In what's being hailed as a significant boon for oncology, the Oak Ridge National Laboratory (ORNL) has made a breakthrough that could expedite cancer research and response times. Their efforts, backed by the Department of Energy, reduced the lag between cancer diagnosis and pathology report processing from 22 months to a markedly improved 14 months. This reduction is largely due to an advanced artificial intelligence technology devised by ORNL's team, as detailed by ORNL News.
The National Cancer Institute's Division of Cancer Control and Population Sciences (DCCPS) has announced this progress, emphasizing the potential for more rapid identification and action on cancer trends nationwide. Cancer registries, which have traditionally been updated by manually sifting through pathology reports, have often harbored a two-year reporting gap, a discrepancy that can delay critical responses to shifts in cancer occurrences. As Heidi Hanson, the senior research scientist who leads the Biostatistics and Biomedical Informatics group at ORNL, noted, the technology resembles a large language model such as ChatGPT but tailored for decoding medical texts. "We can reduce the time it takes to process reports by using artificial intelligence to autocode records," Hanson told ORNL News.
Central to this acceleration in report processing is the Information Extraction API, or OncoIE, an ORNL-developed deep learning algorithm. This AI application can extract vital data from an extensive array of unstructured medical documentation, facilitating a rapid or near real-time response. Hanson hailed the OncoIE algorithm as particularly forward-thinking, stating, "In my opinion, (OncoIE) is really ahead of its time in terms of what can be done with AI in medicine," in a statement obtained by ORNL News.
Highlighted at a function to celebrate the retirement of NCI's Lynne Penberthy, this novel use of AI in cancer reporting holds promise for more effective disease prevention tactics. "When you're 22 months behind in even knowing what's happening on the ground, you can’t prevent it," Hanson asserted. At the event, it was also revealed that the team is not resting on their laurels, with aims to further slash the reporting time down to as little as two months. Hanson continued, conveying the gravity of such goals: "We want to identify changes in disease trends early on so that we can stop whatever's happening and reduce the number of cancer diagnoses." UT-Battelle, ORNL's managing organization, remains committed to addressing pressing scientific challenges through such innovations.
