Tumor Protocol Assistant
Assist vMTB in determining the most appropriate treatment protocol based on patients criteria and guidelines.
- Model retrieves the latest guidelines from NCCN, ASCO, Cancer.gov.
- Extracts, understands and breaks data into independent units.
- Stores data in individual data units that are semantically identified and labelled.
- Builds meaningful and weighted relationships between these various data units.
- Index’s and stores the data in a second database in numerical format.
- Model receives all patient data and extracts relevant information.
- We use a proprietary approach to query the database using the patient extracted data.
- The query returns multiple responses which the model evaluates independently.
- The model review responses and determines the most appropriate treatment protocol
- The model submits the chosen Treatment Protocol along with all relevant references from guidelines sources.
- Model reviews the returned protocol, checks it for faithfulness to guidelines and gives it a score of 0-1.
- The microservice architecture is containerized and orchestrated
- Ongoing remote monitoring of results and reporting
- Continuous model updates, retraining and improvement
- Automated data updates with any changes in guidelines
- Our platform is ideal for medical treatment recommendations, research and development and clinical trials.
- Clinical trial AI module for cancer research will be offered.
- Tools and methodologies used in our platform are also used by major healthcare institutions for similar use cases.
The same tools and methodology are being used in the following Personalized Medicine for Cancer Applications:
- Lung Cancer Property Graph Database: This model allows for personalized treatment strategies by analyzing the complex interrelationships among genetic mutations, clinical histories, and treatment outcomes.
- Gene Co-expression Networks: construct and analyze gene co-expression networks using RNA-seq data from The Cancer Atlas Genome (TCGA). By creating networks for various cancer types, researchers can apply graph algorithms to identify significantly altered genes and relationships in cancer networks. This approach supports personalized medicine by uncovering tumor-specific expression programs and potential therapeutic targets.
- Hetionet: This project models drug efficacy and interactions by connecting data on treatments, compounds, diseases, and pathways. Hetionet’s supports research into drug repurposing and the identification of new therapeutic targets in oncology.
