Artificial Intelligence based biomarkers for breast cancer
Principal Investigator
Name
Nameeta Shah
Degrees
PhD
Institution
Amaranth Medical Analytics
Position Title
CEO
Email
About this CDAS Project
Study
PLCO
(Learn more about this study)
Project ID
PLCOI-1375
Initial CDAS Request Approval
Oct 31, 2023
Title
Artificial Intelligence based biomarkers for breast cancer
Summary
The aim of this study is development of a decision support system based on Artificial Intelligence models derived from histology and radiology data for breast cancer care.
Among all cancers, breast cancer has the highest incidence among Indian females (25.8 per 100,000 women). Due to delayed detection the mortality rate is also very high of 12.7 per 100,000 women. There is a need for better prognostic and predictive biomarkers for improved treatment plan for these patients.
Large-scale molecular analysis of breast cancer tissues have resulted in identification of different molecular subtypes like Basal-like, Her2-enriched, luminal A and luminal B. Depending on the hormone receptor status and molecular subtype data an appropriate treatment plan is devised. Genomic tests like The Oncotype DX Breast Recurrence Score Test or CanAssist test scores are used to make decisions about administration of chemotherapy. These tests can be costly and are not available to all women. Also, the treatment options and companion diagnostic tests are limited for late stage breast cancer. We want to develop Artificial Intelligence based cost-effective tests that take as input routine clinical tests like histopathology slides, biomedical imaging and other clinical parameters like age, weight, etc. to predict treatment and outcome for all different stages of breast cancer. Given that a large number of cases are detected at late stage in Indian women, it is imperative to identify biomarkers for this cancer that are affordable and applicable in the Indian context.
Among all cancers, breast cancer has the highest incidence among Indian females (25.8 per 100,000 women). Due to delayed detection the mortality rate is also very high of 12.7 per 100,000 women. There is a need for better prognostic and predictive biomarkers for improved treatment plan for these patients.
Large-scale molecular analysis of breast cancer tissues have resulted in identification of different molecular subtypes like Basal-like, Her2-enriched, luminal A and luminal B. Depending on the hormone receptor status and molecular subtype data an appropriate treatment plan is devised. Genomic tests like The Oncotype DX Breast Recurrence Score Test or CanAssist test scores are used to make decisions about administration of chemotherapy. These tests can be costly and are not available to all women. Also, the treatment options and companion diagnostic tests are limited for late stage breast cancer. We want to develop Artificial Intelligence based cost-effective tests that take as input routine clinical tests like histopathology slides, biomedical imaging and other clinical parameters like age, weight, etc. to predict treatment and outcome for all different stages of breast cancer. Given that a large number of cases are detected at late stage in Indian women, it is imperative to identify biomarkers for this cancer that are affordable and applicable in the Indian context.
Aims
1. Automated segmentation of breast cancer histology images
2. Cellular phenotyping of breast cancer WSI images
3. AI-based model for prediction of treatment response
Collaborators
NA