Development and Validation of a Deep Learning System for Staging Liver Fibrosis by Using Contrast Agent-enhanced CT Images in the Liver

The application of artificial intelligence in hepatology: A systematic review

The integration of human and artificial intelligence (AI) in medicine has only recently begun but it has already become obvious that intelligent systems can dramatically improve the management of liver diseases. Big data made it possible to envisage transformative developments of the use of AI for diagnosing, predicting prognosis and treating liver diseases, but there is still a lot of work to do. If we want to achieve the 21^st century digital revolution, there is an urgent need for specific national and international rules, and to adhere to bioethical parameters when collecting data. Avoiding misleading results is essential for the effective use of AI. A crucial question is whether it is possible to sustain, technically and morally, the process of integration between man and machine. We present a systematic review on the applications of AI to hepatology, highlighting the current challenges and crucial issues related to the use of such technologies.

Combining Hepatic and Splenic CT Radiomic Features Improves Radiomic Analysis Performance for Liver Fibrosis Staging

Background: The exact focus of computed tomography (CT)-based artificial intelligence techniques when staging liver fibrosis is still not exactly known. This study aimed to determine both the added value of splenic information to hepatic information, and the correlation between important radiomic features and information exploited by deep learning models for liver fibrosis staging by CT-based radiomics. Methods: The study design is retrospective. Radiomic features were extracted from both liver and spleen on portal venous phase CT images of 252 consecutive patients with histologically proven liver fibrosis stages between 2006 and 2018. The radiomics analyses for liver fibrosis staging were done by hepatic and hepatic–splenic features, respectively. The most predictive radiomic features were automatically selected by machine learning models. Results: When using splenic–hepatic features in the CT-based radiomics analysis, the average accuracy rates for significant fibrosis, advanced fibrosis, and cirrhosis were 88%, 82%, and 86%, and area under the receiver operating characteristic curves (AUCs) were 0.92, 0.81, and 0.85. The AUC of hepatic–splenic-based radiomics analysis with the ensemble classifier was 7% larger than that of hepatic-based analysis (p < 0.05). The most important features selected by machine learning models included both hepatic and splenic features, and they were consistent with the location maps indicating the focus of deep learning when predicting liver fibrosis stage. Conclusions: Adding CT-based splenic radiomic features to hepatic radiomic features increases radiomics analysis performance for liver fibrosis staging. The most important features of the radiomics analysis were consistent with the information exploited by deep learning.

Artificial intelligence in liver diseases: improving diagnostics, prognostics and response prediction

Clinical routine in hepatology involves the diagnosis and treatment of a wide spectrum of metabolic, infectious, autoimmune and neoplastic diseases. Clinicians integrate qualitative and quantitative information from multiple data sources to make a diagnosis, prognosticate the disease course, and recommend a treatment. In the last 5 years, advances in artificial intelligence (AI), particularly in deep learning, have made it possible to extract clinically relevant information from complex and diverse clinical datasets. In particular, histopathology and radiology image data contain diagnostic, prognostic and predictive information which AI can extract. Ultimately, such AI systems could be implemented in clinical routine as decision support tools. However, in the context of hepatology, this requires further large-scale clinical validation and regulatory approval. Herein, we summarise the state of the art in AI in hepatology with a particular focus on histopathology and radiology data. We present a roadmap for the further development of novel biomarkers in hepatology and outline critical obstacles which need to be overcome.

Advances in liver US, CT, and MRI: moving toward the future

Over the past two decades, the epidemiology of chronic liver disease has changed with an increase in the prevalence of nonalcoholic fatty liver disease in parallel to the advent of curative treatments for hepatitis C. Recent developments provided new tools for diagnosis and monitoring of liver diseases based on ultrasound (US), computed tomography (CT), and magnetic resonance imaging (MRI), as applied for assessing steatosis, fibrosis, and focal lesions. This narrative review aims to discuss the emerging approaches for qualitative and quantitative liver imaging, focusing on those expected to become adopted in clinical practice in the next 5 to 10 years. While radiomics is an emerging tool for many of these applications, dedicated techniques have been investigated for US (controlled attenuation parameter, backscatter coefficient, elastography methods such as point shear wave elastography [pSWE] and transient elastography [TE], novel Doppler techniques, and three-dimensional contrast-enhanced ultrasound [3D-CEUS]), CT (dual-energy, spectral photon counting, extracellular volume fraction, perfusion, and surface nodularity), and MRI (proton density fat fraction [PDFF], elastography [MRE], contrast enhancement index, relative enhancement, T1 mapping on the hepatobiliary phase, perfusion). Concurrently, the advent of abbreviated MRI protocols will help fulfill an increasing number of examination requests in an era of healthcare resource constraints.

Deep Learning Radiomics to Predict Regional Lymph Node Staging for Hilar Cholangiocarcinoma

Background

Our aim was to establish a deep learning radiomics method to preoperatively evaluate regional lymph node (LN) staging for hilar cholangiocarcinoma (HC) patients.

Methods and Materials

Of the 179 enrolled HC patients, 90 were pathologically diagnosed with lymph node metastasis. Quantitative radiomic features and deep learning features were extracted. An LN metastasis status classifier was developed through integrating support vector machine, high-performance deep learning radiomics signature, and three clinical characteristics. An LN metastasis stratification classifier (N1 vs. N2) was also proposed with subgroup analysis.

Results

The average areas under the receiver operating characteristic curve (AUCs) of the LN metastasis status classifier reached 0.866 in the training cohort and 0.870 in the external test cohorts. Meanwhile, the LN metastasis stratification classifier performed well in predicting the risk of LN metastasis, with an average AUC of 0.946.

Conclusions

Two classifiers derived from computed tomography images performed well in predicting LN staging in HC and will be reliable evaluation tools to improve decision-making.

Deep Learning Prediction of Voxel-Level Liver Stiffness in Patients with Nonalcoholic Fatty Liver Disease

PURPOSE

To reconstruct virtual MR elastography (MRE) images based on traditional MRI inputs with a machine learning algorithm.

MATERIALS AND METHODS

In this single-institution, retrospective study, 149 patients (mean age, 58 years ± 12 [standard deviation]; 71 men) with nonalcoholic fatty liver disease who underwent MRI and MRE between January 2016 and January 2019 were evaluated. Nine conventional MRI sequences and clinical data were used to train a convolutional neural network to reconstruct MRE images at the per-voxel level. The architecture was further modified to accept multichannel three-dimensional inputs and to allow inclusion of clinical and demographic information. Liver stiffness and fibrosis category (F0 [no fibrosis] to F4 [significant fibrosis]) of reconstructed images were assessed by using voxel- and patient-level agreement by correlation, sensitivity, and specificity calculations; in addition, classification by receiver operator characteristic analyses was performed, and Dice score was used to evaluate hepatic stiffness locality.

RESULTS

The model for predicting liver stiffness incorporated four image sequences (precontrast T1-weighted liver acquisition with volume acquisition [LAVA] water and LAVA fat, 120-second-delay T1-weighted LAVA water, and single-shot fast spin-echo T2 weighted) and clinical data. The model had a patient-level and voxel-level correlation of 0.50 ± 0.05 and 0.34 ± 0.03, respectively. By using a stiffness threshold of 3.54 kPa to make a binary classification into no fibrosis or mild fibrosis (F0-F1) versus clinically significant fibrosis (F2-F4), the model had sensitivity of 80% ± 4, specificity of 75% ± 5, accuracy of 78% ± 3, area under the receiver operating characteristic curve of 84 ± 0.04, and a Dice score of 0.74.

CONCLUSION

The generation of virtual elastography images is feasible by using conventional MRI and clinical data with a machine learning algorithm.Keywords: MR Imaging, Abdomen/GI, Liver, Cirrhosis, Computer Applications/Virtual Imaging, Experimental Investigations, Feature Detection, Classification, Reconstruction Algorithms, Supervised Learning, Convolutional Neural Network (CNN) Supplemental material is available for this article. © RSNA, 2021.

Application of artificial intelligence in liver diseases: From diagnosis to treatment

Infectious or noninfectious liver disease has inexorably risen as one of the leading causes of global death and disease burden. There were an estimated 2.14 million liver-related deaths in 2017, representing an 11.4% increase since 2012. Traditional diagnosis and treatment methods have various dilemmas in different causes of liver disease. As a hot research topic in recent years, the application of artificial intelligence (AI) in different fields has attracted extensive attention, and new technologies have brought more ideas for the diagnosis and treatment of some liver diseases. Machine learning (ML) is the core of AI and the basic way to make a computer intelligent. ML technology has many potential uses in hepatology, ranging from exploring new noninvasive means to predict or diagnose different liver diseases to automated image analysis. The application of ML in liver diseases can help clinical staff to diagnose and treat different liver diseases quickly, accurately and scientifically, which is of importance for reducing the incidence and mortality of liver diseases, reducing medical errors, and promoting the development of medicine. This paper reviews the application and prospects of AI in liver diseases, and aims to improve clinicians’ awareness of the importance of AI in the diagnosis and treatment of liver diseases.

Artificial intelligence in gastroenterology: A state-of-the-art review

The development of artificial intelligence (AI) has increased dramatically in the last 20 years, with clinical applications progressively being explored for most of the medical specialties. The field of gastroenterology and hepatology, substantially reliant on vast amounts of imaging studies, is not an exception. The clinical applications of AI systems in this field include the identification of premalignant or malignant lesions (e.g., identification of dysplasia or esophageal adenocarcinoma in Barrett’s esophagus, pancreatic malignancies), detection of lesions (e.g., polyp identification and classification, small-bowel bleeding lesion on capsule endoscopy, pancreatic cystic lesions), development of objective scoring systems for risk stratification, predicting disease prognosis or treatment response [e.g., determining survival in patients post-resection of hepatocellular carcinoma), determining which patients with inflammatory bowel disease (IBD) will benefit from biologic therapy], or evaluation of metrics such as bowel preparation score or quality of endoscopic examination. The objective of this comprehensive review is to analyze the available AI-related studies pertaining to the entirety of the gastrointestinal tract, including the upper, middle and lower tracts; IBD; the hepatobiliary system; and the pancreas, discussing the findings and clinical applications, as well as outlining the current limitations and future directions in this field.

Clinical Artificial Intelligence Applications in Radiology: Chest and Abdomen

Organ segmentation, chest radiograph classification, and lung and liver nodule detections are some of the popular artificial intelligence (AI) tasks in chest and abdominal radiology due to the wide availability of public datasets. AI algorithms have achieved performance comparable to humans in less time for several organ segmentation tasks, and some lesion detection and classification tasks. This article introduces the current published articles of AI applied to chest and abdominal radiology, including organ segmentation, lesion detection, classification, and predicting prognosis.

Enhancement rate of venous phase to portal venous phase computed tomography and its correlation with ultrasound elastography determination of liver fibrosis

INTRODUCTION

This study aimed to compare the correlation between the computed tomography (CT) enhancement rate of the venous to portal venous phase (VP-ER) and the extracellular volume (ECV) fraction with shear-wave ultrasound elastography (USE) findings in patients with liver fibrosis.

METHODS

We included 450 patients with clinically suspected liver cirrhosis who underwent triphasic dynamic CT studies and USE. We compared the USE results with the unenhanced CT phase, with enhancement in the hepatic artery phase (HAP), portal venous phase (PVP), and venous phase (VP), and with the ECV fraction and the VP-ER. We also compared the area under the curve (AUC) of the receiver operating characteristic (ROC) curve of the ECV fraction and VP-ER with that of the values obtained with USE.

RESULTS

The VP-ER was the most highly correlated with the liver stiffness value determined with USE (Pearson's correlation coefficient: r = 0.37), followed by enhancement in the PVP (r = -0.25), CT number on unenhanced CT scans (r = -0.22), the ECV fraction (r = 0.19), enhancement in the VP (r = 0.059), and enhancement in the HAP (r = -0.023) (all p < 0.01). The VP-ER showed a significantly higher AUC than the ECV fraction (0.75 vs 0.62) when the liver stiffness was >15 kPa in USE studies (p = 0.04).

CONCLUSION

Compared to the ECV fraction, the VP-ER is more useful for predicting all degrees of liver fibrosis on routine triphasic dynamic CT images.

IMPLICATIONS FOR PRACTICE

Although improvement is needed, the VP-ER has a higher diagnostic ability for liver fibrosis than the ECV fraction in clinical practice.

Preoperative CT-Based Deep Learning Model for Predicting Risk Stratification in Patients With Gastrointestinal Stromal Tumors

Objective

To develop and evaluate a deep learning model (DLM) for predicting the risk stratification of gastrointestinal stromal tumors (GISTs).

Methods

Preoperative contrast-enhanced CT images of 733 patients with GISTs were retrospectively obtained from two centers between January 2011 and June 2020. The datasets were split into training (n = 241), testing (n = 104), and external validation cohorts (n = 388). A DLM for predicting the risk stratification of GISTs was developed using a convolutional neural network and evaluated in the testing and external validation cohorts. The performance of the DLM was compared with that of radiomics model by using the area under the receiver operating characteristic curves (AUROCs) and the Obuchowski index. The attention area of the DLM was visualized as a heatmap by gradient-weighted class activation mapping.

Results

In the testing cohort, the DLM had AUROCs of 0.90 (95% confidence interval [CI]: 0.84, 0.96), 0.80 (95% CI: 0.72, 0.88), and 0.89 (95% CI: 0.83, 0.95) for low-malignant, intermediate-malignant, and high-malignant GISTs, respectively. In the external validation cohort, the AUROCs of the DLM were 0.87 (95% CI: 0.83, 0.91), 0.64 (95% CI: 0.60, 0.68), and 0.85 (95% CI: 0.81, 0.89) for low-malignant, intermediate-malignant, and high-malignant GISTs, respectively. The DLM (Obuchowski index: training, 0.84; external validation, 0.79) outperformed the radiomics model (Obuchowski index: training, 0.77; external validation, 0.77) for predicting risk stratification of GISTs. The relevant subregions were successfully highlighted with attention heatmap on the CT images for further clinical review.

Conclusion

The DLM showed good performance for predicting the risk stratification of GISTs using CT images and achieved better performance than that of radiomics model.

Challenges and opportunities in the application of artificial intelligence in gastroenterology and hepatology

Artificial intelligence (AI) is an umbrella term used to describe a cluster of interrelated fields. Machine learning (ML) refers to a model that learns from past data to predict future data. Medicine and particularly gastroenterology and hepatology, are data-rich fields with extensive data repositories, and therefore fruitful ground for AI/ML-based software applications. In this study, we comprehensively review the current applications of AI/ML-based models in these fields and the opportunities that arise from their application. Specifically, we refer to the applications of AI/ML-based models in prevention, diagnosis, management, and prognosis of gastrointestinal bleeding, inflammatory bowel diseases, gastrointestinal premalignant and malignant lesions, other nonmalignant gastrointestinal lesions and diseases, hepatitis B and C infection, chronic liver diseases, hepatocellular carcinoma, cholangiocarcinoma, and primary sclerosing cholangitis. At the same time, we identify the major challenges that restrain the widespread use of these models in healthcare in an effort to explore ways to overcome them. Notably, we elaborate on the concerns regarding intrinsic biases, data protection, cybersecurity, intellectual property, liability, ethical challenges, and transparency. Even at a slower pace than anticipated, AI is infiltrating the healthcare industry. AI in healthcare will become a reality, and every physician will have to engage with it by necessity.

Detection of Incidental Esophageal Cancers on Chest CT by Deep Learning

Objective

To develop a deep learning-based model using esophageal thickness to detect esophageal cancer from unenhanced chest CT images.

Methods

We retrospectively identified 141 patients with esophageal cancer and 273 patients negative for esophageal cancer (at the time of imaging) for model training. Unenhanced chest CT images were collected and used to build a convolutional neural network (CNN) model for diagnosing esophageal cancer. The CNN is a VB-Net segmentation network that segments the esophagus and automatically quantifies the thickness of the esophageal wall and detect positions of esophageal lesions. To validate this model, 52 false negatives and 48 normal cases were collected further as the second dataset. The average performance of three radiologists and that of the same radiologists aided by the model were compared.

Results

The sensitivity and specificity of the esophageal cancer detection model were 88.8% and 90.9%, respectively, for the validation dataset set. Of the 52 missed esophageal cancer cases and the 48 normal cases, the sensitivity, specificity, and accuracy of the deep learning esophageal cancer detection model were 69%, 61%, and 65%, respectively. The independent results of the radiologists had a sensitivity of 25%, 31%, and 27%; specificity of 78%, 75%, and 75%; and accuracy of 53%, 54%, and 53%. With the aid of the model, the results of the radiologists were improved to a sensitivity of 77%, 81%, and 75%; specificity of 75%, 74%, and 74%; and accuracy of 76%, 77%, and 75%, respectively.

Conclusions

Deep learning-based model can effectively detect esophageal cancer in unenhanced chest CT scans to improve the incidental detection of esophageal cancer.

Deep learning for noninvasive liver fibrosis classification: A systematic review

BACKGROUND AND AIMS

While biopsy is the gold standard for liver fibrosis staging, it poses significant risks. Noninvasive assessment of liver fibrosis is a growing field. Recently, deep learning (DL) technology has revolutionized medical image analysis. This technology has the potential to enhance noninvasive fibrosis assessment. We systematically examined the application of DL in noninvasive liver fibrosis imaging.

METHODS

Embase, MEDLINE, Web of Science, and IEEE Xplore databases were used to identify studies that reported on the accuracy of DL for classification of liver fibrosis on noninvasive imaging. The search keywords were "liver or hepatic," "fibrosis or cirrhosis," and "neural or DL networks." Risk of bias and applicability were evaluated using the QUADAS-2 tool.

RESULTS

Sixteen studies were retrieved. Imaging modalities included ultrasound (n = 10), computed tomography (n = 3), and magnetic resonance imaging (n = 3). The studies analyzed a total of 40 405 radiological images from 15 853 patients. All but two of the studies were retrospective. In most studies the "ground truth" reference was the METAVIR score for pathological staging (n = 9.56%). The majority of the studies reported an accuracy >85% when compared to histopathology. Fourteen studies (87.5%) had a high risk of bias and concerns regarding applicability.

CONCLUSIONS

Deep learning has the potential to play an emerging role in liver fibrosis classification. Yet, it is still limited by a relatively small number of retrospective studies. Clinicians should facilitate the use of this technology by sharing databases and standardized reports. This may optimize the noninvasive evaluation of liver fibrosis on a large scale.

Automatic discrimination of different sequences and phases of liver MRI using a dense feature fusion neural network: a preliminary study

PURPOSE

To develop and validate a dense feature fusion neural network (DFuNN) to automatically recognize different sequences and phases of liver magnetic resonance imaging (MRI).

MATERIALS AND METHODS

In total, 3869 sequences and phases from 384 liver MRI examinations, divided into training/validation (n = 2886 sequences from 287 patients) and test (n = 983 sequences from 97 patients) sets, were used in this retrospective study. Ten unenhanced sequences and enhanced phases were included. Manual sequence recognition, performed by two radiologists (20 and 10 years of experience) in a consensus reading, was used as the reference standard. The sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) were calculated to evaluate the performance of the DFuNN on an identical unseen test set. Finally, we evaluated the factors impacting the model precision.

RESULTS

A fusion block improved the performance of the DFuNN. DFuNN with a fusion block achieved good recognition performance for both complete and incomplete sequences and phases in the test set. The average sensitivity of recognition performance for complete sequence and phase inputs ranged from 88.06 to 100%, the average specificity ranged from 99.12 to 99.94%, and the median accuracy ranged from 98.02 to 99.95%. The DFuNN prediction accuracy for patients without cirrhosis were significantly higher than those for patients with cirrhosis (P = 0.0153). No significant difference was found in the accuracy across other factors.

CONCLUSION

DFuNN can automatically and accurately identify specific unenhanced MRI sequences and enhanced MRI phases.

Preliminary Study on the Diagnostic Performance of a Deep Learning System for Submandibular Gland Inflammation Using Ultrasonography Images

This study was performed to evaluate the diagnostic performance of deep learning systems using ultrasonography (USG) images of the submandibular glands (SMGs) in three different conditions: obstructive sialoadenitis, Sjögren’s syndrome (SjS), and normal glands. Fifty USG images with a confirmed diagnosis of obstructive sialoadenitis, 50 USG images with a confirmed diagnosis of SjS, and 50 USG images with no SMG abnormalities were included in the study. The training group comprised 40 obstructive sialoadenitis images, 40 SjS images, and 40 control images, and the test group comprised 10 obstructive sialoadenitis images, 10 SjS images, and 10 control images for deep learning analysis. The performance of the deep learning system was calculated and compared between two experienced radiologists. The sensitivity of the deep learning system in the obstructive sialoadenitis group, SjS group, and control group was 55.0%, 83.0%, and 73.0%, respectively, and the total accuracy was 70.3%. The sensitivity of the two radiologists was 64.0%, 72.0%, and 86.0%, respectively, and the total accuracy was 74.0%. This study revealed that the deep learning system was more sensitive than experienced radiologists in diagnosing SjS in USG images of two case groups and a group of healthy subjects in inflammation of SMGs.

Artificial Intelligence in Imaging of Chronic Liver Diseases: Current Update and Future Perspectives

Here we review artificial intelligence (AI) models which aim to assess various aspects of chronic liver disease. Despite the clinical importance of hepatocellular carcinoma in the setting of chronic liver disease, we focus this review on AI models which are not lesion-specific and instead review models developed for liver parenchyma segmentation, evaluation of portal circulation, assessment of hepatic fibrosis, and identification of hepatic steatosis. Optimization of these models offers the opportunity to potentially reduce the need for invasive procedures such as catheterization to measure hepatic venous pressure gradient or biopsy to assess fibrosis and steatosis. We compare the performance of these AI models amongst themselves as well as to radiomics approaches and alternate modality assessments. We conclude that these models show promising performance and merit larger-scale evaluation. We review artificial intelligence models that aim to assess various aspects of chronic liver disease aside from hepatocellular carcinoma. We focus this review on models for liver parenchyma segmentation, evaluation of portal circulation, assessment of hepatic fibrosis, and identification of hepatic steatosis. We conclude that these models show promising performance and merit a larger scale evaluation.

Computed Tomography Techniques, Protocols, Advancements, and Future Directions in Liver Diseases

Computed tomography (CT) is often performed as the initial imaging study for the workup of patients with known or suspected liver disease. Our article reviews liver CT techniques and protocols in clinical practice along with updates on relevant CT advances, including wide-detector CT, radiation dose optimization, and multienergy scanning, that have already shown clinical impact. Particular emphasis is placed on optimizing the late arterial phase of enhancement, which is critical to evaluation of hepatocellular carcinoma. We also discuss emerging techniques that may soon influence clinical care.

Training of computational algorithms to predict NAFLD activity score and fibrosis stage from liver histopathology slides

BACKGROUND

The incidence of non-alcoholic fatty liver disease (NAFLD) and its progressive form, non-alcoholic steatohepatitis (NASH), has been increasing for decades. Since the mainstay is lifestyle modification in this mainly asymptomatic condition, there is a need for accurate diagnostic methods.

OBJECTIVES

This study proposes a method with a computer-aided diagnosis (CAD) system to predict NAFLD Activity score (NAS scores-steatosis, lobular inflammation, and ballooning) and fibrosis stage from histopathology slides.

METHODS

A total of 87 pathology slides pairs (H&E and Trichrome-stained) were used for the study. Ground-truth NAS scores and fibrosis stages were previously identified by a pathologist. Each slide was split into 224 × 224 patches and fed into a feature extraction network to generate local features. These local features were processed and aggregated to obtain a global feature to predict the slide's scores. The effects of different training strategies, as well as training data with different staining and magnifications were explored. Four-fold cross validation was performed due to the small data size. Area Under the Receiver Operating Curve (AUROC) was utilized to evaluate the prediction performance of the machine-learning algorithm.

RESULTS

Predictive accuracy for all subscores was high in comparison with pathologist assessment. There was no difference among the 3 magnifications (5x, 10x, 20x) for NAS-steatosis and fibrosis stage tasks. A larger magnification (20x) achieved better performance for NAS-lobular scores. Middle-level magnification was best for NAS-ballooning task. Trichrome slides are better for fibrosis stage prediction and NAS-ballooning score prediction task. NAS-steatosis prediction had the best performance (AUC 90.48%) in the model. A good performance was observed with fibrosis stage prediction (AUC 83.85%) as well as NAS-ballooning prediction (AUC 81.06%).

CONCLUSIONS

These results were robust. The method proposed proved to be effective in predicting NAFLD Activity score and fibrosis stage from histopathology slides. The algorithms are an aid in having an accurate and systematic diagnosis in a condition that affects hundreds of millions of patients globally.

Predicting the stages of liver fibrosis with multiphase CT radiomics based on volumetric features

PURPOSES

To develop and externally validate a multiphase computed tomography (CT)-based machine learning (ML) model for staging liver fibrosis (LF) by using whole liver slices.

MATERIALS AND METHODS

The development dataset comprised 232 patients with pathological analysis for LF, and the test dataset comprised 100 patients from an independent outside institution. Feature extraction was performed based on the precontrast (PCP), arterial (AP), portal vein (PVP) phase, and three-phase CT images. CatBoost was utilized for ML model investigation by using the features with good reproducibility. The diagnostic performance of ML models based on each single- and three-phase CT image was compared with that of radiologists' interpretations, the aminotransferase-to-platelet ratio index, and the fibrosis index based on four factors (FIB-4) by using the receiver operating characteristic curve with the area under the curve (AUC) value.

RESULTS

Although the ML model based on three-phase CT image (AUC = 0.65-0.80) achieved higher AUC value than that based on PCP (AUC = 0.56-0.69) and PVP (AUC = 0.51-0.74) in predicting various stage of LF, significant difference was not found. The best CT-based ML model (AUC = 0.65-0.80) outperformed the FIB-4 in differentiating advanced LF and cirrhosis and radiologists' interpretation (AUC = 0.50-0.76) in the diagnosis of significant and advanced LF.

CONCLUSION

All PCP, PVP, and three-phase CT-based ML models can be an acceptable in assessing LF, and the performance of the PCP-based ML model is comparable to that of the enhanced CT image-based ML model.

Current Status of Radiomics and Deep Learning in Liver Imaging

ABSTRACT

Artificial intelligence is poised to revolutionize medical image. It takes advantage of the high-dimensional quantitative features present in medical images that may not be fully appreciated by humans. Artificial intelligence has the potential to facilitate automatic organ segmentation, disease detection and characterization, and prediction of disease recurrence. This article reviews the current status of artificial intelligence in liver imaging and reviews the opportunities and challenges in clinical implementation.

The overview of the deep learning integrated into the medical imaging of liver: a review

Deep learning (DL) is a recently developed artificial intelligent method that can be integrated into numerous fields. For the imaging diagnosis of liver disease, several remarkable outcomes have been achieved with the application of DL currently. This advanced algorithm takes part in various sections of imaging processing such as liver segmentation, lesion delineation, disease classification, process optimization, etc. The DL optimized imaging diagnosis shows a broad prospect instead of the pathological biopsy for the advantages of convenience, safety, and inexpensiveness. In this paper, we reviewed the published representative DL-related hepatic imaging works, described the general situation of this new-rising technology in medical liver imaging and explored the future direction of DL development.

Transforming obstetric ultrasound into data science using eye tracking, voice recording, transducer motion and ultrasound video

Ultrasound is the primary modality for obstetric imaging and is highly sonographer dependent. Long training period, insufficient recruitment and poor retention of sonographers are among the global challenges in the expansion of ultrasound use. For the past several decades, technical advancements in clinical obstetric ultrasound scanning have largely concerned improving image quality and processing speed. By contrast, sonographers have been acquiring ultrasound images in a similar fashion for several decades. The PULSE (Perception Ultrasound by Learning Sonographer Experience) project is an interdisciplinary multi-modal imaging study aiming to offer clinical sonography insights and transform the process of obstetric ultrasound acquisition and image analysis by applying deep learning to large-scale multi-modal clinical data. A key novelty of the study is that we record full-length ultrasound video with concurrent tracking of the sonographer's eyes, voice and the transducer while performing routine obstetric scans on pregnant women. We provide a detailed description of the novel acquisition system and illustrate how our data can be used to describe clinical ultrasound. Being able to measure different sonographer actions or model tasks will lead to a better understanding of several topics including how to effectively train new sonographers, monitor the learning progress, and enhance the scanning workflow of experts.

Application of Artificial Intelligence for the Diagnosis and Treatment of Liver Diseases

Modern medical care produces large volumes of multimodal patient data, which many clinicians struggle to process and synthesize into actionable knowledge. In recent years, artificial intelligence (AI) has emerged as an effective tool in this regard. The field of hepatology is no exception, with a growing number of studies published that apply AI techniques to the diagnosis and treatment of liver diseases. These have included machine-learning algorithms (such as regression models, Bayesian networks, and support vector machines) to predict disease progression, the presence of complications, and mortality; deep-learning algorithms to enable rapid, automated interpretation of radiologic and pathologic images; and natural-language processing to extract clinically meaningful concepts from vast quantities of unstructured data in electronic health records. This review article will provide a comprehensive overview of hepatology-focused AI research, discuss some of the barriers to clinical implementation and adoption, and suggest future directions for the field.

A radiomics-based model to classify the etiology of liver cirrhosis using gadoxetic acid-enhanced MRI

The implementation of radiomics in radiology is gaining interest due to its wide range of applications. To develop a radiomics-based model for classifying the etiology of liver cirrhosis using gadoxetic acid-enhanced MRI, 248 patients with a known etiology of liver cirrhosis who underwent 306 gadoxetic acid-enhanced MRI examinations were included in the analysis. MRI examinations were classified into 6 groups according to the etiology of liver cirrhosis: alcoholic cirrhosis, viral hepatitis, cholestatic liver disease, nonalcoholic steatohepatitis (NASH), autoimmune hepatitis, and other. MRI examinations were randomized into training and testing subsets. Radiomics features were extracted from regions of interest segmented in the hepatobiliary phase images. The fivefold cross-validated models (2-dimensional-(2D) and 3-dimensional-(3D) based) differentiating cholestatic cirrhosis from noncholestatic etiologies had the best accuracy (87.5%, 85.6%), sensitivity (97.6%, 95.6%), predictive value (0.883, 0.877), and area under curve (AUC) (0.960, 0.910). The AUC was larger in the 2D-model for viral hepatitis, cholestatic cirrhosis, and NASH-associated cirrhosis (P-value of 0.05, 0.05, 0.87, respectively). In alcoholic cirrhosis, the AUC for the 3D model was larger (P = 0.01). The overall intra-class correlation coefficient (ICC) estimates and their 95% confident intervals (CI) for all features combined was 0.68 (CI 0.56-0.87) for 2D and 0.71 (CI 0.61-0.93) for 3D measurements suggesting moderate reliability. Radiomics-based analysis of hepatobiliary phase images of gadoxetic acid-enhanced MRI may be a promising noninvasive method for identifying the etiology of liver cirrhosis with better performance of the 2D- compared with the 3D-generated models.

Liver fibrosis staging by deep learning: a visual-based explanation of diagnostic decisions of the model

Objectives

Deep learning has been proven to be able to stage liver fibrosis based on contrast-enhanced CT images. However, until now, the algorithm is used as a black box and lacks transparency. This study aimed to provide a visual-based explanation of the diagnostic decisions made by deep learning.

Methods

The liver fibrosis staging network (LFS network) was developed at contrast-enhanced CT images in the portal venous phase in 252 patients with histologically proven liver fibrosis stage. To give a visual explanation of the diagnostic decisions made by the LFS network, Gradient-weighted Class Activation Mapping (Grad-cam) was used to produce location maps indicating where the LFS network focuses on when predicting liver fibrosis stage.

Results

The LFS network had areas under the receiver operating characteristic curve of 0.92, 0.89, and 0.88 for staging significant fibrosis (F2–F4), advanced fibrosis (F3–F4), and cirrhosis (F4), respectively, on the test set. The location maps indicated that the LFS network had more focus on the liver surface in patients without liver fibrosis (F0), while it focused more on the parenchyma of the liver and spleen in case of cirrhosis (F4).

Conclusions

Deep learning methods are able to exploit CT-based information from the liver surface, liver parenchyma, and extrahepatic information to predict liver fibrosis stage. Therefore, we suggest using the entire upper abdomen on CT images when developing deep learning–based liver fibrosis staging algorithms.

Key Points

• Deep learning algorithms can stage liver fibrosis using contrast-enhanced CT images, but the algorithm is still used as a black box and lacks transparency.

• Location maps produced by Gradient-weighted Class Activation Mapping can indicate the focus of the liver fibrosis staging network.

• Deep learning methods use CT-based information from the liver surface, liver parenchyma, and extrahepatic information to predict liver fibrosis stage.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00330-021-08046-x.

Advances in Deep Learning-Based Medical Image Analysis

Importance. With the booming growth of artificial intelligence (AI), especially the recent advancements of deep learning, utilizing advanced deep learning-based methods for medical image analysis has become an active research area both in medical industry and academia. This paper reviewed the recent progress of deep learning research in medical image analysis and clinical applications. It also discussed the existing problems in the field and provided possible solutions and future directions.Highlights. This paper reviewed the advancement of convolutional neural network-based techniques in clinical applications. More specifically, state-of-the-art clinical applications include four major human body systems: the nervous system, the cardiovascular system, the digestive system, and the skeletal system. Overall, according to the best available evidence, deep learning models performed well in medical image analysis, but what cannot be ignored are the algorithms derived from small-scale medical datasets impeding the clinical applicability. Future direction could include federated learning, benchmark dataset collection, and utilizing domain subject knowledge as priors.Conclusion. Recent advanced deep learning technologies have achieved great success in medical image analysis with high accuracy, efficiency, stability, and scalability. Technological advancements that can alleviate the high demands on high-quality large-scale datasets could be one of the future developments in this area.

Artificial intelligence in gastroenterology and hepatology: Status and challenges

Originally proposed by John McCarthy in 1955, artificial intelligence (AI) has achieved a breakthrough and revolutionized the processing methods of clinical medicine with the increasing workloads of medical records and digital images. Doctors are paying attention to AI technologies for various diseases in the fields of gastroenterology and hepatology. This review will illustrate AI technology procedures for medical image analysis, including data processing, model establishment, and model validation. Furthermore, we will summarize AI applications in endoscopy, radiology, and pathology, such as detecting and evaluating lesions, facilitating treatment, and predicting treatment response and prognosis with excellent model performance. The current challenges for AI in clinical application include potential inherent bias in retrospective studies that requires larger samples for validation, ethics and legal concerns, and the incomprehensibility of the output results. Therefore, doctors and researchers should cooperate to address the current challenges and carry out further investigations to develop more accurate AI tools for improved clinical applications.

Quantitative Functional Evaluation of Liver Fibrosis in Mice with Dynamic Contrast-enhanced Photoacoustic Imaging

Background Dynamic contrast-enhanced (DCE) photoacoustic (PA) imaging (PAI) is a novel noninvasive imaging modality that uses the differences in optical absorption of oxyhemoglobin and deoxyhemoglobin and may be performed in a dynamic fashion to image the indocyanine green (ICG) pharmacokinetics in the liver. Purpose To determine whether DCE PAI parameters (maximum peak time [T_max] and half-life [T_1/2]) derived from the PA liver function curve correlate with fibrosis determined using histopathologic analysis. Materials and Methods Between June 2020 and October 2020, 28 male mice aged 8 weeks were intraperitoneally injected with carbon tetrachloride solution every 2 days to establish a liver fibrosis model. At the 1st, 4th, and 8th week of modeling, the changes in liver structure were monitored dynamically by using a PA technique. The T_max and T_1/2 of ICG were calculated at different pathologic stages and within a control group. Corresponding liver histopathologic results and blood biochemical data were obtained. Spearman rank correlation was used to evaluate the relationship between the DCE PAI results and histologic scores. Results The PA liver function curve showed that the T_max and T_1/2 varied among groups (mean T_max: control group, 9 seconds ± 1.8 [standard deviation]; 1 week, 51 seconds ± 4.4; 4 weeks, 73 seconds ± 5.3; 8 weeks, 104 seconds ± 6.6; P < .001) (mean T_1/2: control group, 28 seconds ± 6.5; 1 week, 201 seconds ± 12.4; 4 weeks, 285 seconds ± 11; 8 weeks, 318 seconds ± 30.5; P < .001). There was a positive correlation between the dynamic parameters (T_max and T_1/2) and the histopathologic scores; the Spearman ρ ratios for the Sirius red and α-smooth muscle actin (αSMA)-positive areas versus the T_max were 0.93 and 0.94 (P < .001 for both), and the Spearman ρ ratios for the Sirius red and αSMA-positive areas versus T_1/2 were 0.87 and 0.89 (P < .001 for both). Conclusion Dynamic contrast-enhanced photoacoustic imaging demonstrated a higher maximum peak time and half-life in mice with induced fibrosis compared with control mice without fibrosis, and these values correlated with histologic measures of fibrosis. © RSNA, 2021 Online supplemental material is available for this article.

Comparing radiomics models with different inputs for accurate diagnosis of significant fibrosis in chronic liver disease

OBJECTIVE

The non-invasive discrimination of significant fibrosis (≥ F2) in patients with chronic liver disease (CLD) is clinically critical but technically challenging. We aimed to develop an updated deep learning radiomics model of elastography (DLRE2.0) based on our previous DLRE model to achieve significantly improved performance in ≥ F2 evaluation.

METHODS

This was a retrospective multicenter study with 807 CLD patients and 4842 images from three hospitals. All of these patients have liver biopsy results as referenced standard. Multichannel deep learning radiomics models were developed. Elastography images, gray-scale images of the liver capsule, gray-scale images of the liver parenchyma, and serological results were gradually integrated to establish different diagnosis models, and the optimal model was selected for assessing ≥ F2. Its accuracy was thoroughly investigated by applying different F0-1 prevalence cohorts and independent external test cohorts. Analysis of receiver operating characteristic (ROC) curves was performed to calculate the area under the ROC curve (AUC) for significance of fibrosis (≥ F2) and cirrhosis (F4).

RESULTS

The AUC of the DLRE2.0 model significantly increased to 0.91 compared with the DLRE model (AUC 0.83) when evaluating ≥ F2 (p = 0.0167). However, it did not show statistically significant differences as integrating gray-scale images and serological data into the DLRE2.0 model. AUCs of DLRE and DLRE2.0 increased, when there was higher F0-1 prevalence. All radiomics models had good robustness in the independent external test cohort.

CONCLUSIONS

DLRE2.0 was the most suitable model for staging significant fibrosis while considering the balance of diagnostic accuracy and clinical practicability.

KEY POINTS

• The non-invasive discrimination of significant fibrosis (≥ F2) in patients with chronic liver disease (CLD) is clinically critical but technically challenging. • We aimed to develop an updated deep learning radiomics model of elastography (DLRE2.0) based on our previous DLRE model to achieve significantly improved performance in ≥ F2 evaluation. • Our study based on 807 CLD patients and 4842 images with liver biopsy found that DLRE2.0 was the most suitable model for staging significant fibrosis while considering the balance of diagnostic accuracy and clinical practicability.

Feasibility of automatic detection of small hepatocellular carcinoma (≤2 cm) in cirrhotic liver based on pattern matching and deep learning

Background and objective.Early detection of hepatocellular carcinoma (HCC) is crucial for clinical management. Current studies have reported large HCC detections using automatic algorithms, but there is a lack of research on automatic detection of small HCCs (sHCCs). This study is to investigate the feasibility of automatic detection of sHCC (≤2 cm) based on pattern matching and deep learning (PM-DL) model.Materials and methods. A retrospective study included 5376 image sets from 56 cirrhosis patients (28 sHCC patients with 32 pathologically confirmed lesions and 28 non-HCC cirrhosis patients) in the training-validation cohort to build and validate the model through five-fold cross-validation. In addition, an external test cohort including 6144 image sets from 64 cirrhosis patients (32 sHCC patients with 38 lesions and 32 non-HCC cirrhosis patients) was applied to further verify the generalization ability of the model. The proposed PM-DL model consisted of three main steps: 3D co-registration and liver segmentation, screening of suspicious lesions on diffusion-weighted imaging images based on pattern matching algorithm, and identification/segmentation of sHCC lesions on dynamic contrast-enhanced images with convolutional neural network.Results.The PM-DL model achieved a sensitivity of 89.74% and a positive predictive value of 85.00% in the external test cohort for per-lesion analysis. No significant difference was observed in volumes (P= 0.13) and the largest sizes (P= 0.89) between manually delineated and segmented lesions. The DICE coefficient reached 0.77 ± 0.16. Similar performances were identified in the validation cohort. Moreover, the PM-DL model outperformed Liver Imaging Reporting and Data System (LI-RADS) in sensitivity (probable HCCs: LR-5 or LR-4,P= 0.18; definite HCCs: LR-5,P< 0.001), with a similar high specificity for per-patient analysis.Conclusion. The PM-DL model may be feasible for accurate automatic detection of sHCC in cirrhotic liver.

Hepatic Elastometry and Glissonian Line in the Assessment of Liver Fibrosis

The aim of this study was to identify a method for staging hepatic fibrosis using a non-invasive, rapid and inexpensive technique based on ultrasound morphologic hepatic features. A total of 215 patients with different liver diseases underwent B-mode (2-D brightness mode) ultrasonography, vibration-controlled transient elastography, 2-D shear wave elastography and measurement of the controlled attenuation parameter with transient elastography. B-Mode images of the anterior margin of the left lobe were obtained and processed with automatic Genoa Line Quantification (GLQ) software based on a neural network for staging liver fibrosis. The accuracy of GLQ was 90.6% during model training and 78.9% in 38 different patients with concordant elastometric measures. Receiver operating characteristic curve analysis of GLQ performance using vibration-controlled transient elastography as a reference yielded areas under the curves of 0.851 for F ≥ F1, 0.793 for F ≥ F2, 0.784 for F ≥ F3 and 0.789 for F ≥ F4. GLQ has the potential to be a rapid, easy-to-perform and tolerable method in the staging of liver fibrosis.

An imageomics and multi-network based deep learning model for risk assessment of liver transplantation for hepatocellular cancer

INTRODUCTION

Liver transplantation (LT) is an effective treatment for hepatocellular carcinoma (HCC), the most common type of primary liver cancer. Patients with small HCC (<5 cm) are given priority over others for transplantation due to clinical allocation policies based on tumor size. Attempting to shift from the prevalent paradigm that successful transplantation and longer disease-free survival can only be achieved in patients with small HCC to expanding the transplantation option to patients with HCC of the highest tumor burden (>5 cm), we developed a convergent artificial intelligence (AI) model that combines transient clinical data with quantitative histologic and radiomic features for more objective risk assessment of liver transplantation for HCC patients.

METHODS

Patients who received a LT for HCC between 2008-2019 were eligible for inclusion in the analysis. All patients with post-LT recurrence were included, and those without recurrence were randomly selected for inclusion in the deep learning model. Pre- and post-transplant magnetic resonance imaging (MRI) scans and reports were compressed using CapsNet networks and natural language processing, respectively, as input for a multiple feature radial basis function network. We applied a histological image analysis algorithm to detect pathologic areas of interest from explant tissue of patients who recurred. The multilayer perceptron was designed as a feed-forward, supervised neural network topology, with the final assessment of recurrence risk. We used area under the curve (AUC) and F-1 score to assess the predictability of different network combinations.

RESULTS

A total of 109 patients were included (87 in the training group, 22 in the testing group), of which 20 were positive for cancer recurrence. Seven models (AUC; F-1 score) were generated, including clinical features only (0.55; 0.52), magnetic resonance imaging (MRI) only (0.64; 0.61), pathological images only (0.64; 0.61), MRI plus pathology (0.68; 0.65), MRI plus clinical (0.78, 0.75), pathology plus clinical (0.77; 0.73), and a combination of clinical, MRI, and pathology features (0.87; 0.84). The final combined model showed 80 % recall and 89 % precision. The total accuracy of the implemented model was 82 %.

CONCLUSION

We validated that the deep learning model combining clinical features and multi-scale histopathologic and radiomic image features can be used to discover risk factors for recurrence beyond tumor size and biomarker analysis. Such a predictive, convergent AI model has the potential to alter the LT allocation system for HCC patients and expand the transplantation treatment option to patients with HCC of the highest tumor burden.

MRI-based radiomic feature analysis of end-stage liver disease for severity stratification

Purpose

We aimed to develop a predictive model of disease severity for cirrhosis using MRI-derived radiomic features of the liver and spleen and compared it to the existing disease severity metrics of MELD score and clinical decompensation. The MELD score is compiled solely by blood parameters, and so far, it was not investigated if extracted image-based features have the potential to reflect severity to potentially complement the calculated score.

Methods

This was a retrospective study of eligible patients with cirrhosis (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n=90$$\end{document}n=90) who underwent a contrast-enhanced MR screening protocol for hepatocellular carcinoma (HCC) screening at a tertiary academic center from 2015 to 2018. Radiomic feature analyses were used to train four prediction models for assessing the patient’s condition at time of scan: MELD score, MELD score \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\ge $$\end{document}≥ 9 (median score of the cohort), MELD score \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\ge $$\end{document}≥ 15 (the inflection between the risk and benefit of transplant), and clinical decompensation. Liver and spleen segmentations were used for feature extraction, followed by cross-validated random forest classification.

Results

Radiomic features of the liver and spleen were most predictive of clinical decompensation (AUC 0.84), which the MELD score could predict with an AUC of 0.78. Using liver or spleen features alone had slightly lower discrimination ability (AUC of 0.82 for liver and AUC of 0.78 for spleen features only), although this was not statistically significant on our cohort. When radiomic prediction models were trained to predict continuous MELD scores, there was poor correlation. When stratifying risk by splitting our cohort at the median MELD 9 or at MELD 15, our models achieved AUCs of 0.78 or 0.66, respectively.

Conclusions

We demonstrated that MRI-based radiomic features of the liver and spleen have the potential to predict the severity of liver cirrhosis, using decompensation or MELD status as imperfect surrogate measures for disease severity.

Radiomics and deep learning in liver diseases

Recently, radiomics and deep learning have gained attention as methods for computerized image analysis. Radiomics and deep learning can perform diagnostic or predictive tasks using high-dimensional image-derived features and have the potential to expand the capabilities of liver imaging beyond the scope of traditional visual image analysis. Recent research has demonstrated the potential of these techniques in various fields of liver imaging, including staging of liver fibrosis, prognostication of malignant liver tumors, automated detection and characterization of liver tumors, automated abdominal organ segmentation, and body composition analysis. However, because most of the previous studies were preliminary and focused mainly on technical feasibility, further clinical validation is required for the application of radiomics and deep learning in clinical practice. In this review, we introduce the technical aspects of radiomics and deep learning and summarize the recent studies on the application of these techniques in liver radiology.

Artificial intelligence in prediction of non-alcoholic fatty liver disease and fibrosis

Artificial intelligence (AI) has become increasingly widespread in our daily lives, including healthcare applications. AI has brought many new insights into better ways we care for our patients with chronic liver disease, including non-alcoholic fatty liver disease and liver fibrosis. There are multiple ways to apply the AI technology on top of the conventional invasive (liver biopsy) and noninvasive (transient elastography, serum biomarkers, or clinical prediction models) approaches. In this review article, we discuss the principles of applying AI on electronic health records, liver biopsy, and liver images. A few common AI approaches include logistic regression, decision tree, random forest, and XGBoost for data at a single time stamp, recurrent neural networks for sequential data, and deep neural networks for histology and images.

Imaging-Based Staging of Hepatic Fibrosis in Patients with Hepatitis B: A Dynamic Radiomics Model Based on Gd-EOB-DTPA-Enhanced MRI

Accurate grading of liver fibrosis can effectively assess the severity of liver disease and help doctors make an appropriate diagnosis. This study aimed to perform the automatic staging of hepatic fibrosis on patients with hepatitis B, who underwent gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)-enhanced magnetic resonance imaging with dynamic radiomics analysis. The proposed dynamic radiomics model combined imaging features from multi-phase dynamic contrast-enhanced (DCE) images and time-domain information. Imaging features were extracted from the deep learning-based segmented liver volume, and time-domain features were further explored to analyze the variation in features during contrast enhancement. Model construction and evaluation were based on a 132-case data set. The proposed model achieved remarkable performance in significant fibrosis (fibrosis stage S1 vs. S2–S4; accuracy (ACC) = 0.875, area under the curve (AUC) = 0.867), advanced fibrosis (S1–S2 vs. S3–S4; ACC = 0.825, AUC = 0.874), and cirrhosis (S1–S3 vs. S4; ACC = 0.850, AUC = 0.900) classifications in the test set. It was more dominant compared with the conventional single-phase or multi-phase DCE-based radiomics models, normalized liver enhancement, and some serological indicators. Time-domain features were found to play an important role in the classification models. The dynamic radiomics model can be applied for highly accurate automatic hepatic fibrosis staging.

Utility of Multiparametric CT for Identification of High-Risk NAFLD

OBJECTIVE. The purpose of this study was to evaluate the utility of laboratory and CT metrics in identifying patients with high-risk nonalcoholic fatty liver disease (NAFLD). MATERIALS AND METHODS. Patients with biopsy-proven NAFLD who underwent CT within 1 year of biopsy were included. Histopathologic review was performed by an experienced gastrointestinal pathologist to determine steatosis, inflammation, and fibrosis. The presence of any lobular inflammation and hepatocyte ballooning was categorized as nonalcoholic steatohepatitis (NASH). Patients with NAFLD and advanced fibrosis (stage F3 or higher) were categorized as having high-risk NAFLD. Aspartate transaminase to platelet ratio index and Fibrosis-4 (FIB-4) laboratory scores were calculated. CT metrics included hepatic attenuation, liver segmental volume ratio (LSVR), splenic volume, liver surface nodularity score, and selected texture features. In addition, two readers subjectively assessed the presence of NASH (present or not present) and fibrosis (stages F0-F4). RESULTS. A total of 186 patients with NAFLD (mean age, 49 years; 74 men and 112 women) were included. Of these, 87 (47%) had NASH and 112 (60%) had moderate to severe steatosis. A total of 51 patients were classified as fibrosis stage F0, 42 as F1, 23 as F2, 37 as F3, and 33 as F4. Additionally, 70 (38%) had advanced fibrosis (stage F3 or F4) and were considered to have high-risk NAFLD. FIB-4 score correlated with fibrosis (ROC AUC of 0.75 for identifying high-risk NAFLD). Of the individual CT parameters, LSVR and splenic volume performed best (AUC of 0.69 for both for detecting high-risk NAFLD). Subjective reader assessment performed best among all parameters (AUCs of 0.78 for reader 1 and 0.79 for reader 2 for detecting high-risk NAFLD). FIB-4 and subjective scores were complementary (combined AUC of 0.82 for detecting high-risk NAFLD). For NASH assessment, FIB-4 performed best (AUC of 0.68), whereas the AUCs were less than 0.60 for all individual CT features and subjective assessments. CONCLUSION. FIB-4 and multiple CT findings can identify patients with high-risk NAFLD (advanced fibrosis or cirrhosis). However, the presence of NASH is elusive on CT.

Application of artificial intelligence in chronic liver diseases: a systematic review and meta-analysis

Background

The gold standard for the diagnosis of liver fibrosis and nonalcoholic fatty liver disease (NAFLD) is liver biopsy. Various noninvasive modalities, e.g., ultrasonography, elastography and clinical predictive scores, have been used as alternatives to liver biopsy, with limited performance. Recently, artificial intelligence (AI) models have been developed and integrated into noninvasive diagnostic tools to improve their performance.

Methods

We systematically searched for studies on AI-assisted diagnosis of liver fibrosis and NAFLD on MEDLINE, Scopus, Web of Science and Google Scholar. The pooled sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and diagnostic odds ratio (DOR) with their 95% confidence intervals (95% CIs) were calculated using a random effects model. A summary receiver operating characteristic curve and the area under the curve was generated to determine the diagnostic accuracy of the AI-assisted system. Subgroup analyses by diagnostic modalities, population and AI classifiers were performed.

Results

We included 19 studies reporting the performances of AI-assisted ultrasonography, elastrography, computed tomography, magnetic resonance imaging and clinical parameters for the diagnosis of liver fibrosis and steatosis. For the diagnosis of liver fibrosis, the pooled sensitivity, specificity, PPV, NPV and DOR were 0.78 (0.71–0.85), 0.89 (0.81–0.94), 0.72 (0.58–0.83), 0.92 (0.88–0.94) and 31.58 (11.84–84.25), respectively, for cirrhosis; 0.86 (0.80–0.90), 0.87 (0.80–0.92), 0.85 (0.75–0.91), 0.88 (0.82–0.92) and 37.79 (16.01–89.19), respectively; for advanced fibrosis; and 0.86 (0.78–0.92), 0.81 (0.77–0.84), 0.88 (0.80–0.93), 0.77 (0.58–0.89) and 26.79 (14.47–49.62), respectively, for significant fibrosis. Subgroup analyses showed significant differences in performance for the diagnosis of fibrosis among different modalities. The pooled sensitivity, specificity, PPV, NPV and DOR were 0.97 (0.76–1.00), 0.91 (0.78–0.97), 0.95 (0.87–0.98), 0.93 (0.80–0.98) and 191.52 (38.82–944.81), respectively, for the diagnosis of liver steatosis.

Conclusions

AI-assisted systems have promising potential for the diagnosis of liver fibrosis and NAFLD. Validations of their performances are warranted before implementing these AI-assisted systems in clinical practice.

Trial registration: The protocol was registered with PROSPERO (CRD42020183295).

Evaluation of Diagnostic Tests

As technology advances, diagnostic tests continue to improve and each year, we are presented with new alternatives to standard procedures. Given the plethora of diagnostic alternatives, diagnostic tests must be evaluated to determine their place in the diagnostic armamentarium. The first step involves determining the accuracy of the test, including the sensitivity and specificity, positive and negative predictive values, likelihood ratios for positive and negative tests, and receiver operating characteristic (ROC) curves. The role of the test in a diagnostic pathway has then to be determined, following which the effect on patient outcome should be examined.

Texture Analysis of Gray-Scale Ultrasound Images for Staging of Hepatic Fibrosis

Purpose

Materials and Methods

Results

Conclusion

Deep Learning Algorithm for Automated Segmentation and Volume Measurement of the Liver and Spleen Using Portal Venous Phase Computed Tomography Images

Objective

Measurement of the liver and spleen volumes has clinical implications. Although computed tomography (CT) volumetry is considered to be the most reliable noninvasive method for liver and spleen volume measurement, it has limited application in clinical practice due to its time-consuming segmentation process. We aimed to develop and validate a deep learning algorithm (DLA) for fully automated liver and spleen segmentation using portal venous phase CT images in various liver conditions.

Materials and Methods

A DLA for liver and spleen segmentation was trained using a development dataset of portal venous CT images from 813 patients. Performance of the DLA was evaluated in two separate test datasets: dataset-1 which included 150 CT examinations in patients with various liver conditions (i.e., healthy liver, fatty liver, chronic liver disease, cirrhosis, and post-hepatectomy) and dataset-2 which included 50 pairs of CT examinations performed at ours and other institutions. The performance of the DLA was evaluated using the dice similarity score (DSS) for segmentation and Bland-Altman 95% limits of agreement (LOA) for measurement of the volumetric indices, which was compared with that of ground truth manual segmentation.

Results

In test dataset-1, the DLA achieved a mean DSS of 0.973 and 0.974 for liver and spleen segmentation, respectively, with no significant difference in DSS across different liver conditions (p = 0.60 and 0.26 for the liver and spleen, respectively). For the measurement of volumetric indices, the Bland-Altman 95% LOA was −0.17 ± 3.07% for liver volume and −0.56 ± 3.78% for spleen volume. In test dataset-2, DLA performance using CT images obtained at outside institutions and our institution was comparable for liver (DSS, 0.982 vs. 0.983; p = 0.28) and spleen (DSS, 0.969 vs. 0.968; p = 0.41) segmentation.

Conclusion

The DLA enabled highly accurate segmentation and volume measurement of the liver and spleen using portal venous phase CT images of patients with various liver conditions.

Imaging biomarkers of diffuse liver disease: current status

We are happy to introduce this special issue of Abdominal Radiology on "diffuse liver disease". We have invited imaging experts to discuss various topics pertaining to diffuse liver disease, covering a vast array of imaging techniques including ultrasound (US), CT, MRI and new molecular imaging agents. Below, we briefly discussed the current status, limitations, and future directions of imaging biomarkers of diffuse liver disease.

Use of imaging techniques for non-invasive assessment in the diagnosis and staging of non-alcoholic fatty liver disease

Non-alcoholic fatty liver disease (NAFLD) is a common cause of chronic liver disease and represent a common finding in highly prevalent metabolic disorders (i.e. type 2 diabetes, metabolic syndrome, obesity). Non-alcoholic steatohepatitis (NASH) requires liver biopsy for grading and staging the liver damage by the assessment of steatosis, inflammation and fibrosis. In parallel with the development of numerous 'liquid' biomarkers and algorithms that combine anthropometric and laboratory parameters, innovative hepatic imaging techniques have increasingly been developed to attempt to overcome the need for biopsy, both in diagnosis and staging of NAFLD, and in possible use in the follow-up of the disease. In this review, we focused on the different imaging techniques trying to highlight the strengths and disadvantages of different approaches, particularly for ultrasound techniques, in stratifying liver injury and fibrosis in patients with NAFLD / NASH.

Radiomics Approaches for Predicting Liver Fibrosis With Nonenhanced T1 -Weighted Imaging: Comparison of Different Radiomics Models

BACKGROUND

Liver fibrosis is a common process resulting from various etiologies. Sustained progression of liver fibrosis leads to cirrhosis, even hepatocellular carcinoma. Thus, noninvasive staging of liver fibrosis is of clinical importance. Radiomics is an emerging approach for staging liver fibrosis. However, the feature selection methods and classifier models are complicated, and may result in a discrepancy of diagnostic performance owing to different radiomics models.

PURPOSE

To identify the optimal feature selection and classifier methods for predicting liver fibrosis by using nonenhanced T₁ -weighted imaging.

STUDY TYPE

Prospective.

ANIMAL MODEL

Wistar rats, total 97.

FIELD STRENGTH/SEQUENCE

3T, 3D T₁ -weighted images with fast-spoiled gradient echo (FSPGR).

ASSESSMENT

Liver fibrosis rats were induced via subcutaneous injection of a mixture of carbon tetrachloride. Rats in the control group were injected with saline. Segmentation and feature extraction were performed by 3D slicer and the image biomarker explorer (IBEX) software package. Data preprocessing, feature selection, model building, and model comparative evaluation were conducted with Python. The liver fibrosis stage was determined by pathological examination.

STATISTICAL TESTS

Receiver operating characteristic curve, fuzzy comprehensive evaluation.

RESULTS

For discriminating between F0 and F1-2, F0 and F3-4, F0 and F1-4, F0-1 and F2-4, F0-2 and F3-4, and F0-3 and F4, the accuracies of 12 radiomics models were 77.27-90.91%, 73.33-86.67%, 80.56-91.67%, 74.07-88.89%, 76.47-88.24%, and 79.49-92.31%, respectively. The AUCs of the radiomics models were 0.86-0.97, 0.85-0.95, 0.89-0.97, 0.81-0.96, 0.82-0.93, and 0.85-0.96, respectively. The least absolute shrinkage and selection operator / support vector machine (LASSO-SVM) model had high AUCs of 0.93-0.97. For discriminating between F0 and F1-2, F0 and F3-4, F0 and F1-4, F0-1 and F2-4, and F0-2 and F3-4, the fuzzy comprehensive evaluation showed that the LASSO-SVM model had a high fuzzy score/order of 0.087-0.091/1.

DATA CONCLUSION

LASSO-SVM appears to be the optimal model for predicting liver fibrosis by using nonenhanced T₁ -weighted imaging in a rodent model of liver fibrosis.

LEVEL OF EVIDENCE

2.

TECHNICAL EFFICACY STAGE

2.

Machine-learning-based multiple abnormality prediction with large-scale chest computed tomography volumes

Machine learning models for radiology benefit from large-scale data sets with high quality labels for abnormalities. We curated and analyzed a chest computed tomography (CT) data set of 36,316 volumes from 19,993 unique patients. This is the largest multiply-annotated volumetric medical imaging data set reported. To annotate this data set, we developed a rule-based method for automatically extracting abnormality labels from free-text radiology reports with an average F-score of 0.976 (min 0.941, max 1.0). We also developed a model for multi-organ, multi-disease classification of chest CT volumes that uses a deep convolutional neural network (CNN). This model reached a classification performance of AUROC >0.90 for 18 abnormalities, with an average AUROC of 0.773 for all 83 abnormalities, demonstrating the feasibility of learning from unfiltered whole volume CT data. We show that training on more labels improves performance significantly: for a subset of 9 labels - nodule, opacity, atelectasis, pleural effusion, consolidation, mass, pericardial effusion, cardiomegaly, and pneumothorax - the model's average AUROC increased by 10% when the number of training labels was increased from 9 to all 83. All code for volume preprocessing, automated label extraction, and the volume abnormality prediction model is publicly available. The 36,316 CT volumes and labels will also be made publicly available pending institutional approval.

Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension

The CONSORT 2010 statement provides minimum guidelines for reporting randomised trials. Its widespread use has been instrumental in ensuring transparency in the evaluation of new interventions. More recently, there has been a growing recognition that interventions involving artificial intelligence (AI) need to undergo rigorous, prospective evaluation to demonstrate impact on health outcomes. The CONSORT-AI (Consolidated Standards of Reporting Trials-Artificial Intelligence) extension is a new reporting guideline for clinical trials evaluating interventions with an AI component. It was developed in parallel with its companion statement for clinical trial protocols: SPIRIT-AI (Standard Protocol Items: Recommendations for Interventional Trials-Artificial Intelligence). Both guidelines were developed through a staged consensus process involving literature review and expert consultation to generate 29 candidate items, which were assessed by an international multi-stakeholder group in a two-stage Delphi survey (103 stakeholders), agreed upon in a two-day consensus meeting (31 stakeholders), and refined through a checklist pilot (34 participants). The CONSORT-AI extension includes 14 new items that were considered sufficiently important for AI interventions that they should be routinely reported in addition to the core CONSORT 2010 items. CONSORT-AI recommends that investigators provide clear descriptions of the AI intervention, including instructions and skills required for use, the setting in which the AI intervention is integrated, the handling of inputs and outputs of the AI intervention, the human-AI interaction and provision of an analysis of error cases. CONSORT-AI will help promote transparency and completeness in reporting clinical trials for AI interventions. It will assist editors and peer reviewers, as well as the general readership, to understand, interpret, and critically appraise the quality of clinical trial design and risk of bias in the reported outcomes.

Guidelines for clinical trial protocols for interventions involving artificial intelligence: the SPIRIT-AI Extension

The SPIRIT 2013 (The Standard Protocol Items: Recommendations for Interventional Trials) statement aims to improve the completeness of clinical trial protocol reporting, by providing evidence-based recommendations for the minimum set of items to be addressed. This guidance has been instrumental in promoting transparent evaluation of new interventions. More recently, there is a growing recognition that interventions involving artificial intelligence need to undergo rigorous, prospective evaluation to demonstrate their impact on health outcomes.The SPIRIT-AI extension is a new reporting guideline for clinical trials protocols evaluating interventions with an AI component. It was developed in parallel with its companion statement for trial reports: CONSORT-AI. Both guidelines were developed using a staged consensus process, involving a literature review and expert consultation to generate 26 candidate items, which were consulted on by an international multi-stakeholder group in a 2-stage Delphi survey (103 stakeholders), agreed on in a consensus meeting (31 stakeholders) and refined through a checklist pilot (34 participants).The SPIRIT-AI extension includes 15 new items, which were considered sufficiently important for clinical trial protocols of AI interventions. These new items should be routinely reported in addition to the core SPIRIT 2013 items. SPIRIT-AI recommends that investigators provide clear descriptions of the AI intervention, including instructions and skills required for use, the setting in which the AI intervention will be integrated, considerations around the handling of input and output data, the human-AI interaction and analysis of error cases.SPIRIT-AI will help promote transparency and completeness for clinical trial protocols for AI interventions. Its use will assist editors and peer-reviewers, as well as the general readership, to understand, interpret and critically appraise the design and risk of bias for a planned clinical trial.