Assessment of Hepatic Steatosis Using Ultrasound-Based Techniques: Focus on Fat Quantification
Article information
Abstract
Metabolic dysfunction-associated steatotic liver disease (MASLD) is the most common chronic liver disease worldwide, affecting more than 30% of the global population, and is associated with increased liver-related morbidity and mortality, and extrahepatic complications. With the increasing prevalence of MASLD, there is an urgent need for accessible, non-invasive methods to detect hepatic steatosis. Currently, liver biopsy and magnetic resonance imaging proton density fat fraction are considered reference standards for diagnosing hepatic steatosis; however, their invasiveness and limited accessibility limit their widespread use. Ultrasonography (US) is a promising alternative owing to its cost-effectiveness and widespread accessibility. Recently, quantitative US techniques have been developed and commercialized by several vendors to measure steatosis by detecting changes in various acoustic properties associated with hepatic steatosis, making it readily accessible. Controlled attenuation parameter (CAP), which evaluates hepatic steatosis using the attenuation of the US beam, is the most widely studied algorithm as a non-imaging technique. Several other algorithms are also available on B-mode ultrasound systems, and their diagnostic performance is comparable to or better than that of the CAP. Therefore, we aimed to review current US-based methods for detecting and grading hepatic steatosis, discussing their diagnostic performance and utilization.
INTRODUCTION
Metabolic dysfunction-associated steatotic liver disease (MASLD), proposed as a new nomenclature for non-alcoholic fatty liver disease (NAFLD), is the most common cause of chronic liver disease worldwide [1]. MASLD results from fat accumulation in the liver and is associated with features of metabolic syndromes, including obesity and type 2 diabetes mellitus. Its prevalence is rising worldwide simultaneously with an increase in the prevalence of obesity and comorbid metabolic disease in more than 30% of the general population [2,3]. Liver biopsy remains the gold standard for diagnosing metabolic dysfunction-associated steatohepatitis (MASH) and plays an essential role in staging hepatic steatosis and fibrosis. According to Brunt et al., the degree of steatosis is classified as follows: grade 0, < 5% hepatocytes; grade 1, 5–33% steatosis; grade 2, 34–66% steatosis; and grade 3, > 66% steatosis [4]. However, in practice, it is difficult to perform a liver biopsy in all patients suspected of having MASLD because of high medical costs, sampling errors, low intra- and inter-observer repeatability, invasiveness (e.g., pain, bleeding, or infection), and the difficulty of repeated examination during treatment [5-8]. Moreover, hepatic steatosis is found in other chronic liver diseases, such as chronic hepatitis C, and is possibly associated with hepatic fibrosis progression [9]. Therefore, accurate measurement of liver fat content is crucial for liver donor selection because steatosis may be associated with liver regeneration and perigraft mortality [10,11].
Accordingly, there is an unmet need and growing interest in non-invasive assessment of steatosis. Conventional imaging modalities for assessing the hepatic fat content include B-mode ultrasound (US), computed tomography (CT), and magnetic resonance imaging (MRI). Particularly, MRI proton density fat fraction (PDFF) has been accepted as a reference standard for quantifying hepatic fat content as an alternative to biopsy [12,13]. However, these modalities have some limitations: lower sensitivity in detecting low degrees of fat content and substantial inter-observer variability in conventional B-mode US; exposure to ionizing radiation in CT; and high cost and limited accessibility in MRI. Recently, new quantitative US (QUS) methods have been developed to quantitatively assess hepatic fat content which rely on the analysis of radiofrequency echoes detected by a transducer, enabling the calculation of parameters to quantify the fat content in the liver [14].
Given the lack of studies examining these new US modalities, this article aimed to review various US-based techniques, with a focus on the recently developed QUS for assessing hepatic fat content, providing an overview of the concepts, advantages, limitations, and diagnostic performance of these techniques.
CONVENTIONAL B-MODE US
Conventional B-mode US is commonly used to assess hepatic steatosis because of its cost-effectiveness, safety, and widespread availability [15]. It utilizes qualitative sonographic patterns which include increased hepatic parenchymal echogenicity, blurring of the normal echogenic walls of the portal and hepatic veins, posterior beam attenuation, and poor visualization of the diaphragm, to evaluate the hepatic fat infiltration severity [16]. Additionally, B-mode US can be used to grade steatosis extent based on the degree of liver brightening and/or blurring of the vessels and diaphragm. It is reportedly 90% sensitive in detecting steatosis in at least 20% of hepatocytes but less sensitive in detecting lower degrees of hepatic fat content [17]. Furthermore, conventional B-mode US has inter-observer variability, which may result in overestimation owing to beam attenuation by the overlying non-liver fat and confounded by fibrosis and inflammation [16,18,19].
SEMI-QUANTITATIVE US
The most popular semi-quantitative methods for assessing hepatic steatosis include the hepatorenal index (HRI), Hamaguchi score, ultrasonographic fatty liver index (US-FLI), and US fatty liver score [20]. HRI indirectly measures steatosis as the brightness ratio between the liver and the renal cortex [21]. To measure HRI, two regions of interest (ROIs) are placed in segment VI of the liver and right renal cortex on the same image (Fig. 1). The measurement points should be at the same distance from the transducer to minimize the attenuation difference with depth [21,22]. Various cutoff values of HRI for hepatic steatosis have been proposed, ranging from 1.17–2.2 [23-25]. However, HRI is not suitable for patients with advanced liver fibrosis or severe renal disease because of confounding interpretations and insufficient cortical thickness. Additionally, limitations including technical dependency, intervendor variability, and a lack of large amounts of clinical validation data make it difficult to widely utilize HRI.
The hepatorenal index (HRI) is measured in patients with fatty liver using the subcostal window. It is automatically calculated using the ultrasound (US) machine's embedded formula when regions of interest at the same depth in the kidney and liver parenchyma are placed. (A) Philips US system, (B) Samsung Medison US system. Alternatively, the HRI can be measured by dividing the mean values of the histograms measured in the liver and renal areas, respectively. (C) Cannon Medical US system; (B) and (C) images were provided by Samsung Medison and Cannon Medical Systems Korea, respectively.
The Hamaguchi score is based on four US findings: hepatic parenchymal brightness and hepatorenal echo contrast (0–3), deep attenuation (0–2), and vessel blurring (0–1). According to the Hamaguchi scoring system, hepatic steatosis is graded on a scale of up to six scores [26]. A score ≥ 2 had an area under the receiver operating characteristic curve (AUC) of 0.98, with 91.7% sensitivity and 100% specificity for diagnosing MASLD [26]. US-FLI and US fatty liver scores are semi-quantitative ultrasonographic scoring systems similar to the Hamaguchi score, with slight differences in imaging parameters and the sum of the scores. Although these scoring systems have demonstrated good diagnostic performance in detecting hepatic steatosis in several studies [27-29], they are limited owing to their insufficient reproducibility and potential to underestimate the severity of hepatic steatosis due to recent technological solutions with better US penetration.
QUS PARAMETERS FOR ASSESSING HEPATIC STEATOSIS
Recently, several new US techniques have been developed to quantitatively assess hepatic steatosis. In contrast to conventional B-mode US imaging, which provides qualitative information on hepatic steatosis, QUS aims to quantify the physical phenomena related to the propagation of US into biological tissues. QUS extracts fundamental tissue properties based on the interactions of US waves with the tissue microstructure using raw radiofrequency signals detected by the transducer, allowing for the calculation of parameters to quantify hepatic fat [14,30]. T hese QUS methods include measurement of the attenuation coefficient (AC), backscatter coefficient (BSC), and sound speed [14,31].
Attenuation
Attenuation refer to the acoustic energy loss of a signal resulting from the combination of absorption, reflection, refraction, scattering, and diffusion when an US wave passes through a medium. The presence of fat in the tissue increases the attenuation, causing a signal delay [31,32]. AC is a quantitative measure of acoustic energy loss in tissues and provides a quantitative parameter analogous to the obscuration of liver structures [33]. AC is expressed in dB/cm/MHz. Compared to biologic tissues, which typically have an AC ranging from 0.01–4 dB/cm/MHz, AC in the liver is expected to range from 0.43 to 1.26 dB/cm/MHz [33-36]. The thresholds for grading steatosis are not yet well-defined; therefore, vendor-specified thresholds are recommended. Recently, with improvements in the bandwidth, sensitivity, stability, and reproducibility of commercial systems, more calibration data from reference phantoms (RPs) have been integrated into imaging systems. This facilitates easier clinical implementation without the need for external calibration.
Backscatter
Backscatter refers to the reflection or scattering of US compression waves by tissues or structures [30]. The BSC is a quantitative measurement of the US energy returned from the tissue, defined as the differential scattering cross-section per unit volume in the 180º direction [30,37,38]. Generally, higher echogenicity and BSC are observed with hepatic fat infiltration [38]. Several experimental studies have demonstrated a positive correlation between backscatter and steatosis [39-41]. BSC measurement requires obtaining a reference BSC in units of cm-1 Sr-1 a t a s pecific f requency o r a t other f requencies w ithin t he transducer bandwidth. This is followed by a comparison of the radiofrequency signal of the insonified organ with that of the reference for calibration [30,40]. Unlike the almost linear variation in frequency observed in local attenuation, the BSC displays a more complex, tissue-specific frequency behavior, as mentioned above. Therefore, additional effort is required to calibrate the BSC of the RP at a specific frequency [30]. Recently, more robust estimates have been reported in vivo using a more standardized method to compute BSC with clinical array transducers [42,43].
Speed of sound
Speed of sound is another QUS parameter previously shown to be a biomarker of hepatic steatosis which is expressed in m/s [44,45]. In a healthy human liver, it ranges from approximately 1,538–1,588 m/s. However, fat accumulation in the liver leads to a significant decrease in the speed of sound, with values as low as 1,423–1,567 m/s in fatty liver [41,46-48]; thus, speed of sound has a strong negative correlation with hepatic fat content. Most US scanners assume a constant speed of sound of 1,540 m/s [49]. Although this assumption is demonstrably invalid for an average human, its utilization in image formation still yields reasonable image quality [48].
QUS AND ITS DIAGNOSTIC PERFORMANCE
Controlled attenuated parameter (CAP)
CAP is the most widely clinically studied proprietary algorithm that has been available since 2010, by transient elastography (FibroScan®; EchoSens, Paris, France) using A-mode US (Fig. 2). The CAP has been validated in several studies [50-52]. In a previous study, CAP was measured simultaneously with vibration-controlled transient elastography using a 3.5-MHz standard probe (≥ 2.5 cm skin distance, XL probe and < 2.5 cm skin distance, M probe) [53]. Patients were positioned in the supine or dorsal decubitus position with their right arm fully abducted. The measurements were performed by scanning the right lobe of the liver through the intercostal space. The final CAP results were expressed in dB/m, ranging from 100–400 dB/m. Only examinations with at least 10 valid individual measurements with an interquartile range/median (IQR/M) < 30% or an IQR of CAP < 40 dB/m were deemed valid [51,54-56]. In a recent meta-analysis of 61 studies involving 10,537 adult patients with MASLD/MASH, good performance for steatosis was reported as AUCs of 0.924, 0.794, and 0.778 for steatosis grades ≥ S1, ≥ S2, and = S3, respectively [50]. The cutoff values for steatosis grades ≥ S1, ≥ S2, and = S3 were 302, 331, and 337 dB/m, respectively [50]. However, the cutoff values used for grading steatosis vary with the etiology of the liver disease, body mass index, and population [50,52]. In two other meta-analyses of MASLD, the CAP was unable to accurately grade hepatic steatosis [52,57]. The CAP showed inferior diagnostic performance to that of MRI-based assessment in diagnosing steatosis grade ≥ S1 (AUC 0.77 vs. 0.99) and grading hepatic steatosis (AUC 0.88, 0.73, 0.70 vs. 0.96, 0.90, 0.79 for steatosis grades ≥ S1, ≥ S2, and = S3, respectively) [58,59]. The diagnostic ability of CAP was particularly inferior to that of MRI‑PDFF and decreased with increasing BMI compared to MRI‑PDFF [60].
The controlled attenuation parameter (CAP, Echosens) is measured simultaneously on the same area with liver stiffness by vibration-controlled transient elastography. The interquartile range (IQR) reflects its measurement variability.
B-mode guided QUS
B-mode guided US for measuring hepatic fat content allows visualization of the area to be sampled while avoiding artifacts. Moreover, it can be used for morphological evaluation of the liver, assessment of portal hemodynamics using Doppler US, and evaluation of liver stiffness using two-dimensional shear wave elastography. The most studied acoustic parameters for quantifying hepatic steatosis are AC, BSC, and speed of sound; a combination of these parameters has also been used. These QUS methods of measuring the AC, depending on the vendor, include attenuation imaging (ATI), attenuation measurement (ATT), US-guided attenuation parameter (UGAP), tissue attenuation imaging (TAI), liver fat quantification (LFQ), and so on. Additionally, they allow for the simultaneous measurement of AC and backscatter, and include composite quantitative techniques like US-derived fat fraction (UDFF) and US fat fraction (USFF). The results of the studies using B-mode guided QUS are classified according to quantitative techniques and summarized in Table 1.
Studies since 2020 analyzing the accuracy of QUS in the quantification of hepatic steatosis compared with LB and MRI-based techniques
Although the detailed process for estimating hepatic steatosis differs slightly among vendors, the general measurement process is as follows: patients are positioned in the supine or slightly left lateral decubitus position. Intercostal scanning is performed using a convex probe to visualize the right hepatic lobe. With a midinspiration breath-hold, a large ROI is positioned in the hepatic parenchyma at least 2 cm away from the liver capsule with the transducer perpendicular to the skin, avoiding large vessels and masses. The loss of acoustic magnitude over a specific depth range at a particular frequency is calculated as dB/cm/MHz. According to the manufacturer’s recommendations, 5–10 measurements are obtained, except for UDFF (one measurement), and the median or average values is considered the final result [31,61]. Measurement reliability varies depending on the method used to estimate AC. For AT, a reliability index (R2 value) of ≥ 0.6–0.9 is considered valid [36,62,63], whereas for LFQ and UGAP, a valid measurement is indicated by an IQR/M value < 30% [64].
ATI
ATI is the most studied method for measuring AC, which is implanted in the Aplio i-series machines (Canon Medical Systems, Otawara, Japan) (Fig. 3A). The ATI quantifies the degree of US beam attenuation using the adjusted echo intensity, eliminating the focus-dependent beam profile and the compensated gain profile from the original signal [65]. The correlation coefficients range from 0.47–0.72 [62,66-70] when compared to the degree of histological steatosis as a reference standard and from 0.51–0.89 [35,36,65,70-73] when compared to MRI-PDFF. The AUCs for steatosis grades ≥ S1, ≥ S2, and = S3 ranged from 0.75–0.98, 0.82–0.96, and 0.79–0.95, respectively, indicating good diagnostic performance for grading hepatic steatosis. The cutoff values were 0.62–0.67, 0.64–0.74, and 0.68– 0.91 for grading hepatic steatosis ≥ S1, ≥ S2, and = S3, respectively [62,67,69,71,73-80]. A recent meta-analysis of studies conducted with the AC algorithm in 1,509 patients, reported pooled sensitivity and specificity of 76% and 84%, respectively, for ≥ S1, as well as 87% and 79%, respectively, for ≥ S2 [81]. In 2022, Bae et al. [82] compared the diagnostic performance of conventional B-mode US, CT, MRI-PDFF, CAP, and ATI in detecting steatosis grade ≥ S1 using histology as a reference standard. The MRI-PDFF showed the highest performance, with an AUC of 0.946, while ATI was the second-best modality, with an AUC of 0.892 (compared to CAP [AUC of 0.829], CT [AUC of 0.807], conventional B-mode US [AUC of 0.761]). For steatosis grade ≥ S2, all imaging modalities demonstrated a good diagnostic performance with no significant differences (AUCs of 0.947, 0.914, 0.914, 0.900, and 0.887 for MRI-PDFF, ATI, grayscale US, CAP, and CT, respectively).
Various quantitative ultrasound methods express the attenuation coefficient in dB/cm/MHz. (A) attenuation imaging (ATI), (B) ultrasound-guided attenuation parameter (UGAP), and (C) liver fat quantification (LFQ); (A) and (B) images were provided by Cannon Medical Systems Korea and General Electric Healthcare Korea, respectively.
UGAP
The UGAP is calculated from the slope based on the measured liver signal and the RP signal. It utilizes an US phantom with known attenuation and BSCs to compensate for the characteristics of transmission and reception beamforming [83]. The UGAP is available in the Logic E10 series and P10 (GE Healthcare, New York, USA) (Fig. 3B). In studies using biopsy as a reference standard for grading steatosis, correlation coefficients ranged from 0.70–0.81 [64,84,85]. However, they ranged from 0.50–0.80 when using MRI-PDFF [83,86,87]. The cutoff values for grading steatosis as ≥ S1, ≥ S2, and = S3 were 0.53–0.70, 0.60–0.74, and 0.65–0.77, with corresponding AUCs of 0.89–0.95, 0.87–0.95, and 0.82–0.96 respectively [77,83-87]. In 2023, Kang et al. [64] compared the diagnostic performance of MRI-PDFF, conventional B-mode US, CAP, and UGAP in detecting > 5% steatosis using histology as a reference standard. The AUC of ATI (0.821) was found to be similar to MRI-PDFF (AUC, 0.829) but slightly higher than those of CAP (0.788) and B-mode US (0.766). Additionally, ATI ( AUC, 0 .796) s howed t he s econd h ighest d iagnostic performance for detecting steatosis grade ≥ S2. However, its diagnostic performance was lower than that of the MRI-PDFF (AUC 0.971).
ATT
The ATT is determined by transmitting two ultrasonic waves of different frequencies along the same beam line and calculating the slope of the obtained signal ratio [88]. ATT is available in Fujifilm US systems (Fujifilm, Tokyo, Japan). When using biopsy as the reference standard and PDFF, the correlation coefficient was 0.47 and 0.80 [88,89], respectively. The AUCs for steatosis grades ≥ S1, ≥ S2, and = S3 were 0.74–0.93, 0.87–0.96, and 0.90–0.96, with cutoff values of 0.62–0.68, 0.67–0.74, and 0.73–0.78, respectively [88-90].
TAI
TAI quantifies attenuation based on the slope of the US central frequency downshift along the depth [91,92]. TAI is available in the Samsung Medison US systems (Samsung Medison, Seoul, Korea) (Fig. 4B). When measuring AC using TAI, the correlation coefficients with MRI-PDFF and MRS as reference standards were 0.66–0.78 [91,93] and 0.71 [94], respectively. The AUCs for steatosis grades ≥ S1, ≥ S2 were 0.86–0.86 and 0.70–0.84, with cutoff values of 0.88–0.91 and 0.96–0.98, respectively [91,94].
Two quantitative ultrasound methods assess hepatic fat content, using both attenuation and backscatter information, and report the result as a percentage. (A) The ultrasound-derived fat fraction (UDFF) directly displays the result as a percentage on the monitor. (B, C) After measuring TAI and TSI values using the same ROI, one can confirm the ultrasound fat fraction (USFF) value as a percentage through the ultrasound report; (A), (B), and (C) images were provided by Siemens Healthineers Korea and Samsung Medison, respectively.
LFQ
Philips developed the LFQ method for estimating AC, which is available for the Elite and Affiniti US systems (Philips, Amsterdam, Netherlands) (Fig. 3C). In studies using MRI-PDFF as a reference standard for grading steatosis, the correlation coefficient was 0.76–0.89 [95,96], with AUCs of 0.98 for steatosis grade ≥ S1, 0.96 for steatosis grade ≥ S2, and 0.95 for S3, with cutoff values of 0.63, 0.70, and 0.84, respectively [95].
Composite quantitative techniques (UDFF and USFF)
The individual quantitative parameters are affected by imprecisions in measurements, biological variability, and unmeasured confounders. Therefore, combining simultaneously measured quantitative parameters may enhance hepatic steatosis assessment [97]. QUS techniques utilizing attenuation and backscatter are now commercially available as UDFF and USFF.
UDFF is obtained by combining both the attenuation and BSC information to measure the hepatic fat content, displaying the result as a percentage. QUS measurements of the AC and BSC required an RP. Recently, the RP has been integrated into the US system using a fixed ROI. UDFF is available in the Acuson Sequoia US system (Siemens Healthineer, Erlangen, Germany) (Fig. 4A). UDFF showed a good correlation of 0.71 [98] with histologic steatosis as the reference standard and 0.79–0.87 [98-100] when compared to MRI-PDFF. In previous studies, the AUCs for steatosis grades ≥ S1, ≥ S2, and = S3 were 0.94, 0.88, and 0.83, with cutoff values of 8.1, 15.9, and 16.1, respectively [98]. Recently, UDFF has shown robust agreement with PDFF and good diagnostic performance. The AUCs for steatosis grades ≥ S1, ≥ S2, and = S3 were 0.90, 0.95, and 0.95, with cutoff values of 11.2, 13.6, and 17.2, respectively [100]. However, UDFF showed a bias towards slightly larger values than PDFF did, with intraand inter-operator variation increasing as hepatic steatosis increased [100]. Therefore, further optimization is necessary.
Similar to UDFF, USFF assesses hepatic fat content by measuring two QUS parameters: attenuation and backscatter. It measures AC using TAI and the scatter-distribution coefficient using TSI based on the shape parameter of the Nakagami distribution [91]. These two values are then combined to derive the USFF value. USFF is implemented in the Samsung Medison US systems (Samsung Medison) (Fig. 4B, C). When compared to MRI-PDFF as the reference standard, the correlation coefficient ranged from 0.799– 0.86 [101,102]. The AUCs for steatosis grades ≥ S1, ≥ S2, and = S3 were 0.92–0.97, 0.93–0.96, and 0.91–0.95, with cutoff values of 5.7–8.7, 14.1–14.9, and 16.0–16.7, respectively [101,102].
UDFF and USFF methods offer more practical and intuitive assessment of liver fat content. However, notably, this percentage does not directly correspond to the percentage of liver fat observed histologically.
Comparison among QUS
B-mode guided QUS has the advantage of a more precise measurement of hepatic fat content, along with simultaneous visualization of hepatic parenchyma, compared with CAP. Several studies have shown a better diagnostic performance of ATI and UGAP than that of CAP [77,85,103]; however, some results indicate that the two methods exhibit a similar diagnostic performance [68-70]. In a recent meta-analysis of 13 studies involving 1,422 patients, AC showed a tendency for higher sensitivity and AUC, compared with CAP, however, the difference was not significant. CAP has the advantage of simultaneously measuring liver steatosis and fibrosis, whereas QUS has a lower failure rate and performs better at identifying grade 3 steatosis [77,85,90].
The values measured by B-mode guided QUS from different vendors showed a strong correlation [77,104]. However, significant inter-platform variability were observed, making it difficult to interchangeably use the values measured on different US systems [105,106]. Accordingly, it is difficult to use different US systems for longitudinal follow-up of patients, as specific thresholds recommended by vendors are suggested.
Limitations of B-mode guided QUS and future directions
Despite the advantages of QUS, several barriers to its widespread use exist. The cutoff values for detecting and grading liver steatosis often differ among studies, even when based on algorithms from the same US system. These differences could result from variations in etiology and lack of a standardized measurement protocol. Previous studies using CAP have shown diverse cutoff values depending on the etiology of liver disease. Similarly, although the techniques for estimating hepatic steatosis may differ from CAP, all attenuation-based QUS methods cannot be exempt from this concern. Kubale et al. [100] showed that the cutoff values for detecting mild steatosis using UDFF showed variations compared to Dillman et al. [99] and Labyed et al. [98] The authors suggested that these variations, including AUC and cutoffs, may reflect differences in patient populations, US transducers, and other factors, and proposed that the optimal threshold also depends on specific clinical scenarios. Currently, most QUS studies evaluating hepatic steatosis have focused on patients with MASLD. Therefore, further investigation into QUS techniques for other populations, including those with chronic viral hepatitis, alcoholic liver disease, and the healthy population is necessary. The ROI size and depth also affects the AC values [78,107]. Moreover, several issues that have not yet been clearly defined include greater skin-to-capsule distance [100], number of measurements to be performed [108], respiration [92], patient position [109], fasting conditions [110], and hepatic fibrosis [31,34,62,65,66,80,111]. Recently, several studies have been conducted to standardize the measurement methods [110,112]. Ferraioli et al. [112] demonstrated high repeatability when measuring liver steatosis using ATI, ATT, and UDFF with a 3-cm ROI located 2 cm below the liver capsule.
CONCLUSION
With the prevalence of MASLD continually increasing, there is a growing demand for noninvasive, accurate, and reproducible techniques to estimate hepatic steatosis. QUS is simple, costeffective, and easy to measure. Additionally, it shows a good correlation with histology and MRI-PDFF values, with a diagnostic performance comparable to or better than that of CAP. In recent years, there have been significant advances in QUS technology, resulting in the development of commercially available algorithms for B-mode US systems. Furthermore, the capabilities of US scanners to estimate several quantitative parameters, such as compression and shear wave phenomena, have been steadily improving, leading to a range of techniques and metrics [14,113]. Several studies have attempted to enhance the accuracy of measuring hepatic steatosis by combining multiple parameters, such as the integrated BSC, the signal-to-noise ratio with the AC value, or by improving existing algorithms [89,114]. Moreover, a predictive model utilizing multiparametric US combining dispersion slope, normalized local variance, and other parameters has demonstrated high diagnostic performance for MASH [79,115-117]. However, there is a need to mitigate the variability in cutoff values, develop standardized protocols, and further investigate the effects of confounding factors. As commercialized QUS technology becomes more widely available, further research is required to enable its application not only for accurately measuring hepatic fat but also for diagnosing MASH and utilizing it as a surrogate marker for the effects of MASH-specific treatment.
Acknowledgements
The author is grateful to Philips Korea, Canon Medical Systems Korea, Siemens Healthineers Korea, Samsung Medison, and General Electric Healthcare Korea for supplying information and images for LFQ, ATI, UDFF, TAI plus TSI, and UGAP.