–News Direct–
PathAI, a global leader in AI-powered pathology, today announced it will present research on advances in artificial intelligence (AI) approaches through three posters at the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics on October 11-15 in Boston, Massachusetts, including one poster to be presented in partnership with Flare Therapeutics. The posters showcase how AI may improve the utility of complex and specialized modalities such as multiplexed immunohistochemistry and may enhance the utility of routine modalities, such as augmenting cellular resolution from H&E-stained images. The presentations underline the significant strides in AI that can enhance translational and clinical research applications.
Spatially-resolved prediction of gene expression signatures in H&E whole slide images using additive multiple instance learning models (Poster B010)
In this research, PathAI demonstrates the prediction of TGF-CAF gene expression signature levels in breast cancer from H&E images and links these levels to tumor microenvironment features using PathExplore, PathAIs recently launched structured, standardized and scalable panel of human interpretable features (HIFs) offering unprecedented resolution of the tumor microenvironment (TME) from H&E whole-slide images.
H&E staining is routine for cancer diagnosis but does not provide molecular information. This potentially limits its utility for molecular-targeted therapy development and selection. AI models augment the information that can be extracted from H&E staining, enhancing resolution of H&E data and increasing other applications such as associating spatial TME features with gene expression signatures.
End-to-end additive multiple instance learning (aMIL) models developed and deployed by PathAI performed well at predicting TGF-CAF levels. Importantly, aMIL models link PathExplore HIFs to specific sub-regions within the TME, which allows for granular understanding of specific cellular contributions to various molecular predictions from H&E WSI.
This poster adds to a series of PathAI achievements in predicting molecular phenotypes from digital pathology images, providing a means to harness complex biological signatures as clinical biomarkers for precision medicine. It will be presented by Chintan Parmar, manager of AI research at PathAI, on October 13 from 12:30-4 p.m. ET.
Artificial intelligence (AI) analysis of histological images accurately identifies luminal subtype Urothelial Carcinomas characterized by high Peroxisome Proliferator-Activated Receptor Gamma (PPARG) expression (Poster B016)
This research was completed in collaboration with Flare Therapeutics and highlights an AI-based model that identifies luminal subtype in Urothelial Carcinoma as a novel biomarker approach for its first-in-class clinical candidate FX-909.
The poster will be presented by Stefan Kirov from Flare Therapeutics on October 13 from 12:30-4 p.m. ET.
Machine-learning enabled quantification of colocalized multiplex IHC signals with spectral overlap (Poster B008)
This poster demonstrates novel imaging and AI technology to accurately detect and quantify up to four co-localized antigens or stains from a single multiplex IHC slide. Multi-spectral imaging addresses the challenge of how to separate markers exhibiting both spectral and spatial overlap. Expression or staining of individual markers is identified in spectrally unmixed images and combined with information across all separate marker channels to create a super annotation co-expression map that is then used to train AI models.
PathAIs model utilizing super annotation demonstrated a significantly lower error rate in detecting co-expression of several markers such as ER, PR, and Ki67 compared to an commercially available method, which was consistent on two different scanners, Leica At2 and 3DHistech.
These results may facilitate clinical utilization of higher-plex biomarkers, which could enable faster and more accurate co-expression measurement of various markers, promising cost and time savings by consolidating biomarker measurement to a single slide with automated scoring to assist pathologists.
This poster will be presented by Waleed Tahir, pathology AI research scientist at PathAI, on October 13 from 12:30-4 p.m. ET.
Learn More
PathAI representatives will be on site at the conference for the poster presentations:
Poster B008
- October 13, 12:30-4:00 p.m. ET
- Presenter: Waleed Tahir, PathAI
Poster B010
- October 13, 12:30-4:00 p.m. ET
- Presenter: Chintan Parmar, PathAI
Poster B016 (In partnership with Flare Therapeutics)
- October 13, 12:30-4:00 p.m. ET
- Presenter: Stefan Kirov, Flare Therapeutics
If interested in meeting the PathAI team during or after the conference, contact the team via email at BD@pathai.com.
About PathAI
PathAI is the only AI-focused technology company to provide comprehensive precision pathology solutions from wet lab services to algorithm deployment for clinical trials and laboratory use. Rigorously trained and validated with data from more than 15 million annotations, its AI-powered models can be leveraged to optimize the analysis of pathology samples to improve efficiency and accuracy of pathology interpretation, as well as to better gauge therapeutic efficacy and accelerate drug development for complex diseases.
PathAI, which is headquartered in Boston, MA, and operates a CAP/CLIA-certified laboratory in Memphis, TN, is proud to have a team of 600+ innovative thinkers from around the globe. For more information, please visit www.pathai.com.
Contact Details
SVM Public Relations and Marketing Communications
Maggie Naples
+1 401-490-9700
Company Website
View source version on newsdirect.com: https://newsdirect.com/news/pathai-announces-research-presentations-at-aacr-nci-eortc-international-conference-on-molecular-targets-and-cancer-therapeutics-827400279
PathAI
COMTEX_441742964/2655/2023-10-11T10:00:06
Disclaimer: The views, suggestions, and opinions expressed here are the sole responsibility of the experts. No A News Week journalist was involved in the writing and production of this article.