Nat Commun April 2023 PMID 37055393

Histopathology images predict multi-omics aberrations and prognoses in colorectal cancer patients.

Pei-Chen Tsai, Tsung-Hua Lee, Kun-Chi Kuo, Fang-Yi Su, Tsung-Lu Michael Lee, Eliana Marostica, Tomotaka Ugai, Melissa Zhao, Mai Chan Lau, Juha P Väyrynen, Marios Giannakis, Yasutoshi Takashima, Seyed Mousavi Kahaki, Kana Wu, Mingyang Song

Full text PDF Download social card

Mentions

Peak: 4 in Apr 2023

across 4 sources since Apr 2023

Extended reach

32,074

potential audience

Total likes

Video views

Mentions per month

Apr 2023 – Jul 2023

Where it's discussed

News28 articles
X / Twitter6 posts
Blogs3 articles
YouTube1 videos

AI synopsis

What people are saying

The discussions highlight the promising use of machine learning and histopathology images for predicting colorectal cancer outcomes, with comments like 'new AI tool' and hashtags like #MachineLearning and #AI emphasizing excitement about technological advancements. The tone is optimistic yet cautious, indicating recognition of potential impact while acknowledging ongoing development.

AgreementModerate agreement

Most posts express a positive view, highlighting the potential of the AI tool to predict outcomes and genetic changes in colorectal cancer, indicating general agreement with the article's significance.

InterestHigh interest

Posts show high engagement with enthusiastic language and technical hashtags, demonstrating strong interest in the research and its applications.

EngagementModerate engagement

Discussions include mentions of AI, histopathology, and personalized treatment, reflecting a meaningful level of engagement beyond surface-level reactions.

ImpactModerate impact

Reactions suggest that the publication could influence future research and treatment approaches, though some posts are more aspirational than conclusive.

Abstract in brief

The study introduces the MOMA platform, a machine learning tool designed to analyze histopathology images of colorectal cancer (CRC) to predict patient outcomes and genetic changes, which are crucial for tailoring treatments.
MOMA can accurately forecast survival rates, genetic alterations, and other clinical features across diverse patient groups and imaging methods, demonstrating its robustness and broad applicability.
The approach also identifies specific histological patterns linked to genetic profiles and disease features, making the predictions more interpretable and potentially useful for guiding personalized treatment decisions.]

Read the full abstract

All mentions

Predicting Colorectal Cancer Outcomes Using Histopathology Images and Machine Learning

The study introduces the MOMA platform, a machine learning tool designed to analyze histopathology images of colorectal cancer to predict patient outcomes and genetic changes, crucial for personalized treatment planning.

STEM RTCL TV · Jul 29, 2023 78 views

Reiner 

@erlesen · X

RT @erlesen: Histopathology images predict multi-omics aberrations and prognoses in colorectal cancer patients https://t.co/6iZ70KWzCO New…

May 11, 2023 · 1 share · 7,853 followers View post →

Reiner 

@erlesen · X

May 10, 2023 · 5 likes · 1 share · 7,853 followers View post →

Science Advancement and Outreach

@SAOscience · X

Apr 14, 2023 · 4 likes · 1,237 followers View post →

La veille techno

@laVeilleTechno · X

RT @lopezunwired: Histopathology images predict multi-omics aberrations and prognoses in colorectal cancer patients https://t.co/HmgIO2XrBB…

Apr 13, 2023 · 1 share · 11,305 followers View post →

Paul Lopez

@lopezunwired · X

Histopathology images predict multi-omics aberrations and prognoses in colorectal cancer patients https://t.co/HmgIO2XrBB #MachineLearning #NatureJournal #AI https://t.co/6GX508hTGU

Apr 13, 2023 · 1 like · 1 share · 2,480 followers View post →

Reluctant Quant

@DrMattCrowson · X

RT Histopathology images predict multi-omics aberrations and prognoses in colorectal cancer patients https://t.co/q7SnGTVm3f https://t.co/pnSnRyAk90

Apr 13, 2023 · 1,260 followers View post →