Automated Whole-Liver MRI Segmentation to Assess Steatosis and Iron Quantification in Chronic Liver Disease

Automated hepatic steatosis assessment on dual-energy CT-derived virtual non-contrast images through fully-automated 3D organ segmentation

PURPOSE

To evaluate the efficacy of volumetric CT attenuation-based parameters obtained through automated 3D organ segmentation on virtual non-contrast (VNC) images from dual-energy CT (DECT) for assessing hepatic steatosis.

MATERIALS AND METHODS

This retrospective study included living liver donor candidates having liver DECT and MRI-determined proton density fat fraction (PDFF) assessments. Employing a 3D deep learning algorithm, the liver and spleen were automatically segmented from VNC images (derived from contrast-enhanced DECT scans) and true non-contrast (TNC) images, respectively. Mean volumetric CT attenuation values of each segmented liver (L) and spleen (S) were measured, allowing for liver attenuation index (LAI) calculation, defined as L minus S. Agreements of VNC and TNC parameters for hepatic steatosis, i.e., L and LAI, were assessed using intraclass correlation coefficients (ICC). Correlations between VNC parameters and MRI-PDFF values were assessed using the Pearson's correlation coefficient. Their performance to identify MRI-PDFF ≥ 5% and ≥ 10% was evaluated using receiver operating characteristic (ROC) curve analysis.

RESULTS

Of 252 participants, 56 (22.2%) and 16 (6.3%) had hepatic steatosis with MRI-PDFF ≥ 5% and ≥ 10%, respectively. LVNC and LAIVNC showed excellent agreement with LTNC and LAITNC (ICC = 0.957 and 0.968) and significant correlations with MRI-PDFF values (r = - 0.585 and - 0.588, Ps < 0.001). LVNC and LAIVNC exhibited areas under the ROC curve of 0.795 and 0.806 for MRI-PDFF ≥ 5%; and 0.916 and 0.932, for MRI-PDFF ≥ 10%, respectively.

CONCLUSION

Volumetric CT attenuation-based parameters from VNC images generated by DECT, via automated 3D segmentation of the liver and spleen, have potential for opportunistic hepatic steatosis screening, as an alternative to TNC images.

A fully automatic parenchyma extraction method for MRI T2* relaxometry of iron loaded liver in transfusion-dependent patients

PURPOSE

To develop a fully automatic parenchyma extraction method for the T2* relaxometry of iron overload liver.

METHODS

A retrospective multicenter collection of liver MR examinations from 177 transfusion-dependent patients was conducted. The proposed method extended a semiautomatic parenchyma extraction algorithm to a fully automatic approach by introducing a modified TransUNet on the R2* (1/T2*) map for liver segmentation. Axial liver slices from 129 patients at 1.5 T were allocated to training (85%) and internal test (15%) sets. Two external test sets separately included 1.5 T data from 20 patients and 3.0 T data from 28 patients. The final T2* measurement was obtained by fitting the average signal of the extracted liver parenchyma. The agreement between T2* measurements using fully and semiautomatic parenchyma extraction methods was assessed using coefficient of variation (CoV) and Bland-Altman plots.

RESULTS

Dice of the deep network-based liver segmentation was 0.970 ± 0.019 on the internal dataset, 0.960 ± 0.035 on the external 1.5 T dataset, and 0.958 ± 0.014 on the external 3.0 T dataset. The mean difference bias between T2* measurements of the fully and semiautomatic methods were separately 0.12 (95% CI: -0.37, 0.61) ms, 0.04 (95% CI: -1.0, 1.1) ms, and 0.01 (95% CI: -0.25, 0.23) ms on the three test datasets. The CoVs between the two methods were 4.2%, 4.8% and 2.0% on the internal test set and two external test sets.

CONCLUSIONS

The developed fully automatic parenchyma extraction approach provides an efficient and operator-independent T2* measurement for assessing hepatic iron content in clinical practice.

Using Dual-source Photon-counting Detector CT to Simultaneously Quantify Fat and Iron Content: A Phantom Study

RATIONALE AND OBJECTIVES

To evaluate the feasibility of using photon-counting detector computed tomography (PCD CT) to simultaneously quantify fat and iron content MATERIALS AND METHODS: Phantoms with pure fat, pure iron and fat-iron deposition were scanned by two tube voltages (120 and 140 kV) and two image quality (IQ) settings (80 and 145). Using an iron-specific three-material decomposition algorithm, virtual noniron (VNI) and virtual iron content (VIC) images were generated at quantum iterative reconstruction (QIR) strength levels 1-4.

RESULTS

Significant linear correlations were observed between known fat content (FC) and VNI for pure fat phantoms (r = 0.981-0.999, p < 0.001) and between known iron content (IC) and VIC for pure iron phantoms (r = 0.897-0.975, p < 0.001). In fat-iron phantoms, the measurement for fat content of 5-30% demonstrated good linearity between FC and VNI (r = 0.919-0.990, p < 0.001), and VNI were not affected by 75, 150, and 225 µmol/g iron overload (p = 0.174-0.519). The measurement for iron demonstrated a linear range of 75-225 µmol/g between IC and VIC (r = 0.961-0.994, p < 0.001) and VIC was not confounded by the coexisting 5%, 20%, and 30% fat deposition (p = 0.943-0.999). The Bland-Altman of fat and iron measurements were not significantly different at varying tube voltages and IQ settings (all p > 0.05). No significant difference in VNI and VIC at QIR 1-4.

CONCLUSION

PCD CT can accurately and simultaneously quantify fat and iron, including scan parameters with lower radiation dose.

Effectiveness of deep learning-based reconstruction for improvement of image quality and liver tumor detectability in the hepatobiliary phase of gadoxetic acid-enhanced magnetic resonance imaging

PURPOSE

To evaluate the effectiveness of deep learning-based reconstruction (DLR) in improving image quality and tumor detectability of isovoxel high-resolution breath-hold fat-suppressed T1-weighted imaging (HR-BH-FS-T1WI) in the hepatobiliary phase (HBP) of Gadoxetic acid-enhanced magnetic resonance imaging (Gd-EOB-MRI).

MATERIALS AND METHODS

This retrospective evaluated 42 patients with 98 liver tumors who underwent Gd-EOB-MRI between March 2023 and May 2023 using three techniques based on HBP imaging: isovoxel HR-BH-FS-T1WI reconstructed (1) with DLR (BH-DLR +) and (2) without DLR (BH-DLR -) and (3) HR-FS-T1WI scanned with a free-breathing technique using a navigator-echo-triggered technique and DLR (Navi-DLR +). The three techniques were qualitatively and quantitatively compared by the Friedman test and the Bonferroni post-hoc test. Tumor detectability was compared using the McNemar test.

RESULTS

BH-DLR + (3.85, average score of two radiologists) showed significantly better qualitative scores for image noise than BH-DLR - (2.84) and Navi-DLR + (3.37) (p < 0.0167), and Navi-DLR + showed significantly better scores than BH-DLR - (p < 0.0167). BH-DLR + (3.77) and BH-DLR - (3.77) showed significantly better qualitative scores for respiratory motion artifact than Navi-DLR + (2.75) (p < 0.0167), but there was no significant difference in scores between BH-DLR + and BH-DLR - (p > 0.0167). BH-DLR + (0.32) and Navi-DLR + (0.33) showed significantly higher lesion-to-nonlesion CR than BH-DLR - (0.29) (p < 0.0167), but there was no significant difference in lesion-to-nonlesion CR between BH-DLR + and Navi-DLR + (p > 0.0167). BH-DLR + (89.8%) showed significantly better tumor detectability than BH-DLR - (76.0%) and Navi-DLR + (77.6%) (p < 0.05).

CONCLUSION

The use of DLR for isovoxel HR-BH-FS-T1WI was effective in improving image quality and tumor detectability.

Confidence maps for reliable estimation of proton density fat fraction and R 2 * in the liver

PURPOSE

The objective was to develop a fully automated algorithm that generates confidence maps to identify regions valid for analysis of quantitative proton density fat fraction (PDFF) andR 2 * $$ {R}_2^{\ast } $$ maps of the liver, generated with chemical shift-encoded MRI (CSE-MRI). Confidence maps are urgently needed for automated quality assurance, particularly with the emergence of automated segmentation and analysis algorithms.

METHODS

Confidence maps for both PDFF andR 2 * $$ {R}_2^{\ast } $$ maps are generated based on goodness of fit, measured by normalized RMS error between measured complex signals and the CSE-MRI signal model. Based on Cramér-Rao lower bound and Monte-Carlo simulations, normalized RMS error threshold criteria were developed to identify unreliable regions in quantitative maps. Simulation, phantom, and in vivo clinical studies were included. To analyze the clinical data, a board-certified radiologist delineated regions of interest (ROIs) in each of the nine liver segments for PDFF andR 2 * $$ {R}_2^{\ast } $$ analysis in consecutive clinical CSE-MRI data sets. The percent area of ROIs in areas deemed unreliable by confidence maps was calculated to assess the impact of confidence maps on real-world clinical PDFF andR 2 * $$ {R}_2^{\ast } $$ measurements.

RESULTS

Simulations and phantom studies demonstrated that the proposed algorithm successfully excluded regions with unreliable PDFF andR 2 * $$ {R}_2^{\ast } $$ measurements. ROI analysis by the radiologist revealed that 2.6% and 15% of the ROIs were placed in unreliable areas of PDFF andR 2 * $$ {R}_2^{\ast } $$ maps, as identified by confidence maps.

CONCLUSION

A proposed confidence map algorithm that identifies reliable areas of PDFF andR 2 * $$ {R}_2^{\ast } $$ measurements from CSE-MRI acquisitions was successfully developed. It demonstrated technical and clinical feasibility.

Signal Intensity Trajectories Clustering for Liver Vasculature Segmentation and Labeling (LiVaS) on Contrast-Enhanced MR Images: A Feasibility Pilot Study

This study aims to develop a semiautomated pipeline and user interface (LiVaS) for rapid segmentation and labeling of MRI liver vasculature and evaluate its time efficiency and accuracy against manual reference standard. Retrospective feasibility pilot study. Liver MR images from different scanners from 36 patients were included, and 4/36 patients were randomly selected for manual segmentation as referenced standard. The liver was segmented in each contrast phase and masks registered to the pre-contrast segmentation. Voxel-wise signal trajectories were clustered using the k-means algorithm. Voxel clusters that best segment the liver vessels were selected and labeled by three independent radiologists and a research scientist using LiVaS. Segmentation times were compared using a paired-sample t-test on log-transformed data. The agreement was analyzed qualitatively and quantitatively using DSC for hepatic and portal vein segmentations. The mean segmentation time among four readers was significantly shorter than manual (3.6 ± 1.4 vs. 70.0 ± 29.2 min; p < 0.001), even when using a higher number of clusters to enhance accuracy. The DSC for portal and hepatic veins reached up to 0.69 and 0.70, respectively. LiVaS segmentations were overall of good quality, with variations in performance related to the presence/severity of liver disease, acquisition timing, and image quality. Our semi-automated pipeline was robust to different MRI vendors in producing segmentation and labeling of liver vasculature in agreement with expert manual annotations, with significantly higher time efficiency. LiVaS could facilitate the creation of large, annotated datasets for training and validation of neural networks for automated MRI liver vascularity segmentation. HIGHLIGHTS: Key Finding: In this pilot feasibility study, our semiautomated pipeline for segmentation of liver vascularity (LiVaS) on MR images produced segmentations with simultaneous labeling of portal and hepatic veins in good agreement with the manual reference standard but at significantly shorter times (mean LiVaS 3.6 ± 1.4 vs. mean manual 70.0 ± 29.2 min; p < 0.001). Importance: LiVaS was robust in producing liver MRI vascular segmentations across images from different scanners in agreement with expert manual annotations, with significant ly higher time efficiency, and therefore potential scalability.

Early and accurate diagnosis of steatotic liver by artificial intelligence (AI)-supported ultrasonography

OBJECTIVES

Steatotic liver disease is the most frequent chronic liver disease worldwide. Ultrasonography (US) is commonly employed for the assessment and diagnosis. Few information is available on the possible use of artificial intelligence (AI) to ameliorate the diagnostic accuracy of ultrasonography.

MATERIALS AND METHODS

An AI-based algorithm was developed using a dataset of US images. We prospectively enrolled 134 patients for algorithm validation. Patients underwent abdominal US and Proton Density Fat Fraction MRI scans (MRI-PDFF), assumed as reference technique. The hepatorenal index was manually calculated (HRIM) by 4 operators. An automatic hepatorenal index (HRIA) was obtained by the algorithm. The accuracy of HRIA to discriminate steatosis grades was evaluated by ROC analysis using MRI-PDFF cut-offs.

RESULTS

Overweight was 40 % of subjects (BMI 26.4 kg/cm2). The median HRIA was 1.11 (IQR 0.32) and the average of 4 manually calculated HRIM was 1.08 (IQR 0.26), with a 15 % inter-operator variability. Both HRIA (R = 0.79, P < 0.0001) and HRIM (R = 0.69, P < 0.0001) significantly correlated with liver fat percentage (MRI-PDFF). According to MRI-PDFF, 32 % of enrolled subjects had steatosis. Discrimination capacity by AUC between patient with steatosis and patient without steatosis was better for HRIA than HRIM (AUC: 0.87 vs. 0.82, respectively). ROC analysis showed an AUC = 0.98 for HRIA with 1.64 cut-off in distinguishing between mild and moderate/severe groups.

CONCLUSIONS

The use of AI improves accuracy and speed of ultrasonography in the diagnosis of liver steatosis. Further studies should evaluate the routine use of this technique in the management of liver steatosis at high cardio-metabolic risk.

Development and validation of an image biomarker to identify metabolic dysfunction associated steatohepatitis: MR-MASH score

BACKGROUND AND AIMS

Diagnosis of metabolic dysfunction-associated steatohepatitis (MASH) requires histology. In this study, a magnetic resonance imaging (MRI) score was developed and validated to identify MASH in patients with metabolic dysfunction-associated steatotic liver disease (MASLD). Secondarily, a screening strategy for MASH diagnosis was investigated.

METHODS

This prospective multicentre study included 317 patients with biopsy-proven MASLD and contemporaneous MRI. The discovery cohort (Spain, Portugal) included 194 patients. NAFLD activity score (NAS) and fibrosis were assessed with the NASH-CRN histologic system. MASH was defined by the presence of steatosis, lobular inflammation, and ballooning, with NAS ≥4 with or without fibrosis. An MRI-based composite biomarker of Proton Density Fat Fraction and waist circumference (MR-MASH score) was developed. Findings were afterwards validated in an independent cohort (United States, Spain) with different MRI protocols.

RESULTS

In the derivation cohort, 51% (n = 99) had MASH. The MR-MASH score identified MASH with an AUC = .88 (95% CI .83-.93) and strongly correlated with NAS (r = .69). The MRI score lower cut-off corresponded to 88% sensitivity with 86% NPV, while the upper cut-off corresponded to 92% specificity with 87% PPV. MR-MASH was validated with an AUC = .86 (95% CI .77-.92), 91% sensitivity (lower cut-off) and 87% specificity (upper cut-off). A two-step screening strategy with sequential MR-MASH examination performed in patients with indeterminate-high FIB-4 or transient elastography showed an 83-84% PPV to identify MASH. The AUC of MR-MASH was significantly higher than that of the FAST score (p < .001).

CONCLUSIONS

The MR-MASH score has clinical utility in the identification and management of patients with MASH at risk of progression.

Simulation of Image-Guided Microwave Ablation Therapy Using a Digital Twin Computational Model

Emerging computational tools such as healthcare digital twin modeling are enabling the creation of patient-specific surgical planning, including microwave ablation to treat primary and secondary liver cancers. Healthcare digital twins (DTs) are anatomically one-to-one biophysical models constructed from structural, functional, and biomarker-based imaging data to simulate patient-specific therapies and guide clinical decision-making. In microwave ablation (MWA), tissue-specific factors including tissue perfusion, hepatic steatosis, and fibrosis affect therapeutic extent, but current thermal dosing guidelines do not account for these parameters. This study establishes an MR imaging framework to construct three-dimensional biophysical digital twins to predict ablation delivery in livers with 5 levels of fat content in the presence of a tumor. Four microwave antenna placement strategies were considered, and simulated microwave ablations were then performed using 915 MHz and 2450 MHz antennae in Tumor Naïve DTs (control), and Tumor Informed DTs at five grades of steatosis. Across the range of fatty liver steatosis grades, fat content was found to significantly increase ablation volumes by approximately 29-l42% in the Tumor Naïve and 55-60% in the Tumor Informed DTs in 915 MHz and 2450 MHz antenna simulations. The presence of tumor did not significantly affect ablation volumes within the same steatosis grade in 915 MHz simulations, but did significantly increase ablation volumes within mild-, moderate-, and high-fat steatosis grades in 2450 MHz simulations. An analysis of signed distance to agreement for placement strategies suggests that accounting for patient-specific tumor tissue properties significantly impacts ablation forecasting for the preoperative evaluation of ablation zone coverage.

From NAFLD to HCC: Advances in noninvasive diagnosis

Non-alcoholic fatty liver disease (NAFLD) has gradually become one of the major liver health problems in the world. The dynamic course of the disease goes through steatosis, inflammation, fibrosis, and carcinoma. Before progressing to carcinoma, timely and effective intervention will make the condition better, which highlights the importance of early diagnosis. With the further study of the biological mechanism in the pathogenesis and progression of NAFLD, some potential biomarkers have been discovered, and the possibility of their clinical application is gradually being discussed. At the same time, the progress of imaging technology and the emergence of new materials and methods also provide more possibilities for the diagnosis of NAFLD. This article reviews the diagnostic markers and advanced diagnostic methods of NAFLD in recent years.

Hepatic Involvement across the Metabolic Syndrome Spectrum: Non-Invasive Assessment and Risk Prediction Using Machine Learning

Metabolic-dysfunction-associated steatotic liver disease (MASLD) and metabolic syndrome (MetS) are inextricably linked conditions, both of which are experiencing an upward trend in prevalence, thereby exerting a substantial clinical and economic burden. The presence of MetS should prompt the search for metabolic-associated liver disease. Liver fibrosis is the main predictor of liver-related morbidity and mortality. Non-invasive tests (NIT) such as the Fibrosis-4 index (FIB4), aspartate aminotransferase-to-platelet ratio index (APRI), aspartate aminotransferase-to-alanine aminotransferase ratio (AAR), hepatic steatosis index (HIS), transient elastography (TE), and combined scores (AGILE3+, AGILE4) facilitate the detection of liver fibrosis or steatosis. Our study enrolled 217 patients with suspected MASLD, 109 of whom were diagnosed with MetS. We implemented clinical and biological evaluations complemented by transient elastography (TE) to discern the most robust predictors for liver disease manifestation patterns. Patients with MetS had significantly higher values of FIB4, APRI, HSI, liver stiffness, and steatosis parameters measured by TE, as well as AGILE3+ and AGILE4 scores. Machine-learning algorithms enhanced our evaluation. A two-step cluster algorithm yielded three clusters with reliable model quality. Cluster 1 contained patients without significant fibrosis or steatosis, while clusters 2 and 3 showed a higher prevalence of significant liver fibrosis or at least moderate steatosis as measured by TE. A decision tree algorithm identified age, BMI, liver enzyme levels, and metabolic syndrome characteristics as significant factors in predicting cluster membership with an overall accuracy of 89.4%. Combining NITs improves the accuracy of detecting patterns of liver involvement in patients with suspected MASLD.

Pancreatic steatosis and iron overload increases cardiovascular risk in non-alcoholic fatty liver disease

Objective

To assess the prevalence of pancreatic steatosis and iron overload in non-alcoholic fatty liver disease (NAFLD) and their correlation with liver histology severity and the risk of cardiometabolic diseases.

Method

A prospective, multicenter study including NAFLD patients with biopsy and paired Magnetic Resonance Imaging (MRI) was performed. Liver biopsies were evaluated according to NASH Clinical Research Network, hepatic iron storages were scored, and digital pathology quantified the tissue proportionate areas of fat and iron. MRI-biomarkers of fat fraction (PDFF) and iron accumulation (R2*) were obtained from the liver and pancreas. Different metabolic traits were evaluated, cardiovascular disease (CVD) risk was estimated with the atherosclerotic CVD score, and the severity of iron metabolism alteration was determined by grading metabolic hiperferritinemia (MHF). Associations between CVD, histology and MRI were investigated.

Results

In total, 324 patients were included. MRI-determined pancreatic iron overload and moderate-to severe steatosis were present in 45% and 25%, respectively. Liver and pancreatic MRI-biomarkers showed a weak correlation (r=0.32 for PDFF, r=0.17 for R2*). Pancreatic PDFF increased with hepatic histologic steatosis grades and NASH diagnosis (p<0.001). Prevalence of pancreatic steatosis and iron overload increased with the number of metabolic traits (p<0.001). Liver R2* significantly correlated with MHF (AUC=0.77 [0.72-0.82]). MRI-determined pancreatic steatosis (OR=3.15 [1.63-6.09]), and iron overload (OR=2.39 [1.32-4.37]) were independently associated with high-risk CVD. Histologic diagnosis of NASH and advanced fibrosis were also associated with high-risk CVD.

Conclusion

Pancreatic steatosis and iron overload could be of utility in clinical decision-making and prognostication of NAFLD.

Artificial Intelligence Applications in Hepatology

Over the past 2 decades, the field of hepatology has witnessed major developments in diagnostic tools, prognostic models, and treatment options making it one of the most complex medical subspecialties. Through artificial intelligence (AI) and machine learning, computers are now able to learn from complex and diverse clinical datasets to solve real-world medical problems with performance that surpasses that of physicians in certain areas. AI algorithms are currently being implemented in liver imaging, interpretation of liver histopathology, noninvasive tests, prediction models, and more. In this review, we provide a summary of the state of AI in hepatology and discuss current challenges for large-scale implementation including some ethical aspects. We emphasize to the readers that most AI-based algorithms that are discussed in this review are still considered in early development and their utility and impact on patient outcomes still need to be assessed in future large-scale and inclusive studies. Our vision is that the use of AI in hepatology will enhance physician performance, decrease the burden and time spent on documentation, and reestablish the personalized patient-physician relationship that is of utmost importance for obtaining good outcomes.

Clinical applications of artificial intelligence in liver imaging

This review outlines the current status and challenges of the clinical applications of artificial intelligence in liver imaging using computed tomography or magnetic resonance imaging based on a topic analysis of PubMed search results using latent Dirichlet allocation. LDA revealed that "segmentation," "hepatocellular carcinoma and radiomics," "metastasis," "fibrosis," and "reconstruction" were current main topic keywords. Automatic liver segmentation technology using deep learning is beginning to assume new clinical significance as part of whole-body composition analysis. It has also been applied to the screening of large populations and the acquisition of training data for machine learning models and has resulted in the development of imaging biomarkers that have a significant impact on important clinical issues, such as the estimation of liver fibrosis, recurrence, and prognosis of malignant tumors. Deep learning reconstruction is expanding as a new technological clinical application of artificial intelligence and has shown results in reducing contrast and radiation doses. However, there is much missing evidence, such as external validation of machine learning models and the evaluation of the diagnostic performance of specific diseases using deep learning reconstruction, suggesting that the clinical application of these technologies is still in development.

Quantification of Liver Iron Overload with MRI: Review and Guidelines from the ESGAR and SAR

Accumulation of excess iron in the body, or systemic iron overload, results from a variety of causes. The concentration of iron in the liver is linearly related to the total body iron stores and, for this reason, quantification of liver iron concentration (LIC) is widely regarded as the best surrogate to assess total body iron. Historically assessed using biopsy, there is a clear need for noninvasive quantitative imaging biomarkers of LIC. MRI is highly sensitive to the presence of tissue iron and has been increasingly adopted as a noninvasive alternative to biopsy for detection, severity grading, and treatment monitoring in patients with known or suspected iron overload. Multiple MRI strategies have been developed in the past 2 decades, based on both gradient-echo and spin-echo imaging, including signal intensity ratio and relaxometry strategies. However, there is a general lack of consensus regarding the appropriate use of these methods. The overall goal of this article is to summarize the current state of the art in the clinical use of MRI to quantify liver iron content and to assess the overall level of evidence of these various methods. Based on this summary, expert consensus panel recommendations on best practices for MRI-based quantification of liver iron are provided.

Automatic segmentation of hepatic metastases on DWI images based on a deep learning method: assessment of tumor treatment response according to the RECIST 1.1 criteria

BACKGROUND

Evaluation of treated tumors according to Response Evaluation Criteria in Solid Tumors (RECIST) criteria is an important but time-consuming task in medical imaging. Deep learning methods are expected to automate the evaluation process and improve the efficiency of imaging interpretation.

OBJECTIVE

To develop an automated algorithm for segmentation of liver metastases based on a deep learning method and assess its efficacy for treatment response assessment according to the RECIST 1.1 criteria.

METHODS

One hundred and sixteen treated patients with clinically confirmed liver metastases were enrolled. All patients had baseline and post-treatment MR images. They were divided into an initial (n = 86) and validation cohort (n = 30) according to the examined time. The metastatic foci on DWI images were annotated by two researchers in consensus. Then the treatment responses were assessed by the two researchers according to RECIST 1.1 criteria. A 3D U-Net algorithm was trained for automated liver metastases segmentation using the initial cohort. Based on the segmentation of liver metastases, the treatment response was assessed automatically with a rule-based program according to the RECIST 1.1 criteria. The segmentation performance was evaluated using the Dice similarity coefficient (DSC), volumetric similarity (VS), and Hausdorff distance (HD). The area under the curve (AUC) and Kappa statistics were used to assess the accuracy and consistency of the treatment response assessment by the deep learning model and compared with two radiologists [attending radiologist (R1) and fellow radiologist (R2)] in the validation cohort.

RESULTS

In the validation cohort, the mean DSC, VS, and HD were 0.85 ± 0.08, 0.89 ± 0.09, and 25.53 ± 12.11 mm for the liver metastases segmentation. The accuracies of R1, R2 and automated segmentation-based assessment were 0.77, 0.65, and 0.74, respectively, and the AUC values were 0.81, 0.73, and 0.83, respectively. The consistency of treatment response assessment based on automated segmentation and manual annotation was moderate [K value: 0.60 (0.34-0.84)].

CONCLUSION

The deep learning-based liver metastases segmentation was capable of evaluating treatment response according to RECIST 1.1 criteria, with comparable results to the junior radiologist and superior to that of the fellow radiologist.

Standardised lesion segmentation for imaging biomarker quantitation: a consensus recommendation from ESR and EORTC

BACKGROUND

Lesion/tissue segmentation on digital medical images enables biomarker extraction, image-guided therapy delivery, treatment response measurement, and training/validation for developing artificial intelligence algorithms and workflows. To ensure data reproducibility, criteria for standardised segmentation are critical but currently unavailable.

METHODS

A modified Delphi process initiated by the European Imaging Biomarker Alliance (EIBALL) of the European Society of Radiology (ESR) and the European Organisation for Research and Treatment of Cancer (EORTC) Imaging Group was undertaken. Three multidisciplinary task forces addressed modality and image acquisition, segmentation methodology itself, and standards and logistics. Devised survey questions were fed via a facilitator to expert participants. The 58 respondents to Round 1 were invited to participate in Rounds 2-4. Subsequent rounds were informed by responses of previous rounds.

RESULTS/CONCLUSIONS

Items with ≥ 75% consensus are considered a recommendation. These include system performance certification, thresholds for image signal-to-noise, contrast-to-noise and tumour-to-background ratios, spatial resolution, and artefact levels. Direct, iterative, and machine or deep learning reconstruction methods, use of a mixture of CE marked and verified research tools were agreed and use of specified reference standards and validation processes considered essential. Operator training and refreshment were considered mandatory for clinical trials and clinical research. Items with a 60-74% agreement require reporting (site-specific accreditation for clinical research, minimal pixel number within lesion segmented, use of post-reconstruction algorithms, operator training refreshment for clinical practice). Items with ≤ 60% agreement are outside current recommendations for segmentation (frequency of system performance tests, use of only CE-marked tools, board certification of operators, frequency of operator refresher training). Recommendations by anatomical area are also specified.

Deep Learning for the Automatic Segmentation of Extracranial Venous Malformations of the Head and Neck from MR Images Using 3D U-Net

Background: It is difficult to characterize extracranial venous malformations (VMs) of the head and neck region from magnetic resonance imaging (MRI) manually and one at a time. We attempted to perform the automatic segmentation of lesions from MRI of extracranial VMs using a convolutional neural network as a deep learning tool. Methods: T2-weighted MRI from 53 patients with extracranial VMs in the head and neck region was used for annotations. Preprocessing management was performed before training. Three-dimensional U-Net was used as a segmentation model. Dice similarity coefficients were evaluated along with other indicators. Results: Dice similarity coefficients in 3D U-Net were found to be 99.75% in the training set and 60.62% in the test set. The models showed overfitting, which can be resolved with a larger number of objects, i.e., MRI VM images. Conclusions: Our pilot study showed sufficient potential for the automatic segmentation of extracranial VMs through deep learning using MR images from VM patients. The overfitting phenomenon observed will be resolved with a larger number of MRI VM images.

Comparison of reader agreement, correlation with liver biopsy, and time-burden sampling strategies for liver proton density fat fraction measured using magnetic resonance imaging in patients with obesity: a secondary cross-sectional study

Background

The magnetic resonance imaging (MRI)-based proton density fat fraction (PDFF) has become popular for quantifying liver fat content. However, the variability of the region-of-interest (ROI) sampling strategy may result in a lack of standardisation of this technology. In an effort to establish an accurate and effective PDFF measurement scheme, this study assessed the pathological correlation, the reader agreement, and time-burden of different sampling strategies with variable ROI size, location, and number.

Methods

Six-echo spoiled gradient-recalled-echo magnitude-based fat quantification was performed for 50 patients with obesity, using a 3.0-T MRI scanner. Two readers used different ROI sampling strategies to measure liver PDFF, three times. Intra-reader and inter-reader agreement was evaluated using intra-class correlation coefficients and Bland‒Altman analysis. Pearson correlations were used to assess the correlation between PDFFs and liver biopsy. Time-burden was recorded.

Results

For pathological correlations, the correlations for the strategy of using three large ROIs in Couinaud segment 3 (S3 3L-ROI) were significantly greater than those for all sampling strategies at the whole-liver level (P < 0.05). For inter-reader agreement, the sampling strategies at the segmental level for S3 3L-ROI and using three large ROIs in Couinaud segment 6 (S6 3L-ROI) and the sampling strategies at the whole-liver level for three small ROIs per Couinaud segment (27S-ROI), one large ROI per Couinaud segment (9L-ROI), and three large ROIs per Couinaud segment (27S-ROI) had limits of agreement (LOA) < 1.5%. For intra-reader agreement, the sampling strategies at the whole-liver level for 27S-ROI, 9L-ROI, and 27L-ROI had both intraclass coefficients > 0.995 and LOAs < 1.5%. The change in the time-burden was the largest (100.80 s) when 9L-ROI was changed to 27L-ROI.

Conclusions

For hepatic PDFF measurement without liver puncture biopsy as the gold standard, and for general hepatic PDFF assessment, 9L-ROI sampling strategy at the whole-liver level should be used preferentially. For hepatic PDFF with liver puncture biopsy as the gold standard, 3L-ROI sampling strategy at the puncture site segment is recommended.