Tumor biology and immune infiltration define primary liver cancer subsets linked to overall survival after immunotherapy DOI Creative Commons

Anuradha Budhu,

Erica C. Pehrsson, Aiwu Ruth He

et al.

Cell Reports Medicine, Journal Year: 2023, Volume and Issue: 4(6), P. 101052 - 101052

Published: May 23, 2023

Primary liver cancer is a rising cause of deaths in the US. Although immunotherapy with immune checkpoint inhibitors induces potent response subset patients, rates vary among individuals. Predicting which patients will respond to great interest field. In retrospective arm National Cancer Institute Cancers Liver: Accelerating Research Immunotherapy by Transdisciplinary Network (NCI-CLARITY) study, we use archived formalin-fixed, paraffin-embedded samples profile transcriptome and genomic alterations 86 hepatocellular carcinoma cholangiocarcinoma prior following inhibitor treatment. Using supervised unsupervised approaches, identify stable molecular subtypes linked overall survival distinguished two axes aggressive tumor biology microenvironmental features. Moreover, responses treatment differ between subtypes. Thus, heterogeneous may be stratified status indicative inhibitors.

Language: Английский

A Survey of Deep Learning for Alzheimer’s Disease DOI Creative Commons
Qinghua Zhou, Jiaji Wang, Xiang Yu

et al.

Machine Learning and Knowledge Extraction, Journal Year: 2023, Volume and Issue: 5(2), P. 611 - 668

Published: June 9, 2023

Alzheimer’s and related diseases are significant health issues of this era. The interdisciplinary use deep learning in field has shown great promise gathered considerable interest. This paper surveys literature to disease, mild cognitive impairment, from 2010 early 2023. We identify the major types unsupervised, supervised, semi-supervised methods developed for various tasks field, including most recent developments, such as application recurrent neural networks, graph-neural generative models. also provide a summary data sources, processing, training protocols, evaluation guide future research into disease. Although promising performance across studies tasks, it is limited by interpretation generalization challenges. survey provides brief insight these challenges possible pathways studies.

Language: Английский

Citations

22
Artificial Intelligence and liver: Opportunities and barriers DOI Creative Commons
Clara Balsano, Patrizia Burra, Christophe Duvoux

et al.

Digestive and Liver Disease, Journal Year: 2023, Volume and Issue: 55(11), P. 1455 - 1461

Published: Sept. 16, 2023

Artificial Intelligence (AI) has recently been shown as an excellent tool for the study of liver; however, many obstacles still have to be overcome digitalization real-world hepatology. The authors present overview current state art on use innovative technologies in different areas (big data, translational hepatology, imaging, and transplant setting). In clinical practice, physicians must integrate a vast array data modalities (medical history, laboratory tests, pathology slides) achieve diagnostic or therapeutic decision. Unfortunately, machine learning deep are far from really supporting clinicians real life. fact, accuracy any technological support no value medicine without clinicians. To make better new technologies, it is essential improve clinicians’ knowledge about them. this end, propose that collaborative networks multidisciplinary approaches will rapid implementation AI systems developing disease-customized AI-powered decision tools. also discuss ethical, educational, legal challenges build robust bridges deploy potentially effective settings.

Language: Английский

Citations

17
The future of artificial intelligence in digital pathology – results of a survey across stakeholder groups DOI Creative Commons

Céline N. Heinz,

Amelie Echle, Sebastian Foersch

et al.

Histopathology, Journal Year: 2022, Volume and Issue: 80(7), P. 1121 - 1127

Published: April 4, 2022

Artificial intelligence (AI) provides a powerful tool to extract information from digitised histopathology whole slide images. During the last 5 years, academic and commercial actors have developed new technical solutions for diverse set of tasks, including tissue segmentation, cell detection, mutation prediction, prognostication prediction treatment response. In light limited overall resources, it is presently unclear researchers, practitioners policymakers which these topics are stable enough clinical use in near future still experimental, but worth investing time effort into.To identify potentially promising applications AI pathology, we performed an anonymous online survey 75 computational pathology domain experts academia industry. Participants enrolled 2021 were queried about their subjective opinion on appealing subfields with focus upon solid tumours. The results this indicate that response directly routine slides regarded as most application. This item was ranked highest analysis subgroups by age professional background. Furthermore, genetic alterations, gene expression survival images scored consistently high throughout subgroups.Together, data demonstrate possible direction development systems clinical, industrial research future.

Language: Английский

Citations

26
Deep learning-based quantification of NAFLD/NASH progression in human liver biopsies DOI Creative Commons
Fabian Heinemann, Peter Groß,

Svetlana Zeveleva

et al.

Scientific Reports, Journal Year: 2022, Volume and Issue: 12(1)

Published: Nov. 10, 2022

Non-alcoholic fatty liver disease (NAFLD) affects about 24% of the world's population. Progression early stages NAFLD can lead to more advanced form non-alcoholic steatohepatitis (NASH), and ultimately cirrhosis or cancer. The current gold standard for diagnosis assessment NAFLD/NASH is biopsy followed by microscopic analysis a pathologist. Kleiner score frequently used semi-quantitative progression. In this scoring system features active injury (steatosis, inflammation, ballooning) separated fibrosis are quantified. procedure time consuming pathologists, scores have limited resolution subject variation. We developed an automated deep learning method that provides full reproducibility higher resolution. was established with 296 human biopsies tested on 171 pathologist ground truth scores. inspired way pathologist's analyze biopsies. First, analyzed microscopically relevant histopathological features. Subsequently, aggregated per-biopsy score. Scores in identical numeric range as ballooning, steatosis, scores, but continuous scale. Resulting (quadratic weighted Cohen's κ test set: steatosis 0.66, inflammation 0.24, ballooning 0.43, 0.62, activity (NAS) 0.52. Mean absolute errors 0.29, 0.53, 0.61, 0.78, NAS 0.77).

Language: Английский

Citations

23
Tumor biology and immune infiltration define primary liver cancer subsets linked to overall survival after immunotherapy DOI Creative Commons

Anuradha Budhu,

Erica C. Pehrsson, Aiwu Ruth He

et al.

Cell Reports Medicine, Journal Year: 2023, Volume and Issue: 4(6), P. 101052 - 101052

Published: May 23, 2023

Primary liver cancer is a rising cause of deaths in the US. Although immunotherapy with immune checkpoint inhibitors induces potent response subset patients, rates vary among individuals. Predicting which patients will respond to great interest field. In retrospective arm National Cancer Institute Cancers Liver: Accelerating Research Immunotherapy by Transdisciplinary Network (NCI-CLARITY) study, we use archived formalin-fixed, paraffin-embedded samples profile transcriptome and genomic alterations 86 hepatocellular carcinoma cholangiocarcinoma prior following inhibitor treatment. Using supervised unsupervised approaches, identify stable molecular subtypes linked overall survival distinguished two axes aggressive tumor biology microenvironmental features. Moreover, responses treatment differ between subtypes. Thus, heterogeneous may be stratified status indicative inhibitors.

Language: Английский

Citations

15