AI workflow enhances precision of progression-free survival estimates in metastatic breast cancer, enabling more effective and personalized treatments

nference, a company dedicated to transforming healthcare by making biomedical data computable, will present research at the 2024 American Society of Clinical Oncology (ASCO) Annual Meeting, May 31 to June 4 in Chicago. The poster, titled A deep learning-enabled workflow to estimate real world progression-free survival in patients with metastatic breast cancer, will be presented on June 3, 9 a.m. to 12 p.m.

"Progression-free survival is a critical endpoint in oncology, and our deep learning-enabled workflow represents a significant advancement in how we evaluate real-world patient outcomes," said Maulik Nanavaty, President of Ventures at nference & CEO at Anumana. "This recent study demonstrates how AI and real-world data can drive innovation in oncology, enabling more precise and efficient cancer care and research."

Utilizing nference’s advanced natural language processing and structured EHR data from its federated platform, researchers analyzed data from 316 patients with metastatic breast cancer treated between 2015 and 2021 to estimate real-world progression-free survival rates. The study accurately captured progression events in clinical notes and radiology reports, closely matching expert manual reviews. Results showed 99.8% sensitivity and 98.2% accuracy at the sentence level, and 92.5% sensitivity and 88.0% accuracy at the patient level.

Researchers demonstrated that this innovative workflow, combining manual curation with AI algorithms, can accurately and quickly determine progression-free survival in breast cancer patients and could potentially improve the tracking of progression for various types of cancer and treatments.

A second abstract was accepted by ASCO for online publication, titled Clinical nSights: A software platform to accelerate real world oncology analyses.

nference's federated AI platform, nSights, provides researchers with access to insights from a curated, de-identified dataset spanning all therapeutic areas, including gastroenterology, cardiology, oncology, metabolic diseases, immunology, and infectious diseases. The platform makes the complete patient journey readily accessible for research, enabling breakthrough discoveries and collaborations as well as the creation of new venture companies that harness various data modalities.

For more information, visit www.nference.com.

About nference

nference is unlocking the potential of healthcare data with AI, delivering real-world evidence, insights, and solutions informed by the most comprehensive multimodal data across all therapeutic areas. Our extensive de-identified, longitudinal data includes decades of clinical notes along with rich patient histories, vitals, lab tests, radiology images, digital pathology, genomics, and electrophysiology waveforms. Our federated data platform, comprised of premier healthcare organizations, including Mayo Clinic, Duke Health, Banner Health, Vanderbilt University Medical Center, and Emory Healthcare, enables advanced research and custom AI model development. With unprecedented access to data from over 40 million patient journeys, our partners can accelerate target discovery, optimize clinical trial design, and enhance lifecycle management. Follow nference on LinkedIn. Visit us at www.nference.com.

View source version on businesswire.com: https://www.businesswire.com/news/home/20240531955525/en/