AI has found a wider use in healthcare, not only for rapid drug development
In the life science community, the way AI can speed up drug research, allowing big pharma companies and emerging biotechnology to more effectively discover new molecules to advance into clinical testing. But drug discovery alone won’t lead to more drugs or even faster drug development, said Liz Beatty, chief strategy officer at clinical trial technology startup.
No matter how fast a drug is discovered, it must be tested in humans. Beatty's experience includes a 16-year clinical trial at Bristol Myers Squibb, who said more than 80% of clinical trials missed the schedule due to enrollment issues. The clinical trials of drug development remain very human-dependent. Reviewing charts and lab reports, often hundreds of pages, has been working manually, Beatty said. Inato's technology platform uses AI to automate the process. Humans still make final decisions about whether patients meet the criteria for clinical trials, but the technology reduces the time it took to the past several hours to minutes
“In fact, we can speed up research by using AI in this part of the ecosystem, which historically has been a pain point that cannot be solved until new advances in AI,” Beatty said.
Beatty's comments were held during this week's panel discussion, with Medcity News' investment conference in Chicago. Chelsea Vane, vice president of product management at GE Healthcare, and Bobby Reddy, co-founder and CEO of Prenosis, also joined her. Michelle Hoffman, executive director of the Chicago Biomedical Consortium, chaired the group “How to Reshape the Healthcare Industry.”
AI is not only a tool for drug discovery and clinical trials. Technology that incorporates AI is already in contact with patients. Prenosis has commercial technology that guides clinicians to diagnose sepsis, a dangerous immune system response to infection. Sepsis can cause inflammation and organ damage and can be life-threatening. Historically, diagnosis has been a human effort, conducted through physician reviews of clinical discoveries and laboratory testing.
Prenosis's technology, sepsis, combines different types of data such as vital signs, standard laboratory tests, demographic information and biomarkers. AI analyses these data to give clinicians a deeper understanding of patient biology. Due to the nature of sepsis, this approach is necessary. Reddy said it is not a syndrome, but a disease, but a collection of various diseases.
Last year, the FDA was granted a relapse authorization by the Department of Sepsis Immunization. Reddy says the technology. The traditional way of diagnosing sepsis depends on human judgment and experience, which varies from clinician to clinician, and Prenosis's technology makes the diagnosis of sepsis more consistent.
“It’s more standardized, it’s based on thousands of patients from the past,” Reddy said. “So it’s more accurate, more unified, more realistic.”
For GE Healthcare, AI has the effect of increasing patient access to care. Vane points to Air Recon DL, a deep learning image reconstruction technique for MRI. This technology eliminates noise and distortion in the image, resulting in faster and clearer images. Vanne said the Air Reconnaissance DL accelerates scanning time by up to 50%. Therefore, more scans can be performed and clinicians can support more patients. While Air Recon DL is specifically for MRI, GE Healthcare also has AI applications for CT scans.
GE Healthcare also uses AI to improve cancer care. The company’s CareIntlect Oncology is an application that aggregates different types of patient data from different sources, such as medical images and electronic medical records, and provides a single vision for clinicians. With this technology, clinicians no longer need to jump between systems to get the full story of a patient’s medical history, reducing the time by several minutes, Vine said. In addition to summarizing complex medical history, the application can also help evaluate patients’ clinical trial qualifications.
“By summarizing all this multimodal data into a single unified view and then summarizing using AI, we are actually able to reduce the speed of speeding up that patient and increase the time the provider can spend with that patient,” Vane said.
Photo: Nick Fanion, breaking the media
