Currently, there are 569 clinical trials focused on developing in vitro diagnostic (IVD) devices in oncology out of 1,490 active clinical trials. Out of the total IVD trials, nine are being conducted to test partial or full analysis by artificial intelligence (AI) to improve oncology diagnosis. AI-driven machine learning is pushing the advancement of IVD device development in terms of accuracy and reliability. Considering the advantages provided by AI, the incorporation of this technology in oncology IVD products is expected to result in advancements in cancer diagnosis, according to GlobalData, a leading data and analytics company.
Recently, AI pathology provider Mindpeak and Porscia, a computational and digital solutions provider, partnered to improve cancer diagnosis with AI-powered workflows to help pathologists make more efficient clinical decisions. This is done by AI analysis of digital pathology images from glass slides with biopsy pathology samples. Both Mindpeak’s breast cancer cell detection software, BreastIHC, and Poscia’s digital diagnostic device, Concentriq Dx, have their CE-IVD and CE-IVDR marks and thus helping pave the way for similar technologies to reach the public.
Selena Yu, Medical Analyst at GlobalData, comments: “Some of the active oncology clinical trials with AI are focusing on using AI to optimise workflows, such as which patients require additional analysis and prioritizing severe cases. Another use for AI in oncology IVD devices is to predict which treatment responses would be the most effective using patient samples.”
The trial “An Observational Study to Evaluate the Clinical Utility of the Oncomine Precision Assay within the Exactis Network” is using AI to assess treatment responses in patient samples. Another study, “Artificial Intelligence Neuropathologist” is currently testing their AI’s ability to detect central nervous system (CNS) tumors unsupervised and fully automated.
Yu concludes: “These advancements will push more manufacturers to partner with AI algorithm providers to improve their existing cancer diagnostic software. This, in turn, will not only decrease the time needed for diagnosis feedback to patients and early diagnosis with more sensitive software, but also provide effective treatment options for patients and decrease the workload for healthcare professionals.”
