AIpredict: characterization of a deep learning tool to predict molecular features and clinical outcomes directly from routine breast cancer histopathology digital images
Project name: AIpredict: characterization of a deep learning tool to predict molecular features and clinical outcomes directly from routine breast cancer histopathology digital images
Call: 09I03-03-V03 - Large projects for excellent researchers
Investment: 3. Excellent science
Component: 9. More effective Management and strengthening of Research, Development and Innovation funding
Type of action: Recovery and Resilience Plan of the Slovak Republic 2021 - 2027 (RRP, Component 9)
Granting authority: Research Agency
Project starting date: fixed date: 01/2024
Project end date: 06/2026
Project duration: 30 months
Estimated budget: 3 049 874,50 €
Abstract
With nearly 360k breast cancer (BCa) cases in 2020, BCa is the most frequently diagnosed cancer in EU-27 countries, accounting for 28.7% of all new cancer cases in women and leading to a shocking 93k mortalities in 2020 alone. These more challenging BCa cases pose the greatest challenge for society and healthcare systems.
To solve this problem, new molecular diagnostics are needed to alleviate diagnostic irregularities, identify vulnerable patients, and provide them with targeted and effective therapy.
In this project, we propose to train artificial intelligence (AI) to predict molecular properties and outcome results directly from common hematoxylin and eosin preparations. To accomplish this, target our unique data set of 1082 breast tumors with extensive molecular and clinical data.
To determine the parameters of the AI model then we will use orthogonal methods including spatial multiomics. This innovative diagnostic tool is an ideal approach to rapidly identify high-risk BCa patients with a greater probability of relapse or stop responding to treatments and triage them for scrutiny with more tailored precision oncology care.
In conclusion, we will perform external validations to ensure a robust AI stability model. This AI model, called AIpredict, has the potential to cause a revolution in the early diagnosis of breast cancer, reduce the socio-economic burden, and save patients' lives.
Related news
https://www.multiplexdx.com/products?tab=breast-cancer-diagnostics
Funded by the EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia under the project No. 09I03-03-V03-00101.
About MultiplexDX
MultiplexDX is one of the most innovative biotech corporations, created to bring its revolutionary technologies to the market of personalized molecular diagnostics. The company has representation in both U.S. and European markets. The collaborators of MultiplexDX are from the world’s most prestigious scientific organizations including the National Cancer Institute, Rockefeller University, Albert Einstein University, Vanderbilt University, Cornell University, Queens University (Canada), Hebrew University of Jerusalem (Israel), and the Max Delbrück Center for Molecular Medicine (Germany).
MultiplexDX IP-based and innovative platform merges histopathology methods, biomarker quantification, visualization and gene expression with a single-cell resolution by combining MDX proprietary visual and sequencing technologies into one diagnostic test. This cross-validation approach eliminates diagnostic errors and creates 100% precise cancer profiling for each patient which allows clinicians to suggest specific, personalized cancer treatment.
Find out more about MultiplexDX on Corporate website, Facebook, LinkedIn, Twitter
