NCI-DOE Collaboration, MOSSAIC project (Modeling Outcomes using Surveillance Data and Scalable AI for Cancer)
Description
MOSSAIC applies deep learning natural language processing (NLP) and foundation models to population-based cancer data collected by NCI's Surveillance, Epidemiology, and End Results (SEER) program. DOE's Oak Ridge National Lab (ORNL) has data use agreements (DUAs) with multiple SEER registries to access and train models using SEER data. Two APIs are in production use in the data management system used by the SEER registries -- OncoID which predicts whether a pathology report is related to cancer and OncoIE which extracts key tumor characteristics from unstructured pathology report text. Together these APIs are an important part of moving the US towards near real-time cancer incidence reporting. In addition, a third API OncoMetsID, which predicts whether a pathology report is indicative of metastatic disease, is in a pilot phase to use in conjunction with other sources of information in the registries to identify recurrent disease.
Detailed example
Input: unstructured (free text) cancer pathology reports. Output: varies depending on the algorithm but generally a predicted class (eg, tumor site) and associated relative confidence score that can be used to tune accuracy
AI / analytics pattern
Natural Language Processing: AI that processes, interprets, and shares information in human language.
Automation level / stage
c) Deployed – The use case is being actively authorized or utilized to support the functions or mission of an agency.
Expected benefit
MOSSAIC enhances the infrastructure of the SEER cancer registries by providing tools that can increase the efficiency and accuracy of manual data abstraction by automatically extracting cancer surveillance data elements. SEER registries receive millions of unstructured clinical text documents that must be manually reviewed, leading to a lag in reporting of US cancer incidence trends. Automated tools such as those developed by MOSSAIC will help us achieve near real-time incidence trends and ultimately a more meaningful report card on the status of cancer in the US.
Controls / human review
ATO: Yes; PIA: Not published
Data needed
Data is owned by the NCI SEER registries, which are funded by the NCI