Machine learning predicts liver cancer risk from routine clinical data: a large population-based multicentric study DOI Open Access
Jan Clusmann, Paul‐Henry Koop, David Zhang

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 4, 2024

Abstract Background and aims Hepatocellular carcinoma (HCC) is a highly fatal tumor, for which early detection risk stratification crucial, yet remains challenging. We aimed to develop an interpretable machine-learning framework HCC based on routinely collected clinical data. Methods leverage data obtained from over 900,000 individuals 983 cases of across two large-scale population-based cohorts: the UK Biobank study “All Of Us Research Program”. For all these patients, timepoints years before diagnosis was available. integrate modalities including demographics, electronic health records, lifestyle, routine blood tests, genomics metabolomics offer unique, multi-modal perspective risk. Results Our random-forest-based model significantly outperforms publicly available state-of-the-art risk-scores, with AUROC 0.88 both internal external test sets. demonstrate robustness our ethnic subgroups, major advance previous models variable performance by ethnicity. Further, we perform extensive feature-importance analysis, showcasing approach as framework. provide weights open-source web calculator facili-tate further validation model. Conclusion presents robust stratification, offers potential improve could ultimately reduce disease burden through targeted interventions. Lay summary Finding liver cancer crucial successful treatment. Therefore, screening abdominal ultra-sound can be performed. However, it not clear who should receive ultrasound screening, current standard only patients cirrhosis, severe disease, many are diagnosed in late stages. trained machine learning model, acting like decision trees at same time, detect high looking patterns almost 1000 population 900.000 individuals. In separate set has seen during training, worked better than models. Additionally, investigated 1. how comes its prediction, 2. whether works males females alike 3. most relevant Like this, help sort into categories “high-risk”, “medium-risk” “low-risk”, via strategies then decided, cancer.

Language: Английский

Gut Microbiome-Liver-Brain axis in Alcohol Use Disorder. The role of gut dysbiosis and stress in alcohol-related cognitive impairment progression: possible therapeutic approaches. DOI Creative Commons
Emilio Merlo Pich, Ioannis Tarnanas, Patrizia Brigidi

et al.

Neurobiology of Stress, Journal Year: 2025, Volume and Issue: 35, P. 100713 - 100713

Published: Feb. 8, 2025

The Gut Microbiome-Liver-Brain Axis is a relatively novel construct with promising potential to enhance our understanding of Alcohol Use Disorder (AUD), and its therapeutic approaches. Significant alterations in the gut microbiome occur AUD even before any other systemic signs or symptoms manifest. Prolonged inappropriate alcohol consumption, by affecting microbiota mucosa permeability, thought contribute development behavioral cognitive impairments, leading Alcohol-Related Liver Disorders potentially progressing into alcoholic cirrhosis, which often associated severe impairment related neurodegeneration, such as hepatic encephalopathy dementia. critical role further supported efficacy FDA-approved treatments for cirrhosis (i.e., lactulose rifaximin). To stimulate new research, we hypothesize that interactions between maladaptive stress response constitutional predisposition neurodegeneration underlie progression conditions Clinical Concerns impairment, represent significant costly burden society. Early identification individuals at risk developing these could help prioritize integrated interventions targeting different substrates axis. Specifically, addiction medications, modulators, stress-reducing interventions, and, possibly soon, agents reduce steatosis/fibrosis will be discussed context digitally explicit goal this treatment performed on early stage disorder would transition from those Common strategy recommended most neurological neurodegenerative disorders.

Language: Английский

Citations

0
ASS1 is a hub gene and possible therapeutic target for regulating metabolic dysfunction-associated steatotic liver disease modulated by a carbohydrate-restricted diet DOI
Shaojun Chen,

Yanhua Bi,

Lihua Zhang

et al.

Molecular Diversity, Journal Year: 2025, Volume and Issue: unknown

Published: April 17, 2025

Language: Английский

Citations

0
Advancements in Artificial Intelligence-Enhanced Imaging Diagnostics for the Management of Liver Disease—Applications and Challenges in Personalized Care DOI Creative Commons
Naoshi Nishida

Bioengineering, Journal Year: 2024, Volume and Issue: 11(12), P. 1243 - 1243

Published: Dec. 9, 2024

Liver disease can significantly impact life expectancy, making early diagnosis and therapeutic intervention critical challenges in medical care. Imaging diagnostics play a crucial role diagnosing managing liver diseases. Recently, the application of artificial intelligence (AI) imaging analysis has become indispensable healthcare. AI, trained on vast datasets images, sometimes demonstrated diagnostic accuracy that surpasses human experts. AI-assisted are expected to contribute standardization quality. Furthermore, AI potential identify image features imperceptible humans, thereby playing an essential clinical decision-making. This capability enables physicians make more accurate diagnoses develop effective treatment strategies, ultimately improving patient outcomes. Additionally, is anticipated powerful tool personalized medicine. By integrating individual data with information, propose optimal plans for treatment, it component provision most appropriate care each patient. Current reports highlight advantages As technology continues evolve, advance treatments overall improvements healthcare

Language: Английский

Citations

1
Machine learning predicts liver cancer risk from routine clinical data: a large population-based multicentric study DOI Open Access
Jan Clusmann, Paul‐Henry Koop, David Zhang

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 4, 2024

Abstract Background and aims Hepatocellular carcinoma (HCC) is a highly fatal tumor, for which early detection risk stratification crucial, yet remains challenging. We aimed to develop an interpretable machine-learning framework HCC based on routinely collected clinical data. Methods leverage data obtained from over 900,000 individuals 983 cases of across two large-scale population-based cohorts: the UK Biobank study “All Of Us Research Program”. For all these patients, timepoints years before diagnosis was available. integrate modalities including demographics, electronic health records, lifestyle, routine blood tests, genomics metabolomics offer unique, multi-modal perspective risk. Results Our random-forest-based model significantly outperforms publicly available state-of-the-art risk-scores, with AUROC 0.88 both internal external test sets. demonstrate robustness our ethnic subgroups, major advance previous models variable performance by ethnicity. Further, we perform extensive feature-importance analysis, showcasing approach as framework. provide weights open-source web calculator facili-tate further validation model. Conclusion presents robust stratification, offers potential improve could ultimately reduce disease burden through targeted interventions. Lay summary Finding liver cancer crucial successful treatment. Therefore, screening abdominal ultra-sound can be performed. However, it not clear who should receive ultrasound screening, current standard only patients cirrhosis, severe disease, many are diagnosed in late stages. trained machine learning model, acting like decision trees at same time, detect high looking patterns almost 1000 population 900.000 individuals. In separate set has seen during training, worked better than models. Additionally, investigated 1. how comes its prediction, 2. whether works males females alike 3. most relevant Like this, help sort into categories “high-risk”, “medium-risk” “low-risk”, via strategies then decided, cancer.

Language: Английский

Citations

0