A novel quantitative ultrasound technique for identifying non-alcoholic steatohepatitis

Accuracy of Non-Invasive Imaging Techniques for the Diagnosis of MASH in Patients With MASLD: A Systematic Review

BACKGROUND AND AIMS

Metabolic dysfunction-associated steatotic liver disease (MASLD) is a growing public health problem. The secondary stage in MASLD is steatohepatitis (MASH), the co-existence of steatosis and inflammation, a leading cause of progression to fibrosis and mortality. MASH resolution alone improves survival. Currently, MASH diagnosis is via liver biopsy. This study sought to evaluate the accuracy of imaging-based tests for MASH diagnosis, which offer a non-invasive method of diagnosis.

METHODS

Eight academic literature databases were searched and references of previous systematic reviews and included papers were checked for additional papers. Liver biopsy was used for reference standard.

RESULTS

We report on 69 imaging-based studies. There were 31 studies on MRI, 27 on ultrasound, five on CT, 13 on transient elastography, eight on controlled attenuation parameter (CAP) and two on scintigraphy. The pathological definition of MASH was inconsistent, making it difficult to compare studies. 55/69 studies (79.71%) were deemed high-risk of bias as they had no preset thresholds and no validation. The two largest groups of imaging papers were on MRI and ultrasound. AUROCs were up to 0.93 for MRE, 0.90 for MRI, 1.0 for magnetic resonance spectroscopy (MRS) and 0.94 for ultrasound-based studies.

CONCLUSIONS

Our study found that the most promising imaging tools are MRI techniques or ultrasound-based scores and confirmed there is potential to utilise these for MASH diagnosis. However, many publications are single studies without independent prospective validation. Without this, there is no clear imaging tool or score currently available that is reliably tested to diagnose MASH.

Ultrasonic Assessment of Liver Fibrosis Using One-Dimensional Convolutional Neural Networks Based on Frequency Spectra of Radiofrequency Signals with Deep Learning Segmentation of Liver Regions in B-Mode Images: A Feasibility Study

The early detection of liver fibrosis is of significant importance. Deep learning analysis of ultrasound backscattered radiofrequency (RF) signals is emerging for tissue characterization as the RF signals carry abundant information related to tissue microstructures. However, the existing methods only used the time-domain information of the RF signals for liver fibrosis assessment, and the liver region of interest (ROI) is outlined manually. In this study, we proposed an approach for liver fibrosis assessment using deep learning models on ultrasound RF signals. The proposed method consisted of two-dimensional (2D) convolutional neural networks (CNNs) for automatic liver ROI segmentation from reconstructed B-mode ultrasound images and one-dimensional (1D) CNNs for liver fibrosis stage classification based on the frequency spectra (amplitude, phase, and power) of the segmented ROI signals. The Fourier transform was used to obtain the three kinds of frequency spectra. Two classical 2D CNNs were employed for liver ROI segmentation: U-Net and Attention U-Net. ROI spectrum signals were normalized and augmented using a sliding window technique. Ultrasound RF signals collected (with a 3-MHz transducer) from 613 participants (Group A) were included for liver ROI segmentation and those from 237 participants (Group B) for liver fibrosis stage classification, with a liver biopsy as the reference standard (Fibrosis stage: F0 = 27, F1 = 49, F2 = 51, F3 = 49, F4 = 61). In the test set of Group A, U-Net and Attention U-Net yielded Dice similarity coefficients of 95.05% and 94.68%, respectively. In the test set of Group B, the 1D CNN performed the best when using ROI phase spectrum signals to evaluate liver fibrosis stages ≥F1 (area under the receive operating characteristic curve, AUC: 0.957; accuracy: 89.19%; sensitivity: 85.17%; specificity: 93.75%), ≥F2 (AUC: 0.808; accuracy: 83.34%; sensitivity: 87.50%; specificity: 78.57%), and ≥F4 (AUC: 0.876; accuracy: 85.71%; sensitivity: 77.78%; specificity: 94.12%), and when using the power spectrum signals to evaluate ≥F3 (AUC: 0.729; accuracy: 77.14%; sensitivity: 77.27%; specificity: 76.92%). The experimental results demonstrated the feasibility of both the 2D and 1D CNNs in liver parenchyma detection and liver fibrosis characterization. The proposed methods have provided a new strategy for liver fibrosis assessment based on ultrasound RF signals, especially for early fibrosis detection. The findings of this study shed light on deep learning analysis of ultrasound RF signals in the frequency domain with automatic ROI segmentation.

Feasibility of ultrasound radiomics based models for classification of liver fibrosis due to Schistosoma japonicum infection

BACKGROUND

Schistosomiasis japonica represents a significant public health concern in South Asia. There is an urgent need to optimize existing schistosomiasis diagnostic techniques. This study aims to develop models for the different stages of liver fibrosis caused by Schistosoma infection utilizing ultrasound radiomics and machine learning techniques.

METHODS

From 2018 to 2022, we retrospectively collected data on 1,531 patients and 5,671 B-mode ultrasound images from the Second People's Hospital of Duchang City, Jiangxi Province, China. The datasets were screened based on inclusion and exclusion criteria suitable for radiomics models. Liver fibrosis due to Schistosoma infection (LFSI) was categorized into four stages: grade 0, grade 1, grade 2, and grade 3. The data were divided into six binary classification problems, such as group 1 (grade 0 vs. grade 1) and group 2 (grade 0 vs. grade 2). Key radiomic features were extracted using Pyradiomics, the Mann-Whitney U test, and the Least Absolute Shrinkage and Selection Operator (LASSO). Machine learning models were constructed using Support Vector Machine (SVM), and the contribution of different features in the model was described by applying Shapley Additive Explanations (SHAP).

RESULTS

This study ultimately included 1,388 patients and their corresponding images. A total of 851 radiomics features were extracted for each binary classification problems. Following feature selection, 18 to 76 features were retained from each groups. The area under the receiver operating characteristic curve (AUC) for the validation cohorts was 0.834 (95% CI: 0.779-0.885) for the LFSI grade 0 vs. LFSI grade 1, 0.771 (95% CI: 0.713-0.835) for LFSI grade 1 vs. LFSI grade 2, and 0.830 (95% CI: 0.762-0.885) for LFSI grade 2 vs. LFSI grade 3.

CONCLUSION

Machine learning models based on ultrasound radiomics are feasible for classifying different stages of liver fibrosis caused by Schistosoma infection.

AutoFibroNet: A deep learning and multi-photon microscopy-derived automated network for liver fibrosis quantification in MAFLD

BACKGROUND

Liver fibrosis is the strongest histological risk factor for liver-related complications and mortality in metabolic dysfunction-associated fatty liver disease (MAFLD). Second harmonic generation/two-photon excitation fluorescence (SHG/TPEF) is a powerful tool for label-free two-dimensional and three-dimensional tissue visualisation that shows promise in liver fibrosis assessment.

AIM

To investigate combining multi-photon microscopy (MPM) and deep learning techniques to develop and validate a new automated quantitative histological classification tool, named AutoFibroNet (Automated Liver Fibrosis Grading Network), for accurately staging liver fibrosis in MAFLD.

METHODS

AutoFibroNet was developed in a training cohort that consisted of 203 Chinese adults with biopsy-confirmed MAFLD. Three deep learning models (VGG16, ResNet34, and MobileNet V3) were used to train pre-processed images and test data sets. Multi-layer perceptrons were used to fuse data (deep learning features, clinical features, and manual features) to build a joint model. This model was then validated in two further independent cohorts.

RESULTS

AutoFibroNet showed good discrimination in the training set. For F0, F1, F2 and F3-4 fibrosis stages, the area under the receiver operating characteristic curves (AUROC) of AutoFibroNet were 1.00, 0.99, 0.98 and 0.98. The AUROCs of F0, F1, F2 and F3-4 fibrosis stages for AutoFibroNet in the two validation cohorts were 0.99, 0.83, 0.80 and 0.90 and 1.00, 0.83, 0.80 and 0.94, respectively, showing a good discriminatory ability in different cohorts.

CONCLUSION

AutoFibroNet is an automated quantitative tool that accurately identifies histological stages of liver fibrosis in Chinese individuals with MAFLD.

The Past, Present, and Future of Noninvasive Test in Chronic Liver Diseases

Chronic liver disease is a major global health threat and is the 11th leading cause of death globally. A liver biopsy is frequently required in assessing the degree of steatosis and fibrosis, information that is important in diagnosis, management, and prognostication. However, liver biopsies have limitations and carry a considerable risk, leading to the development of various modalities of noninvasive testing tools. These tools have been developed in recent years and have improved markedly in diagnostic accuracy. Moving forward, they may change the practice of hepatology.

Fully connected neural network-based serum surface-enhanced Raman spectroscopy accurately identifies non-alcoholic steatohepatitis

BACKGROUND/PURPOSE OF THE STUDY

There is a need to find a standardized and low-risk diagnostic tool that can non-invasively detect non-alcoholic steatohepatitis (NASH). Surface enhanced Raman spectroscopy (SERS), which is a technique combining Raman spectroscopy (RS) with nanotechnology, has recently received considerable attention due to its potential for improving medical diagnostics. We aimed to investigate combining SERS and neural network approaches, using a liver biopsy dataset to develop and validate a new diagnostic model for non-invasively identifying NASH.

METHODS

Silver nanoparticles as the SERS-active nanostructures were mixed with blood serum to enhance the Raman scattering signals. The spectral data set was used to train the NASH classification model by a neural network primarily consisting of a fully connected residual module.

RESULTS

Data on 261 Chinese individuals with biopsy-proven NAFLD were included and a prediction model for NASH was built based on SERS spectra and neural network approaches. The model yielded an AUROC of 0.83 (95% confidence interval [CI] 0.70-0.92) in the validation set, which was better than AUROCs of both serum CK-18-M30 levels (AUROC 0.63, 95% CI 0.48-0.76, p = 0.044) and the HAIR score (AUROC 0.65, 95% CI 0.51-0.77, p = 0.040). Subgroup analyses showed that the model performed well in different patient subgroups.

CONCLUSIONS

Fully connected neural network-based serum SERS analysis is a rapid and practical tool for the non-invasive identification of NASH. The online calculator website for the estimated risk of NASH is freely available to healthcare providers and researchers ( http://www.pan-chess.cn/calculator/RAMAN_score ).

Comparative Study of Raw Ultrasound Data Representations in Deep Learning to Classify Hepatic Steatosis

Adiposity accumulation in the liver is an early-stage indicator of non-alcoholic fatty liver disease. Analysis of ultrasound (US) backscatter echoes from liver parenchyma with deep learning (DL) may offer an affordable alternative for hepatic steatosis staging. The aim of this work was to compare DL classification scores for liver steatosis using different data representations constructed from raw US data. Steatosis in N = 31 patients with confirmed or suspected non-alcoholic fatty liver disease was stratified based on fat-fraction cutoff values using magnetic resonance imaging as a reference standard. US radiofrequency (RF) frames (raw data) and clinical B-mode images were acquired. Intermediate image formation stages were modeled from RF data. Power spectrum representations and phase representations were also calculated. Co-registered patches were used to independently train 1-, 2- and 3-D convolutional neural networks (CNNs), and classifications scores were compared with cross-validation. There were 67,800 patches available for 2-D/3-D classification and 1,830,600 patches for 1-D classification. The results were also compared with radiologist B-mode annotations and quantitative ultrasound (QUS) metrics. Patch classification scores (area under the receiver operating characteristic curve [AUROC]) revealed significant reductions along successive stages of the image formation process (p < 0.001). Patient AUROCs were 0.994 for RF data and 0.938 for clinical B-mode images. For all image formation stages, 2-D CNNs revealed higher patch and patient AUROCs than 1-D CNNs. CNNs trained with power spectrum representations converged faster than those trained with RF data. Phase information, which is usually discarded in the image formation process, provided a patient AUROC of 0.988. DL models trained with RF and power spectrum data (AUROC = 0.998) provided higher scores than conventional QUS metrics and multiparametric combinations thereof (AUROC = 0.986). Radiologist annotations indicated lower hepatic steatosis classification accuracies (Acc = 0.914) with respect to magnetic resonance imaging proton density fat fraction that DL models (Acc = 0.989). Access to raw ultrasound data combined with artificial intelligence techniques may offer superior opportunities for quantitative tissue diagnostics than conventional sonographic images.

Non-invasive tests of non-alcoholic fatty liver disease

ABSTRACT

For the detection of steatosis, quantitative ultrasound imaging techniques have achieved great progress in past years. Magnetic resonance imaging proton density fat fraction is currently the most accurate test to detect hepatic steatosis. Some blood biomarkers correlate with non-alcoholic steatohepatitis, but the accuracy is modest. Regarding liver fibrosis, liver stiffness measurement by transient elastography (TE) has high accuracy and is widely used across the world. Magnetic resonance elastography is marginally better than TE but is limited by its cost and availability. Several blood biomarkers of fibrosis have been used in clinical trials and hold promise for selecting patients for treatment and monitoring treatment response. This article reviews new developments in the non-invasive assessment of non-alcoholic fatty liver disease (NAFLD). Accumulating evidence suggests that various non-invasive tests can be used to diagnose NAFLD, assess its severity, and predict the prognosis. Further studies are needed to determine the role of the tests as monitoring tools. We cannot overemphasize the importance of context in selecting appropriate tests.