Publications

A 12-month fasting-mimicking diet programme reduced liver fat (MRI-PDFF) and inflammation/fibrosis (cT1) in type 2 diabetes patients. Improvements in cT1 and PDFF correlated with reductions in HbA1c, fasting glucose, triglycerides, weight, and total cholesterol. This dietary approach shows promise as an adjunct therapy to standard diabetes care.

Quantitative MRCP-derived metrics showed high accuracy, repeatability, and reproducibility across scanners, effectively distinguishing healthy individuals from PSC patients. These findings support the use of MRCP+ metrics for assessing biliary health and their suitability for multi-centre clinical trials and longitudinal monitoring in liver disease

This study presents a deep learning framework for accurate, personalized Couinaud liver segment localization using MRI data. It outperforms existing methods in landmark placement and segment volume estimation, is explainable, adaptable across imaging types, and scalable, potentially improving liver surgery planning and clinical workflow efficiency

Among UK Biobank participants with MRI, elevated liver cT1 and reduced left-ventricular ejection fraction independently—or in combination—predicted cardiovascular and liver events and all-cause mortality. Cardiac and hepatic impairment often coexisted, supporting multi-organ MRI as a tool for integrated risk stratification and prevention pathways spanning heart and liver disease.

Treatment interventions in MASLD significantly reduce liver cT1 by an average of 57 ms, indicating improved liver health. Fibroblast growth factor analogues, GLP-1 receptor agonists, and FXR agonists showed notable effects, while placebo showed no change. These results inform study design, supporting cT1 as a useful non-invasive treatment response marker.

Adding quantitative MRCP+ metrics to standard MRCP improves radiologists’ confidence and agreement in detecting high-grade strictures in primary sclerosing cholangitis patients. MRCP+ reduces variability, enhances diagnostic reliability, and may help standardize assessments, potentially improving patient management and treatment decisions

Multiparametric MRI, including cT1, was used to non-invasively monitor treatment response to lanifibranor in NASH. cT1 was significantly reduced after 24 weeks of treatment compared to placebo, indicating reduced hepatic fibroinflammation. Moreover, cT1 change was associated with histologic improvements in SAF activity score, supporting cT1 as a sensitive, non-invasive imaging biomarker for assessing therapeutic response in NASH

Genetic risk and lifestyle factors influence steatotic liver disease and cardiovascular events. Lifestyle effects were stronger in individuals with higher genetic risk, especially if overweight. Genetic factors, physical activity, and sleep duration interacted to influence cT1 and cardiovascular events. Targeted lifestyle interventions could be particularly effective in genetically susceptible populations.

Psoriatic disease (PsD) patients show a high prevalence of metabolic liver conditions (44% with MASLD and 22% with MASH) despite normal blood tests. Rates exceed those in matched controls and occur regardless of methotrexate use, highlighting the need for routine liver screening in PsD management e.g. by multiparametric MRI

Left atrial volume (LAV) is key in diagnosing heart failure with preserved ejection fraction. Compared to cardiovascular magnetic resonance (CMR), transthoracic echocardiography frequently fails or underestimates LAV, misclassifying enlargement in 38% of cases. CMR remains the more accurate gold standard for LAV assessment

A 37-year-old woman with psoriatic arthritis and undiagnosed liver disease showed significant reversal of liver fibro-inflammation after 6 months of IL-17A inhibitor (secukinumab) treatment. Quantitative mpMRI and blood tests confirmed improved liver health, suggesting biologic disease-modifying antirheumatic drugs  may benefit coexisting metabolic liver disease and psoriatic disease.

Multiparametric MRI detected increased risk of MASH or advanced fibrosis in over half of MASLD patients classified as low risk by FIB-4. Early use of multiparametric MRI could identify at-risk patients, particularly women and those with diabetes, enabling timely intervention and better disease management compared to current guideline-based risk stratification.

Once-weekly semaglutide significantly improved liver histology in patients with MASH and fibrosis, compared to placebo. It also achieved substantial weight loss, although gastrointestinal side effects were more common. Semaglutide improved noninvasive liver markers, though the high placebo response underscores the need for more precise monitoring tools.

In suspected acute biliary or ductal gallstone presentations, first-line MRCP outperformed ultrasound-led standard care—delivering quicker, more cost-effective diagnosis and obstruction detection. MRCP+ also quantified gallbladder volume and cystic duct width, adding value for triage and management and supporting direct-access MRCP pathways.

In the Precision1 trial, preoperative mpMRI and whole-genome sequencing (WGS) informed personalized surgery for operable colorectal liver metastases. mpMRI frequently influenced surgical plans (e.g., embolization, parenchymal-sparing or more extensive resections, weight management). WGS provided therapeutic and prognostic insights, supporting tailored, multidisciplinary decision-making.

Quantitative MRCP metrics, especially total biliary tree volume, effectively distinguished malignant from benign biliary obstructions. A volume ≥25 ml predicted malignancy with good accuracy (AUC 0.79). These findings support MRCP+ as a non-invasive, contrast-free tool for identifying cholangiocarcinoma, potentially aiding earlier diagnosis and improving patient outcomes.

In this randomised control trial in people with MASLD without diabetes, LiverMultiScan monitored improvements in liver fat and liver disease activity following 12-week treatment with liraglutide, and the subsequent rebound after cessation. A low-calorie diet of similar duration achieved equivalent improvements.

Guidelines prioritise early detection of chronic liver disease progression risk. To avoid biopsy, NITs have become part of standard‐of‐care. Across etiologies, patients with low fibrosis risk by MRE but high disease activity by cT1 have ~threefold higher progression, showing cT1 flags risk earlier than stiffness alone. Integrating cT1 into care pathways improves stratification and proactive management without biopsy.

Single-slice MRI at the L3 vertebral level provides accurate and reproducible area measurements of fat and muscle across diverse body types, multiple MRI scanner field strengths, and manufacturers. The measurements are consistent between trained analysts and experienced radiologists.

Severe energy restriction safely induced significant weight loss and reduced liver fat in patients with compensated cirrhosis due to MASLD (CC-MASLD). No adverse effects or liver deterioration were observed, while both MRI-PDFF and cT1 were significantly reduced. These promising results support further research into diet-based interventions for CC-MASLD management

Manual srT1, a standardized biomarker of pancreatic fibro-inflammation, showed excellent intra- and inter-observer reproducibility and strong scan–rescan reliability. A semi-automated workflow further improved scan–rescan agreement.

TERN-501, a THRβ agonist, significantly reduced liver fat in a dose-dependent manner over 12 weeks in patients with MASH. Compared to placebo, MRI-PDFF and cT1 reductions were substantial, and significantly more participants receiving TERN-501 achieved a clinically significant change in cT1 by 6 weeks.

Incorporating LiverMultiScan into the MASLD diagnostic pathway is cost-effective, significantly reduces the number of specialist appointments, patient assessments, and rate of diagnosis without increasing other liver-related health care resource use. cT1 also correctly classified 50% of patients without MASH with fibrosis who could avoid biopsy. Standardized adoption of LiverMultiScan can streamline clinical pathways, lower costs, and improve long-term management.

Higher dietary flavonoid intake was linked to lower risk of NAFLD. Participants with the highest flavonoid-rich diet scores had significantly lower cT1 values than those with the lowest scores. Tea, red wine, and sweet peppers were particularly associated with reduced cT1, suggesting flavonoid-rich diets may help lower risk for NAFLD.

An AI-assisted workflow allowed non-radiologists to perform liver delineation, Couinaud segmentation, and lesion detection on MRI with accuracy comparable to radiologists, reducing workload. Agreement on whole-liver volume and segment measurements was high, and all lesions identified by radiologists were found; non-radiologists also flagged additional lesions confirmed by external adjudicator radiologists.

Across three multicentre trials, changes in cT1 and PDFF predicted histologic response to MASH therapy. A reduction in cT1 of -80 ms was associated with a 5-fold increase in the odds of achieving resolution of steatohepatitis, a 30% PDFF reduction had roughly fourfold higher odds, while the optimal threshold of -58% had sevenfold higher odds.

Guidelines note an indirect MRI method for MASH and fibrosis diagnostics, LiverMultiScan, that can be performed on standard scanners without an elastography unit. By generating iron-corrected T1 (cT1) maps from T1 and T2* data, it estimates extracellular volume, reflecting inflammatory activity and fibrosis, aiding MASH and fibrosis assessment.

In a 12-week MASLD trial, pemvidutide (a GLP-1/glucagon dual agonist) markedly reduced liver fat and improved cT1 versus placebo, alongside improvements in body weight and liver enzymes. A high proportion achieved clinically meaningful cT1 reductions. Treatment was well tolerated without serious adverse events, supporting pemvidutide as a promising therapy.

In liver metastases, functional liver volumes automatically segmented using cT1 maps were used to create liver-sparing plans, which significantly reduced the functional liver mean dose versus standard planning. This first demonstration of cT1-guided stereotactic ablative radiotherapy suggests a route to spare healthy liver, though validation in liver cancer cohorts is required.

This study demonstrates clinical benefits of combining imaging markers with blood markers, such as GGT, in patients without a history of pancreatitis who have obesity and MASLD. Pairing LiverMultiScan with standard blood markers enables early identification and risk stratification of fatty pancreas in individuals with MASLD.

Poor diet quality and specific genetic alleles are associated with steatotic liver disease. Diet has a much greater impact on hepatic steatosis in those at higher genetic risk, affecting both liver fat content and cT1 (a measure of fibrosis and inflammation). Thus, dietary interventions may yield greater benefit in genetically susceptible groups.

Two new  MRI-based scoring strategies—single-step MPcT (MRE+PDFF+cT1) and two-step M-PcT (LSM by MRE then PDFF with cT1)—were evaluated to detect at-risk MASH. M-PcT showed superior diagnostic accuracy over FAST, MAST, and MPcT. PDFF plus cT1 also performed strongly for diagnosing active MASH and severe hepatic inflammation, outperforming FAST and MAST scores.

The study included pancreatic T1/PDFF and hepatic cT1/PDFF measures to assess pancreas–liver relationships in MASLD. Intrapancreatic fat deposition was not associated with hepatic steatosis. Pancreatic T1 was increased across disease stages, consistent with pancreatic inflammation in individuals with impaired glucose homeostasis.

Multiparametric MRI assessed liver fat (PDFF) and fibroinflammation (cT1) in Hispanic prepubertal boys at risk for MASLD. Despite normal transaminases, mpMRI identified MASLD/MASH, with higher cT1/PDFF in those with metabolic risk factors. Elevated cT1 aligned with insulin resistance, supporting mpMRI, particularly cT1, for routine MASLD screening of high-risk pediatric groups.

Using multiparametric MRI metrics alongside routine blood markers supports identification of liver transplant patients without rejection, enabling many to avoid a liver biopsy. Findings support a practical, non-invasive pathway to monitor graft health and reduce unnecessary invasive procedures.

CoverScan efficiently detects multi-organ abnormalities in patients with blood cancer, enabling more personalised treatment and potentially better outcomes. Compared with healthy controls, multi-organ impairment was more frequent, predominantly affecting the spleen and kidneys.

Hepatica showed a significant improvement in future liver remnant (FLR) health, measured by cT1 and PDFF, in 10 consecutive patients with liver metastasis and insufficient FLR after undergoing dual vein embolization in the Precision1 clinical trial. “This is a significant advance in our understanding of how to prevent liver failure in patients undergoing major liver surgery.”

Quantitative digital pathology objectively measures inflammatory activity in autoimmune hepatitis, showing strong concordance with consensus histological assessment. By quantifying inflamed tissue area and inflammatory foci density on whole-slide images, the approach enables automated assessment, better stratification, and identification of patients with more advanced inflammatory disease.

The first comprehensive analysis of multi-organ steatosis and fibro-inflammation in people living with obesity and type 2 diabetes receiving routine hospital care. Those on weight loss treatments saw improvements in organ health that did not occur in patients on metformin, highlighting the measurable and visible impact of weight-loss therapies in real-world patients.

MRCP+ metrics differentiated between AIH, PBC and PSC, highlighting its potential as an adjunctive diagnostic tool. In PSC, baseline strictures and bile-duct metrics correlated with Modified Amsterdam Score, transient elastography, and ALP. Over one year, strictures increased in more than half of patients, while biochemistry, elastography, or Amsterdam Score showed no change.

This study demonstrates the accuracy and reproducibility of LiverMultiScan metrics (cT1 and PDFF) in patients with MASH and MASLD. The results quantify the minimal detectable (absolute) changes in cT1 (32 ms) and liver fat fraction (0.6%) that discern measurement error from real change when monitoring patients with MASLD on a single scanner.

Multi-centre pooled analysis established reference ranges for cT1 and PDFF in healthy children and young adults. Values were normally distributed (median cT1 748 ms; PDFF 1.7%) and within the reported normal limits (<800 ms; <5%). Age comparisons showed clinically negligible cT1 and PDFF differences differences between paediatric and adult groups.

Kidney cortical T1 mapping is a highly repeatable and reproducible method across MRI manufacturers, field strengths, and operator conditions. The standardized methodological consistency paves the way for its adoption in multi-centre clinical trials.

In UK Biobank, women with PCOS had greater prevalence and incidence of type 2 diabetes, steatotic liver disease, and all-cause cardiovascular disease. Liver MRI showed more severe steatosis and higher fibroinflammation by cT1. These comorbidities are increased in older PCOS patients with a disease phenotype-specific pattern.

Technical overview of quantitative MRI advancements including abdominal T1 mapping of liver, pancreas, and kidneys. T1 mapping produces color-coded maps that highlight changes related to inflammation and fibrosis, aiding in diagnoses like liver fibro-inflammation. Standardized methods are emphasized to improve diagnostic accuracy and enable reliable longitudinal assessment.

Among adults with obesity and MASLD, MRI biomarkers, including cT1, assessed liver health and identified MASH with high accuracy. Bariatric surgery achieved greater reductions in cT1 and PDFF than lifestyle modification, indicating superior improvement in liver health and supporting cT1 for monitoring treatment response.

In a longitudinal bariatric surgery case series, liver fat (PDFF), cT1, and the glutamate–serine–glycine (GSG) index all showed reductions postoperatively. These findings support integrating blood and imaging biomarkers to monitor improvement and guide interventions.

Over 52 weeks, weekly tirzepatide significantly increased MASH resolution without fibrosis worsening in biopsy-confirmed MASH with F2–F3 fibrosis. It also showed improvement of at least one fibrosis stage without worsening MASH. cT1 was significantly reduced in the treatment group compared to placebo, showing a dose-dependent reduction.

First validated prognostic model of clinical outcomes in PSC using a novel composite marker derived from quantitative MRCP metrics, bilirubin, and AST. It outperformed the traditional Mayo risk score in predicting transplant-free survival in patients with PSC, suggesting improved risk stratification and potential to improve clinical management.

In NASH patients with obesity and fibrosis on a low-energy total diet replacement, LiverMultiScan cT1 detected significant reductions in liver disease activity. Compared to other NITs (CAP, MRE, VCTE), cT1 was the only consistent and reliable biomarker that captured change in liver disease severity, highlighting its utility for monitoring response to diet and lifestyle interventions.

An editorial on UK digital hepatology highlights the diagnostic and prognostic value of LiverMultiScan and MRCP+ in AILDs. LiverMultiScan’s cT1 has a prognostic value as it correlates with the risk of loss of biochemical remission in AIH. MRCP+ correlates with validated risk scoring systems and may improve diagnosis and prognostic models in primary PSC.

AASLD’s new practice guidance on the clinical management of NAFLD recommends sequentially integrating cT1 with other noninvasive tests for the identification and risk-stratification of patients with “at-risk” NASH in secondary/speciality care to facilitate better clinical management of these patients.

Using Bayesian-network modeling, this study links body-fat distribution to COVID-19 hospitalization risk. Obesity increased risk versus normal weight; across all BMI groups, higher visceral fat and liver fat were associated with markedly greater hospitalization probability. Findings underscore multi-organ imaging and Bayesian models for understanding and stratifying COVID-19 risk.

CoverScan identified cardiac abnormalities in ~20%patients at 6 months, persisting in over half of those at 12 months following initial COVID-19 symptoms. Importantly, blood biomarkers were not as sensitive as MRI in identifying cardiac abnormalities, highlighting its utility in the clinical diagnosis and management of patients with multi-system conditions.

This editorial provides insights into the clinical utility of multi-modal imaging to study metabolic diseases with organ-specific and extra-organ manifestations requiring integrated care: “the multisystem nature of metabolic disease puts multiorgan imaging front and centre for integration of care to avoid putting patients at risk of the complications associated with biopsy”.

Across large population cohorts, clonal haematopoiesis of indeterminate potential was linked to higher risk of chronic liver disease, including NASH. Specifically, LiverMultiScan findings in UK Biobank participants revealed a higher prevalence of liver disease activity and liver fat in individuals with CHIP. LiverMultiScan may represent a modifiable risk factor for e.g. blood cancers and cardiovascular disease.

Among children and young adults with autoimmune liver disease (AILD), cT1 mapping correlated strongly with histology, highlighting its ability to accurately assess liver inflammatory activity and fibrosis stage. Findings support cT1 as a useful, noninvasive tool that could reduce reliance on liver biopsy for assessing and monitoring AILD.

This state-of-the-art review highlights the clinical utility of cT1 in AIH diagnosis and management: it provides a quantitative, accurate, and reliable measurement of liver fibroinflammation, which correlates with histology, and predicts clinical outcomes. cT1 serves as a “virtual biopsy”, providing a “panoramic view” of liver tissue characteristics; it helps assess the heterogenous distribution of disease across the liver, supports monitoring, risk-stratification, and informs treatment-withdrawal decisions.

Growth hormone (GH) administration for six months showed promising results in ameliorating NAFLD in overweight and obese individuals. Specifically, cT1 measurements revealed significant improvements in liver disease activity, suggesting the potential of GH therapy as a novel approach for managing NAFLD in this population.

Individuals with suspected fibrotic NASH treated with FXR agonist vonafexor showed a mean 80 ms absolute reduction in cT1 at week 12 from baseline, indicating a significant reduction in liver disease activity post-treatment. These findings suggest that vonafexor is a promising therapeutic for reversing liver fibro-inflammation in NASH.

Patients with increased liver disease activity (cT1) were more likely to develop cardiac conditions. Specifically, individuals with increased cT1 were ~30% more likely to be hospitalized due to a major adverse cardiovascular event. Findings suggest liver disease activity is a meaningful risk factor for heart disease and support cT1 for cardiovascular risk stratification.

The protocol for the LITMUS Imaging Study, a prospective multi-centre cohort study on adults with NAFLD, evaluating non-invasive imaging tools (LiverMultiScan, MR elastography, and ultrasound elastography) against liver histology as the reference standard. The goal is to identify the imaging tools that accurately stage the severity of liver fibrosis in people with NAFLD.

Among people recovering from acute COVID, multi-organ MRI (CoverScan) detected organ impairment in nearly 70% of patients at six months, persisting in over half at one year. Some were asymptomatic, yet impairment also tracked symptoms, especially high liver fat with severe breathlessness, supporting integration of multi-organ MRI into routine follow-up care.

This first-of-its-kind study used LiverMultiScan and MRE to assess NAFLD in patients with morbid obesity over a six-month period following bariatric surgery. Liver disease activity (cT1) and liver fat (PDFF) significantly improved, but there was no change in liver stiffness (MRE). These results highlight LiverMultiScan’s utility for detecting early, clinically meaningful change during postoperative monitoring.

This review distills LITMUS lessons on biomarker qualification to speed NAFLD/NASH drug development. cT1 (LiverMultiScan) was submitted to EMA/FDA for prognostic enrichment and now holds an FDA-accepted Biomarker Qualification Program, the first for a diagnostic enrichment biomarker in NASH trials, positioning cT1 to streamline recruitment and improve trial success.

This review highlights the importance of assessing liver inflammation in NAFLD with non-invasive biomarkers, especially in the drug development process, owing to the inaccuracies of liver biopsies. LiverMultiScan’s cT1 emerges as a robust biomarker that correlates with necroinflammation, fibrosis, and liver-related clinical outcomes, with fewer technical failures than ultrasound-based techniques.

An expert review on advances in noninvasive tests for NAFLD recognizes cT1 as a validated tool for detecting NASH and at-risk NASH, correlating with NAS features and fibrosis. cT1 showed the highest rule-in specificity (90%) among the reviewed noninvasive tests for at-risk NASH at a cut-off of ≥ 875 ms.

Using a Bayesian-network model, this study shows that in high-risk adults with type 2 diabetes, excess liver fat, poor glycaemic control, and excess visceral fat raise NASH risk by 31%, 22%, and 20%, respectively, with important implications for the development of targeted interventions to prevent NASH in high-risk populations.

Review of prognostic tools for the clinical management of patients with AILD. In AIH, LiverMultiScan cT1, an objective, accurate measure of fibro-inflammation, may be determinant in clinical decision-making and could substitute liver biopsy in follow-up. It also outlines the prognostic value of MRCP+ for PSC management.

This editorial reviews imaging modalities for NAFLD/NASH, endorsing PDFF for diagnosing NAFLD and positioning cT1 as the leading biomarker for NASH. It highlights cT1's utility for identifying patients with 'at-risk' NASH (with fibrosis) and for risk-stratifying patients with intermediate or indeterminate results on other guideline-recognized tests (AACE, AGA, DGVS).

In this phase 2b randomized controlled trial, treatment with FGF21 analogue pegozafermin, a novel therapy for treating NASH, led to improvements in liver fat and fibro-inflammation (liver disease activity) quantified by cT1 from baseline up to 24 weeks. These results support the advancement of pegozafermin into phase 3 development.

This opinion piece states that MASLD is an obesity-related condition that requires both imaging- and genetic-risk tools for a comprehensive picture of multi-organ associations. Without imaging data, genetic tools such as the Mendelian randomisation method alone may not provide clinically relevant information on liver health and cardiovascular risk in patients with MASLD.

An invasive minilaparoscopic biopsy of both liver lobes in AIH (AIH, AIH-PBC, and AIH-PSC) was more accurate for grading/staging disease activity than percutaneous biopsy, reflecting intrahepatic heterogeneity. Noninvasive LiverMultiScan cT1 demonstrated this heterogeneity, and its extent independently predicted disease flares, supporting multi-lobe sampling and cT1 for comprehensive assessment.

This expert review proposes a clinical care pathway for suspected NAFLD, positioning cT1 as the key individual biomarker to risk-stratify patients and flag high-risk NASH to facilitate timely management. It further suggests that cT1 could replace or precede liver biopsy in all patients at intermediate/high risk of fibrosis, subject to hepatologist judgment.

This review on the value of AI-supported digital applications in radiology describes LiverMultiScan as a tool detecting liver diseases with a high degree of accuracy compared to gold-standard liver biopsies, delivering comparable diagnostic information without the risks associated with an invasive procedure. It also reports that LiverMultiScan-derived parameters strongly predict clinical outcomes.

In Fontan circulation, higher liver cT1 (fibrosis/congestion) correlated with impaired exercise capacity and adverse 4D blood flow measured with CMR. Findings link hepatic injury to hemodynamic inefficiency, supporting combined 4D flow CMR and LiverMultiScan for noninvasive risk assessment.

Retrospective PSC cohort: quantitative MRCP (MRCP+) stricture count and spleen length independently associated with adverse events. MRCP+ outperformed manual radiological (ANALI) scores for risk stratification in PSC, supporting use for longitudinal monitoring and as a surrogate endpoint in clinical trials.

A rare genetic variant (SLC30A10 p.Thr95Ile) is associated with elevated ALT and higher cT1 measured by LiverMultiScan, markers of liver fibro-inflammation, indicating the variant may increase liver-disease risk in the general population.

This American Gastroenterological Association (AGA) clinical practice update provides best practice advice on evidence-based approaches diagnosis, staging, and management of NAFLD in lean individuals. It recommends combining noninvasive biomarkers, such as cT1, with fibrosis tests (VCTE, MRE, and EFL) to identify and risk-stratify lean patients with NASH for most appropriate treatment.

AACE guideline provides evidence-based recommendations for NAFLD/NASH diagnosis and management in primary care and endocrinology. Its cirrhosis-prevention algorithm advises referring patients at indeterminate or high fibrosis risk for specialist evaluation using advanced noninvasive biomarkers, including cT1 (LiverMultiScan), to refine staging and guide care.

This international survey revealed major limitations in NAFLD/NASH management: delayed referrals, suboptimal screening, and absence of reliable tools for diagnose active steatohepatitis. Most respondents used multiple noninvasive fibrosis tests but had not used cT1, underscoring unmet need; LiverMultiScan cT1 could fill this diagnostic and monitoring gap.

This review outlines the evidence on how AI and machine learning can address unmet needs in autoimmune liver diseases. In PSC, MRCP+ metrics can significantly improve the current diagnostic approach and risk prediction. In AIH, LiverMultiScan’s cT1 is described as an accurate non-invasive biomarker for risk stratification and monitoring, enabling precision-medicine workflows.

In paediatric and young adult with autoimmune liver disease AILD, multiparametric MRI (LiverMultiScan) and MRCP+ showed robust associations with laboratory markers. cT1 correlated with markers of liver disease activity, and cT1 IQR appeared less confounded than MRE. MRCP+ biliary metrics discriminated AIH from PSC/ASC, supporting combined imaging for noninvasive stratification and monitoring.

In a population of children and young adults with autoimmune liver disease (AILD), quantitative metrics from LiverMultiScan and MRCP+ were significantly associated with the MAYO and SCOPE risk scores and were independent predictors. These findings support multiparametric MRI using LiverMultiScan and MRCP+ as noninvasive tools for diagnosis risk stratification, and prognosis in AILD.

In a UK Biobank secondary analysis using LiverMultiScan, variants in glycogen-metabolism genes associated with MRI-PDFF were not linked to NAFLD risk, suggesting hepatic glycogen can bias PDFF and should be considered when interpreting MRI-based liver fat measurements.

UK Biobank analysis shows that people with type 2 diabetes have higher liver fat (PDFF) and fibro-inflammation (cT1) and lower skeletal muscle index than non-diabetic controls. In addition, LiverMultiScan detected greater liver damage in participants with T2D despite normal liver biochemistry (AST/ALT).

Early economic modelling from NHS England’s Oxford AHSN found LiverMultiScan to be a cost-effective, noninvasive alternative to painful, invasive surveillance liver biopsy for monitoring autoimmune hepatitis (AIH) across disease stages, saving up to £336,926, thus lowering NHS costs and procedure burden while enabling repeatable follow-up in routine care.

Longitudinal mpMRI in paediatric/young adult autoimmune liver disease showed individual, but not group, changes in cT1, MRE, and quantitative MRCP over two years. Changes in biliary MRCP+ metrics correlated with changes in liver stiffness (MRE) and fibro-inflammation (MRE, cT1), supporting personalized trajectory tracking and noninvasive monitoring.

In treated chronic hepatitis B with advanced fibrosis, liver iron quantified by LiverMultiScan predicted hepatocellular carcinoma (HCC) occurrence over 2 years. The same scan also assessed the pancreas, supporting a single, noninvasive workflow for longitudinal monitoring and prognostication in chronic hepatitis B.

Review advocates streamlined, noninvasive pathways for cirrhosis assessment, replacing fragmented diagnostics and biopsies. LiverMultiScan cT1 is highlighted as a fast, fully noninvasive biomarker that can simplify clinical pathways, support diagnosis and screening of liver diseases, and reduce clinic burden.

This large UK Biobank study explored the relationship between heart, brain, and liver health. Individuals with poor liver health — quantified by cT1 mapping — were more likely to have brain abnormalities (e.g., cognitive decline and dementia) and poor heart health, underscoring the importance of multi-organ assessment when assessing an individual's overall health.

This article reviews the evaluation, diagnosis, and management of adults with NAFLD/NASH in the primary care setting and endocrinology clinics. LiverMultiScan was cited as a potential tool for liver imaging and assessment to ensure prompt diagnosis of NAFLD/NASH in routine practice.

A novel MRI segmentation algorithm divides the pancreas into head, body, and tail, enabling regional quantification. It achieved excellent accuracy and reproducibility, supporting assessment of heterogeneous pancreatic disease and revealing subsegment-specific alterations in diabetes.

Study protocol for Precision1 trial: an observational radiogenomics cohort in liver tumours. Primary endpoint: proportion of patients whose management would change based on whole-genome sequencing. Secondary aims assess decision impact of LiverMultiScan- and Hepatica, and correlate quantitative MRI with genomic and histologic features to guide precision care.

Quantitative MRCP metrics in large-duct PSC correlated with established prognostic factors, ANALI (morphological scores based on MRCP images), liver stiffness (VCTE), and biochemical scores. Median duct diameter was associated with outcomes, underscoring quantitative MRCP’s prognostic value in patients with PSC.

First-of-its-kind evaluation in biopsy-proven AIH: LiverMultiScan’s cT1 increases clinician confidence and influences management by improving disease characterization and early detection of patients in biochemical remission with undetected, active sub-clinical disease at a high risk of relapse, compared to VCTE and serum biomarkers alone.

In pediatric Wilson disease, LiverMultiScan’s cT1 and PDFF correlated with histological scores of fibrosis, inflammation, and steatosis, whereas ultrasound did not. MRCP+ identified dilatations in the biliary tree, not detected during the standard of care examination.

Opinion piece on regulating Artificial Intelligence (AI) and Machine Learning (ML) in software as a medical device (SaMD): advocates stronger technology literacy, clear evidence standards across the full product lifecycle, and continuous engagement of end-users to demonstrate safety and effectiveness—shifting SaMD oversight to be technology-enabled, evidence-driven, and patient-centred.

This review highlights CMR’s value in convalescent COVID-19, revealing frequent myocardial abnormalities months after infection. CMR also enabled the identification of multi-organ damage, indicating the importance of further investigation into the relationship between cardiac injury, ongoing symptoms, pre-existing risk factors, and damage to other organs.

Case report using MRCP+ to convert MRCP into 3D biliary maps with quantitative metrics. Intrahepatic biliary stricture disease showed distinct ductal abnormalities versus controls. MRCP+ visualized and quantified strictures and dilatations across the entire tree at a glance, supporting earlier detection of subtle intrahepatic biliary disease.

In large-duct PSC, MRCP+ quantification of biliary dilatation identified high-risk patients and showed superior interobserver agreement when compared to qualitative MRCP, supporting MRCP+ for objective risk stratification and consistent reporting.

Multicenter, biopsy-paired NAFLD analysis shows MRI PDFF and cT1 accurately identify NASH; cT1 better pinpoints patients at high risk of progression than liver fat alone. The data supports cT1 as a noninvasive biomarker for triage, risk stratification, and trial enrichment.

Biopsy-referenced comparison in NAFLD: LiverMultiScan PDFF performed well across all steatosis grades, while CAP (ultrasound) was adequate at lower fat levels but inferior to liver PDFF for moderate–severe steatosis. This supports MRI-PDFF as the preferred noninvasive metric when precise quantification is needed.

This study compares quantitative biliary measurements obtained using MRCP+ and three different MRCP acquisition methods, showing that MRCP with compressed sensing to reduce acquisition time produced comparable biliary diameter metrics and global duct quantification to clinically indicated MRCP.

The correlations between cT1 and PDFF with the histopathological hallmarks of NASH demonstrate the potential utility of both cT1 and PDFF as non-invasive biomarkers to detect a pharmacodynamic change in NASH, with cT1 showing superiority for detecting changes in inflammation and fibrosis, rather than liver fat alone.

LiverMultiScan’s cT1 was the primary endpoint in a multicenter, randomized, double-blind, placebo-controlled trial on the effect of Mastiha on liver inflammation and fibrosis. Over six months, cT1 decreased in participants with severe obesity, with no significant change on placebo.

In PSC, MRCP+ quantitative duct metrics correlated with biochemical risk scores and MR elastography, with performance comparable to ANALI morphological scores. Combining MRCP+ with ANALI or with MRE improved association with the Mayo Risk Score, supporting integrated imaging and morphology for noninvasive risk assessment.

In a large-scale UK Biobank study, hepatic steatosis, not obesity alone, predicted COVID-19 severity: obese individuals with elevated liver fat had ~5-fold higher hospitalization risk. LiverMultiScan can quantify liver fat to stratify risk and support public health policy, clinical triage, and targeted lifestyle interventions.

Synoptic reporting provides a comprehensive and structured format, which results in a complete assessment of liver disease. This opinion piece proposes using synoptic reporting, in place of the existing narrative style, in liver biopsies for NAFLD and NASH clinical trials to improve trial success rates and accelerate drug development.

Review of incidence of all complications and technical failure associated with percutaneous liver biopsy shows notable risk: major complications ~2% (mostly bleeding), minor ~10% (mostly pain), and ~1% technical failure. These findings support consideration of standardized, noninvasive alternatives for diagnosis and monitoring.

In LITMUS, serum miRNA profiling identified elevated miR-193a-5p with worsening fibrosis and higher NAS/SAF scores, replicated in an independent cohort. Findings support circulating miRNAs, especially miR-193a-5p, as noninvasive biomarkers to detect advanced NAFLD without a liver biopsy.

LiverMultiScan’s cT1 decreased in patients with chronic hepatitis C virus (HCV) undergoing direct-acting antiviral therapy. The fast reduction in cT1 is consistent with reduced inflammation rather than fibrosis regression, supporting LiverMultiScan as a sensitive biomarker for treatment monitoring.

This study highlights the utility of Hepatica for identifying patients at increased risk of poor post-operative liver performance and longer hospital stay. Pre-operative cT1 predicts length of post-operative hospital stay and has potential to inform the individualised risk assessment and surgical decision-making for liver cancer surgery.

Prospective community study of low-risk individuals with persistent post-COVID symptoms found widespread multi-organ involvement months after infection, affecting the heart, lungs, kidneys, pancreas and liver, with cardiac impairment more frequent in severe cases. These findings demonstrate feasibility and value of community-based, multiparametric assessment to guide management before and after infection.

In paediatric autoimmune hepatitis (AIH), LiverMultiScan’s cT1 correlated with biochemical markers of disease severity and detected active disease even when routine labs were normal. These findings support cT1 as a sensitive, non-invasive tool to aid monitoring in AIH and guide timely management.

In AIH, failure to maintain remission was associated with increased cT1, not FibroScan liver stiffness or ELF. Baseline cT1 also predicted subsequent biochemical relapse, whereas liver stiffness or ELF did not, supporting LiverMultiScan as a sensitive tool for risk stratification and earlier intervention.

Multistakeholder NAFLD guideline—co-developed by patients, patient representatives, clinicians, and scientists—summarizes current care pathways. The emerging noninvasive tests section highlights multiparametric MRI as a promising tool to identify NASH with fibrosis, referencing Schaapman et al. (2020), which integrated LiverMultiScan cT1 into a clinical risk-assessment algorithm for NAFLD.

Automated Hepatica delivered accurate, repeatable liver and Couinaud segment volumes, matching expert radiologists while reducing inter-operator variability and measurement time. Standardized, deep-learning outputs support robust pre-surgical planning and comprehensive liver health assessment.

In asymptomatic, middle-aged U.S. adults, LiverMultiScan PDFF estimated NAFLD prevalence at 38% and biopsy-confirmed NASH prevalence was 14%. NASH associated with race, obesity, and diabetes. These findings underscore LiverMultiScan’s suitability for scalable population assessment to identify at-risk individuals and enrich NASH cohorts.

Case report. Two colorectal liver metastasis resections with contrasting outcomes: retrospective pre-op Hepatica revealed high cT1 and PDFF in the patient who developed complications and prolonged stay, versus normal cT1 and PDFF in the short-stay case with no complications, indicating quantitative mpMRI can flag at-risk patients and guide safer surgical planning.

Paediatric AILD cohort with AIH and ASC underwent LiverMultiScan, MRCP+, biopsy, and blood-based liver function tests. Quantitative MRCP+ biliary metrics discriminated AIH from ASC, supporting noninvasive subtype stratification. Although cT1 did not differ between subtypes, LiverMultiScan provided parenchymal characterization and longitudinal monitoring, complementing MRCP+ for comprehensive assessment.

Comparative study in patients with historical diagnosis of NAFLD or suspected of NAFLD showed that cT1 and PDFF (LiverMultiScan) outperformed FibroScan, MRE, and SWE for identifying NASH, providing strong evidence for clinical adoption.

Automated Hepatica delivered accurate, repeatable liver and Couinaud segment volumes, matching expert radiologists while reducing inter-operator variability and measurement time. Standardized, deep-learning outputs support robust pre-surgical planning and comprehensive liver health assessment.

cT1 (LiverMultiScan) was significantly reduced after 16 weeks of novel NASH treatment, AXA1125, versus placebo, with twice as many patients achieving a reduction of cT1 ≥ 80 ms, previously linked to improved histopathological outcomes. AXA1125 also reduced liver fat (MRI-PDFF) and improved metabolic and fibro-inflammatory markers (HOMA-IR, ALT, CK18-M65, Pro-C3).

Serum MMP7 was evaluated as a biomarker for sclerosing cholangitis in paediatric patients with autoimmune liver disease (AILD). Higher sMMP7 levels associated with histologic bile duct injury and hepatic fibrosis, and on MRCP+ correlated with the number of biliary dilatations and strictures, supporting sMMP7 as a noninvasive indicator of ductal disease severity.

In biopsy-confirmed NAFLD, LiverMultiScan’s cT1, blood-based AST, GGT, and fasting glucose predicted significant fibrosis. The combined cTAG score (cT1+AST+glucose) outperformed individual markers, highlighting its utility for screening patients prior to biopsy to identify those suitable for NASH clinical trial enrolment.

In pediatric autoimmune liver disease, LiverMultiScan showed an excellent positive correlation between liver native T1 and iron-corrected cT1 in a cross-sectional study, suggesting limited iron confounding in this cohort.

Quantitative MRCP+ outperformed serum biochemistry in differentiating paediatric PSC/ASC from AIH: all but one MRCP+ parameter significantly differed between cohorts and predicted diagnosis, whereas no laboratory values were significantly different between cohorts. Findings highlight MRCP+ as a useful non-invasive tool providing multiple imaging biomarkers of autoimmune liver disease.

This study across two independent UK cohorts found limited ethnic differences in body-fat depot distribution linked to metabolic disease, suggesting that alternative mechanisms should be investigated. In the UK Biobank cohort, LiverMultiScan measured PDFF using a standardized workflow, scalable for large populations and enabling consistent, cross-site comparisons.

GWAS study of cT1 (LiverMultiScan) in 14,440 UK Biobank participants identified SLC39A as a genetic variant associated with fibro-inflammatory liver disease. Findings highlight SLC39A8 as a potential drug target and reinforce cT1 as a genetics-informed biomarker.

This review discusses bidirectional links between NAFLD and type 2 diabetes, recommends MAFLD nomenclature, and details screening with noninvasive tests/imaging for advanced fibrosis. Outlines lifestyle, glucose-lowering drugs, and metabolic surgery, noting extra-hepatic risks and medication considerations.

This review of techniques for clinical decision-making in diffuse liver disease highlights the advantages of cT1 (LiverMultiScan), showing superior NASH diagnostic performance compared with MRE, which requires additional hardware and is confounded by even mild iron overload and inflammation, and compared with DWI, which currently lacks standardization.

This review contains the first European guidance on multiparametric MRI, including cT1, for suspected NAFLD: outlines how to assess fibrosis and inflammation risk and suggests how clinical management and treatment monitoring can be improved.

Modern liver imaging pairs structure with function: techniques such as multiparametric MRI deliver noninvasive surrogate biomarkers for focal and diffuse disease, supporting diagnosis, staging, and treatment monitoring.

Nutrition-focused review of NAFLD/NASH: how macro- and micronutrients shape disease biology and overlap with drug targets. Lifestyle modification remains first-line therapy. The review also offers a practical dietary framework to integrate with medical management and address cardiometabolic risk.

A new technique for group-wise registration and manifold learning of pancreas images could be used for automated quality control and to determine abnormal pancreas morphology, which might reflect different patterns of fat infiltration.

LiverMultiScan’s noninvasive MRI biomarker cT1 predicts clinical outcomes in liver disease patients, outperforming FibroScan’s transient elastography and FIB-4, with performance comparable to invasive liver biopsy. After accounting for high technical failure rates, FibroScan lost prognostic value—underscoring cT1 as a robust, noninvasive risk-stratification biomarker.

Quantitative MRCP (MRCP+) is an accurate, repeatable and reproducible method. Compared with qualitative MRCP, MRCP+ reduces subjectivity, quantifies duct diameters along the biliary tree, and detects candidate strictures and dilatations with high sensitivity and specificity. Quantitative MRCP improves upon the current “gold standard” of noninvasive imaging of the biliary tree.

In pediatric and young adult autoimmune liver disease, splenic T1 was associated with radiologic and clinical fibrosis markers, including radiologic portal hypertension.

Meta-analysis of more than 42 studies, including UK Biobank and cohorts with LiverMultiScan fat quantification confirms that a common genetic variant is a risk factor for presence and severity of NAFLD in individuals of European descent.

Monogenic signal–based MRCP segmentation accurately detects bile ducts and quantifies duct diameters, outperforming the widely used Frangi vesselness filter—enhancing automated biliary tree delineation for quantitative MRCP.

Radiology-inspired machine learning method combining complementary tumour characteristics from both arterial phase and portal venous phase CT images offers improved accuracy in automatically delineating liver lesions.

Protocol for RADIcAL1 study: a multi-centre, two-arm RCT across four European countries testing whether standardized mpMRI (LiverMultiScan) as a diagnostic test for suspected NAFLD in tertiary hepatology clinics is cost-effective versus current pathways (consultations and/or liver biopsy) in over 1072 adults.

Evaluates the utility of quantitative MRI biomarkers including elastography and cT1 mapping for predicting portal hypertension in a paediatric population. Liver and spleen stiffness, along with liver cT1, predict radiologic portal hypertension with good accuracy.

Three genetic variants that are linked to an increased risk of developing higher liver iron content were identified. Mendelian randomisation analysis provided evidence that higher central obesity plays a causal role in increased liver iron content.

Technical validation of LiverMultiScan technique (MAGO) for measurement of liver fat: MAGO showed excellent accuracy and reproducibility across MR vendors at different field strengths. Technically challenging cases were processable with MAGO, demonstrating robustness of the technique.

Results from a Phase 2 NASH trial showing that LiverMultiScan (cT1) can detect drug efficacy in as little as 6 weeks to serve as an early go/no-go signal. cT1 differentiated responders, tracked changes, and correlated with histology and serum biomarkers, enabling sensitive, non-invasive longitudinal monitoring.

Proof-of-concept evaluation of multi-parametric MRI (LiverMultiScan) for non-invasive assessment of portal hypertension. Spleen cT1 emerged as a promising marker of portal pressure, outperforming established non-invasive scores and indicating potential for safer, more accurate monitoring.

Review of non-invasive methods for assessing liver disease in NAFLD and how they could be used in clinical practice. Highlights the key issues of differentiating NASH from simple steatosis and identifying advanced fibrosis.

This study establishes cT1 reference ranges in a large, low-risk UK Biobank cohort, which serves as a benchmark for normal liver, aids clinical interpretation and study design, reinforcing cT1 as a reliable quantitative biomarker for liver health and disease.

Technical validation of LiverMultiScan demonstrating good reproducibility and repeatability for quantifying liver tissue characteristics across different MR vendors and field strengths. Standardized biomarkers with low variability, like LiverMultiScan, are crucial for reliable readings in longitudinal monitoring in multicentre clinical trials.

LiverMultiScan’s cT1 was used alongside other technologies to identify new indications for existing type 2 diabetes drugs. cT1 confirmed an absence of treatment effect of dapagliflozin in this small NAFLD cohort.

Review of imaging methods for NAFLD risk stratification and management, including multiparametric MRI (LiverMultiScan), MR Elastography and Ultrasound Elastography.

Describes a method for computing a “water T1” by correcting the MOLLI T1 for the confounding influence of iron, fat, and frequency offsets. This has the potential to improve the characterisation of fibro-inflammatory liver disease.

LiverMultiScan was used to quantify hepatic iron levels in 9,108 UK Biobank participants, demonstrating feasibility for large-scale, non-invasive population assessment of liver iron with this technique.

From a radiologist’s perspective, LiverMultiScan delivers early, non-invasive detection of diffuse liver disease with reproducible, decision-ready biomarkers. It reduces biopsies, outperforms other non-invasive tests, and drives cost-effective patient stratification and monitoring, helping providers act sooner and health systems save more.

LiverMultiScan achieved a 98.1% technical success rate—including in obesity—outperforming transient elastography (85%). It showed excellent repeatability and reproducibility and strong diagnostic accuracy for hepatic fibroinflammation, fat, and iron.

LiverMultiScan accurately quantified levels of hepatic fibroinflammation, fat, and iron in preclinical models.

This ongoing trial will refine the technology and clinical application of LiverMultiScan in quantifying pre-existing liver health and predicting post-intervention outcomes following liver resection in cancer.

cT1 showed superior reproducibility (CoV 3.1%) compared to the other non-invasive liver tests used in this NASH clinical trial: magnetic resonance elastography (MRE, CoV 11%) and transient elastography (TE, 40%). This underscores utility of cT1 as a sensitive, reliable marker for monitoring longitudinal change in NASH patients.

Technical validation of the LiverMultiScan technique (LMS IDEAL) for liver fat quantification: high accuracy and excellent reproducibility across major MR vendors and field strengths and over a wide range of values. Perspectum is the only imaging provider with the IDEAL technique standardized across vendors, enabling consistent, scalable multi-site deployment.

Validation of the technology underpinning cT1: iron correction of liver T1 values produces more accurate measurement of hepatic fibroinflammation.

LiverMultiScan was used to measure hepatic fat in ~5,000 participants in the UK Biobank with a success rate of 96.8% and acquisition time of 3 minutes, enabling cost-effective population screening at scale.

Assessment of liver fibrosis by digital imaging analysis of Collagen Proportionate Area (CPA) followed by visual assessment of CPA is more robust than visual analysis alone with improved agreement between pathologists.

LiverMultiScan had a higher success rate in NAFLD patients (95%) compared to transient elastography (59%), and showed high accuracy for the diagnosis of NASH and ballooning. This demonstrates potential application in patient management and supports use as a surrogate endpoint in clinical trials.

LiverMultiScan accurately identified patients with steatosis, stratified those with NASH, and reliably excluded clinically significant liver disease, with superior negative predictive value (83.3%) vs. transient elastography (42.9%) and ELF (57.1%). For NAFLD risk stratification, LiverMultiScan was cost-effective and, when combined with transient elastography, delivered the lowest cost per correct diagnosis.

The inclusion of LiverMultiScan, either as an adjunct to or replacement of transient elastography (TE) in the diagnostic pathway of NAFLD, leads to cost savings for the NHS by reducing the number of liver biopsies required. When used in place of TE and liver biopsy, LiverMultiScan remains cost-effective up to a price of £672.

LiverMultiScan showed elevated hepatic fibroinflammatory and fat levels in asymptomatic patients with type 2 diabetes, suggesting significant NAFLD and NASH. This highlights an urgent need – and potential application of the technology – in screening, staging and monitoring diabetic liver disease.

Hepatic fat measured by 1H-MRS is related to increased aortic stiffness in both adults and children. Hepatic fat content is a potential therapeutic target to treat the elevated vascular risk in obesity, which requires reliable biomarkers for monitoring response to treatment.

LiverMultiScan demonstrated 100% negative predictive value for liver-related clinical outcomes, in patients monitored for up to 40 months. This holds prognostic potential for stratification of patients to facilitate clinical management and clinical trials.

Proton MRS (1H-MRS) and cardiac MRI showed that even in the absence of other comorbidities, adult and childhood obesity is related to cardiac steatosis. Higher myocardial triglyceride content relates to diastolic dysfunction, stronger in males. The findings suggest that cardiac steatosis occurs early in obesity and significantly affects childhood.

First study to report that abdominal obesity in women, and post-menopausal weight gain, are major drivers of very low-density lipoprotein secretion, which may increase risk of cardiovascular disease.

Men were observed to have a higher prevalence and greater risk of NAFLD compared with women despite similar levels of hepatic fat content. This is potentially driven by sex- specific differences in fatty acid metabolism.

First clinical validation showing that LiverMultiScan accurately stages chronic liver disease and has high diagnostic accuracy for assessment of liver fibrosis, steatosis and haemosiderosis. This is the first demonstration of a non-invasive test differentiating early stages of fibrosis from normal liver with high diagnostic accuracy.

Cardiac lipid levels measured using rapid single breath-hold 1H-magnetic resonance spectroscopy (1H-MRS) at 3T shows excellent correlation with results obtained by averaging over seven multiple breath-holds. This study demonstrates that single breath-hold 1H-MRS at 3T is a rapid and reliable method to quantify myocardial lipids.