Comparison of Quantitative Liver US and MRI in Patients with Liver Disease

Adaptive Evolutionary Optimization of Deep Learning Architectures for Focused Liver Ultrasound Image Segmentation

Background: Liver ultrasound segmentation is challenging due to low image quality and variability. While deep learning (DL) models have been widely applied for medical segmentation, generic pre-configured models may not meet the specific requirements for targeted areas in liver ultrasound. Quantitative ultrasound (QUS) is emerging as a promising tool for liver fat measurement; however, accurately segmenting regions of interest within liver ultrasound images remains a challenge. Methods: We introduce a generalizable framework using an adaptive evolutionary genetic algorithm to optimize deep learning models, specifically U-Net, for focused liver segmentation. The algorithm simultaneously adjusts the depth (number of layers) and width (neurons per layer) of the network, dropout, and skip connections. Various architecture configurations are evaluated based on segmentation performance to find the optimal model for liver ultrasound images. Results: The model with a depth of 4 and filter sizes of [16, 64, 128, 256] achieved the highest mean adjusted Dice score of 0.921, outperforming the other configurations, using three-fold cross-validation with early stoppage. Conclusions: Adaptive evolutionary optimization enhances the deep learning architecture for liver ultrasound segmentation. Future work may extend this optimization to other imaging modalities and deep learning architectures.

Quantitative Liver Imaging in Children

ABSTRACT

In children and adults, quantitative imaging examinations determine the effectiveness of treatment for liver disease. However, pediatric liver disease differs in presentation from liver disease in adults. Children also needed to be followed for a longer period from onset and have less control of their bodies, showing more movement than adults during imaging examinations, which leads to a greater need for sedation. Thus, it is essential to appropriately tailor and accurately perform noninvasive imaging tests in these younger patients. This article is an overview of updated imaging techniques used to assess liver disease quantitatively in children. The common initial imaging study for diffuse liver disease in pediatric patients is ultrasound. In addition to preexisting echo analysis, newly developed attenuation imaging techniques have been introduced to evaluate fatty liver. Ultrasound elastography is also now actively used to evaluate liver conditions, and the broad age spectrum of the pediatric population requires caution to be taken even in the selection of probes. Magnetic resonance imaging (MRI) is another important imaging tool used to evaluate liver disease despite requiring sedation or anesthesia in young children because it allows quantitative analysis with sequences such as fat analysis and MR elastography. In addition to ultrasound and MRI, we review quantitative imaging methods specifically for fatty liver, Wilson disease, biliary atresia, hepatic fibrosis, Fontan-associated liver disease, autoimmune hepatitis, sinusoidal obstruction syndrome, and the transplanted liver. Lastly, concerns such as growth and motion that need to be addressed specifically for children are summarized.

A Comparative Study of Three Systems for Liver Magnetic Resonance Elastography

BACKGROUND

Different MR elastography (MRE) systems may produce different stiffness measurements, making direct comparison difficult in multi-center investigations.

PURPOSE

To assess the repeatability and reproducibility of liver stiffness measured by three typical MRE systems.

STUDY TYPE

Prospective.

POPULATION/PHANTOMS

Thirty volunteers without liver disease history (20 males, aged 21-28)/5 gel phantoms.

FIELD STRENGTH/SEQUENCE

3.0 T United Imaging Healthcare (UIH), 1.5 T Siemens Healthcare, 3.0 T General Electric Healthcare (GE)/Echo planar imaging-based MRE sequence.

ASSESSMENT

Wave images of volunteers and phantoms were acquired by three MRE systems. Tissue stiffness was evaluated by two observers, while phantom stiffness was assessed automatically by code. The reproducibility across three MRE systems was quantified based on the mean stiffness of each volunteer and phantom.

STATISTICAL TESTS

Intraclass correlation coefficients (ICC), coefficients of variation (CV), and Bland-Altman analyses were used to assess the interobserver reproducibility, the interscan repeatability, and the intersystem reproducibility. Paired t-tests were performed to assess the interobserver and interscan variation. Friedman tests with Dunn's multiple comparison correction were performed to assess the intersystem variation. P values less than 0.05 indicated significant difference.

RESULTS

The reproducibility of stiffness measured by the two observers demonstrated consistency with ICC > 0.92, CV < 4.32%, Mean bias < 2.23%, and P > 0.06. The repeatability of measurements obtained using the electromagnetic system for the liver revealed ICC > 0.96, CV < 3.86%, Mean bias < 0.19%, P > 0.90. When considering the range of reproducibility across the three systems for liver evaluations, results ranged with ICCs from 0.70 to 0.87, CVs from 6.46% to 10.99%, and Mean biases between 1.89% and 6.30%. Phantom studies showed similar results. The values of measured stiffness differed across all three systems significantly.

DATA CONCLUSION

Liver stiffness values measured from different MRE systems can be different, but the measurements across the three MRE systems produced consistent results with excellent reproducibility.

EVIDENCE LEVEL

1 TECHNICAL EFFICACY: Stage 2.

Pediatric MASLD: current understanding and practical approach

Metabolic dysfunction-associated steatotic liver disease (MASLD) is now the most prevalent chronic liver disease in children in industrialized countries mainly due to the rise in obesity and overweight. Besides risk of progressive liver damage, MASLD also carries an increased risk of extra-hepatic morbidity, most importantly type 2 diabetes mellitus and cardiovascular disease. Important challenges remain in the prevention, detection, and treatment of this prevalent disorder. This review outlines the epidemiology and risk factors of MASLD and provides an approach to screening, diagnosis, and treatment based on current best available evidence and expert opinion. What is known: • NAFLD/MASLD is a common disorder in children strongly related to obesity/overweight and insulin resistance. • This silent disorder is underdiagnosed due to lack of awareness and lack of simple diagnostic criteria. What is new: • New diagnostic criteria have transformed NAFLD/MASLD from a diagnosis of exclusion to a positive diagnosis with simple criteria. • Effective treatments are emerging for adults and will likely become available for children. • Identifying children with NAFLD/MASLD has become even more important due to this new treatment perspective.

Comparison of Ultrasound Attenuation Imaging Using a Linear versus a Conventional Convex Probe: A Volunteer Study

The study aimed to investigate the feasibility of attenuation imaging (ATI) measurements using a linear probe on healthy volunteers and compare measurements with the conventional convex probe. Attenuation imaging measurements of the liver tissue were taken using ultrasound with a convex and a linear probe in 33 volunteers by two examining doctors, and the measurements were repeated 4-5 weeks later by one of them. The ATI values for the linear probe were in the range of the values for the convex probe for both examiners. Measurements did not change significantly for 32 out of 33 volunteers after 4-5 weeks when using the linear probe. The size of the region of interest (ROI) only impacted the ATI values for the convex probe; it did not affect the values taken with the linear probe. Healthy volunteers were measured, and their attenuation values were compared to those from a convex probe, commonly used in steatosis evaluation. When both probes were positioned in the same liver area, they showed good agreement in attenuation values, though depth significantly affected the measurements, with both probes providing different values at different depths. The study's results aligned with previous research using the same system. Operator A and B's results were compared, demonstrating similar ranges of values for both probes. The linear probe has been demonstrated to allow for superficial measurements and attain ATI values in line with that of the convex probe in the liver.

Prospective study of quantitative liver MRI in cystic fibrosis: feasibility and comparison to PUSH cohort ultrasound

BACKGROUND

Pediatric radiologists can identify a liver ultrasound (US) pattern predictive of progression to advanced liver disease. However, reliably discriminating these US patterns remains difficult. Quantitative magnetic resonance imaging (MRI) may provide an objective measure of liver disease in cystic fibrosis (CF).

OBJECTIVE

The purpose of this study was to determine if quantitative MRI, including MR elastography, is feasible in children with CF and to determine how quantitative MRI-derived metrics compared to a research US.

MATERIALS AND METHODS

A prospective, multi-institutional trial was performed evaluating CF participants who underwent a standardized MRI. At central review, liver stiffness, fat fraction, liver volume, and spleen volume were obtained. Participants whose MRI was performed within 1 year of US were classified by US pattern as normal, homogeneous hyperechoic, heterogeneous, or nodular. Each MRI measure was compared among US grade groups using the Kruskal-Wallis test.

RESULTS

Ninety-three participants (51 females [54.8%]; mean 15.6 years [range 8.1-21.7 years]) underwent MRI. MR elastography was feasible in 87 participants (93.5%). Fifty-eight participants had an US within 1 year of MRI. In these participants, a nodular liver had significantly higher stiffness (P<0.01) than normal or homogeneous hyperechoic livers. Participants with a homogeneous hyperechoic liver had a higher fat fraction (P<0.005) than others.

CONCLUSION

MR elastography is feasible in children with CF. Participants with a nodular pattern had higher liver stiffness supporting the US determination of advanced liver disease. Participants with a homogeneous hyperechoic pattern had higher fat fractions supporting the diagnosis of steatosis.

Ultrasound attenuation imaging: a reproducible alternative for the noninvasive quantitative assessment of hepatic steatosis in children

BACKGROUND

Pediatric hepatic steatosis is a global public health concern, as an increasing number of children are affected by this condition. Liver biopsy is the gold standard diagnostic method; however, this procedure is invasive. Magnetic resonance imaging (MRI)-derived proton density fat fraction has been accepted as an alternative to biopsy. However, this method is limited by cost and availability. Ultrasound (US) attenuation imaging is an upcoming tool for noninvasive quantitative assessment of hepatic steatosis in children. A limited number of publications have focused on US attenuation imaging and the stages of hepatic steatosis in children.

OBJECTIVE

To analyze the usefulness of ultrasound attenuation imaging for the diagnosis and quantification of hepatic steatosis in children.

MATERIAL AND METHODS

Between July and November 2021, 174 patients were included and divided into two groups: group 1, patients with risk factors for steatosis (n = 147), and group 2, patients without risk factors for steatosis (n = 27). In all cases, age, sex, weight, body mass index (BMI), and BMI percentile were determined. B-mode US (two observers) and US attenuation imaging with attenuation coefficient acquisition (two independent sessions, two different observers) were performed in both groups. Steatosis was classified into four grades (0: absent, 1: mild, 2: moderate and 3: severe) using B-mode US. Attenuation coefficient acquisition was correlated with steatosis score according to Spearman's correlation. Attenuation coefficient acquisition measurements' interobserver agreement was assessed using intraclass correlation coefficients (ICC).

RESULTS

All attenuation coefficient acquisition measurements were satisfactory without technical failures. The median values for group 1 for the first session were 0.64 (0.57-0.69) dB/cm/MHz and 0.64 (0.60-0.70) dB/cm/MHz for the second session. The median values for group 2 for the first session were 0.54 (0.51-0.56) dB/cm/MHz and 0.54 (0.51-0.56) dB/cm/MHz for the second. The average attenuation coefficient acquisition was 0.65 (0.59-0.69) dB/cm/MHz for group 1 and 0.54 (0.52-0.56) dB/cm/MHz for group 2. There was excellent interobserver agreement at 0.94 (95% CI 0.92-0.96). There was substantial agreement between both observers (κ = 0.77, with a P < 0.001). There was a positive correlation between ultrasound attenuation imaging and B-mode scores for both observers (r = 0.87, P < 0.001 for observer 1; r = 0.86, P < 0.001 for observer 2). Attenuation coefficient acquisition median values were significantly different for each steatosis grade (P < 0.001). In the assessment of steatosis by B-mode US, the agreement between the two observers was moderate (κ = 0.49 and κ = 0.55, respectively, with a P < 0.001 in both cases).

CONCLUSION

US attenuation imaging is a promising tool for the diagnosis and follow-up of pediatric steatosis, which provides a more repeatable form of classification, especially at low levels of steatosis detectable in B-mode US.