Healthcare Startup Uses AI to Analyze Cancer Cells in the Operating Room
Explore how Invenio Imaging's AI accelerates cancer cell analysis in operating rooms, offering insights in minutes instead of weeks.
Invenio Imaging is developing technology that empowers surgeons to assess tissue biopsies in the operating room immediately after they are collected. This innovation is groundbreaking because it provides AI-accelerated insights in a mere three minutes, a process traditionally taking weeks in a pathology lab.
This compact, AI-powered imaging system is designed to support swift clinical decision-making right within the treatment room. Chris Freudiger, the Chief Technology Officer of Invenio, based in Silicon Valley, believes this technology will aid surgeons in making intraoperative decisions effectively during biopsies or surgeries. It enables the rapid evaluation of whether the tissue sample contains cancerous cells and, with the AI models Invenio is developing, potentially allows for a molecular diagnosis for personalised medical treatment within minutes.
This quicker diagnosis is crucial, particularly for aggressive types of cancer that could proliferate or metastasise significantly in the weeks it usually takes for biopsy results to return from a dedicated pathology lab. Invenio is part of NVIDIA Inception, a programme that offers state-of-the-art startups with technological support and AI platform guidance, accelerating AI training and inference using NVIDIA GPUs and software libraries.
Invenio’s NIO Laser Imaging System is a revolutionary digital pathology tool that accelerates the imaging of fresh tissue biopsies. It has been utilised in thousands of procedures across the U.S. and Europe and received the CE Mark of regulatory approval in Europe in 2021. The company plans to integrate the NVIDIA Jetson Orin series of edge AI modules for its next-generation imaging system, which will feature near real-time AI inference accelerated by the NVIDIA TensorRT SDK.
Invenio provides diagnostic images and an analysis of what physicians are observing, thanks to a layer of AI models built on top of their imaging capabilities. With the AI performance provided by NVIDIA Jetson at the edge, physicians can quickly determine the types of cancer cells present in a biopsy image. Invenio employs a cluster of NVIDIA RTX A6000 GPUs to train neural networks with tens of millions of parameters on pathologist-annotated images, developed using the TensorFlow deep learning framework and trained on images acquired with NIO imaging systems.
One of Invenio’s remarkable AI products, NIO Glioma Reveal, is approved for clinical use in Europe and available for research use in the U.S. to help identify areas of cancerous cells in brain tissue. Collaborators from renowned universities have developed a deep learning model to find biomarkers of cancerous tumours with 93% accuracy in 90 seconds. This capability to analyse different molecular subtypes of cancer within a tissue sample enables doctors to predict a patient's response to chemotherapy or determine whether a tumour has been successfully removed during surgery.
Beyond brain tissue analysis, Invenio has announced a clinical research collaboration with Johnson & Johnson’s Lung Cancer Initiative to develop and validate an AI solution to help evaluate lung biopsies. Lung cancer is the deadliest form of cancer globally, with lung nodules found in over 1.5 million patients yearly in the U.S. alone. Invenio’s NIO Lung Cancer Reveal tool aims to reduce the time needed to analyse tissue biopsies for these patients once approved for clinical use.
Source: NVIDIA Blog