). Patients without and with at least moderate steatosis, respectively, had mean postcontrast liver attenuation of 104.5 ± 18.1 (SD) HU and 67.1 ± 18.6 HU (p < .001); a mean difference in postcontrast attenuation between the liver and the spleen (hereafter, postcontrast liver-spleen attenuation difference) of -7.6 ± 16.4 (SD) HU and -31.8 ± 20.3 HU (p < .001); and mean liver enhancement of 49.3 ± 15.9 (SD) HU versus 38.6 ± 13.6 HU (p < .001). Diagnostic performance for the detection of at least moderate steatosis was higher for postcontrast liver attenuation (AUC = 0.938) than for the postcontrast liver-spleen attenuation difference (AUC = 0.832) (p < .001). For detection of at least moderate steatosis, postcontrast liver attenuation had sensitivity and specificity of 77.8% and 93.2%, respectively, at less than 80 HU and 90.5% and 78.4%, respectively, at less than 90 HU; the postcontrast liver-spleen attenuation difference had sensitivity and specificity of 71.4% and 79.3%, respectively, at less than -20 HU and 87.0% and 62.1%, respectively, at less than -10 HU. CONCLUSION. Postcontrast liver attenuation outperformed the postcontrast liver-spleen attenuation difference for detecting at least moderate steatosis in a heterogeneous patient sample, as evaluated using an automated AI tool. Splenic attenuation likely is not needed to assess for at least moderate steatosis on postcontrast images. CLINICAL IMPACT. The technique could promote early detection of clinically significant nonalcoholic fatty liver disease through individualized or large-scale opportunistic evaluation.

The Future Is Beyond Bright: The Evolving Role of Quantitative US for Fatty Liver Disease

Nonalcoholic fatty liver disease (NAFLD) is a common cause of morbidity and mortality. Nonfocal liver biopsy is the historical reference standard for evaluating NAFLD, but it is limited by invasiveness, high cost, and sampling error. Imaging methods are ideally situated to provide quantifiable results and rule out other anatomic diseases of the liver. MRI and US have shown great promise for the noninvasive evaluation of NAFLD. US is particularly well suited to address the population-level problem of NAFLD because it is lower-cost, more available, and more tolerable to a broader range of patients than MRI. Noninvasive US methods to evaluate liver fibrosis are widely available, and US-based tools to evaluate steatosis and inflammation are gaining traction. US techniques including shear-wave elastography, Doppler spectral imaging, attenuation coefficient, hepatorenal index, speed of sound, and backscatter-based estimation have regulatory clearance and are in clinical use. New methods based on channel and radiofrequency data analysis approaches have shown promise but are mostly experimental. This review discusses the advantages and limitations of clinically available and experimental approaches to sonographic liver tissue characterization for NAFLD diagnosis as well as future applications and strategies to overcome current limitations.

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.