iGrok builds software and hardware/software solutions for image-guided medical devices, from classical computer vision to modern machine learning. Exploratory research is fast and broad-based, while production-oriented development is driven within your quality system or the iGrok QMS, built for life-critical products.
Our products rarely carry the iGrok name. We measure success by how proud our partners are to put their own name on our work, from university spin-offs to multinational device makers.
Projects start near the finish line. Our continuously refreshed libraries in imaging analysis, registration, treatment planning and workflow give new development a real head start, and we still benchmark against the state of the art.
Engineering, research, project management, quality assurance and technical communication, staffed by people with deep experience in critical healthcare systems and the domain knowledge to get traction from day one.
A quality management system designed to work our way, your way, or a hybrid. That covers everything from start-ups new to device regulation to multinationals with highly regimented processes.
"iGrok" comes from image understanding. Our work combines image processing, computer vision, machine learning, rapid search and optimization to turn torrents of multi-modal data, from CT, MR and ultrasound to endoscopic video, into information clinicians can act on in seconds.
We deliver across desktop, embedded, web and mobile platforms, whichever vehicle best fits the device and the clinical workflow.
Image understanding has always been machine intelligence: segmentation, registration, classification, and learned models of anatomy and motion. The methods have moved from hand-built features to deep networks and foundation models, and they will keep moving. Medical device R&D still needs the same four things from them: an efficient path to an optimal solution, evidence that it works, a clear account of how it may fail, and a design that keeps patients safe.
AI tools in the R&D process and AI techniques in the end product both move fast: in capability, in expectations, and in regulatory requirements. We continually tune our toolsets and workflows and track guidance as it changes, so we reach safe, effective product quickly, with submission strategies that hold up.
High-quality medical software and hardware demand an enthusiastic focus on user experience and exceptional attention to detail, verification and validation. We're committed to a quality process that our clients can see in our work product, and rely on in theirs.
iGrok engages flexibly: entirely within our own Quality Management System, within the framework of yours, or in a hybrid.
Quality is never an accident; it is always the result of high intention, sincere effort, intelligent direction and skillful execution; it represents the wise choice of many alternatives.William A. Foster
iGrok grew from new treatment-time imaging in radiation therapy. That imaging promised to transform how physicians track clinically significant change over a course of treatment, but it buried them in gigabytes of data. Our first tools automatically analyzed and summarized those images so physicians could review weeks of data in seconds.
The technology was first demonstrated in 2004, and the company was founded in 2006 in Sewickley, Pennsylvania, just outside Pittsburgh. Since then our work has spread from cancer care into neurosurgery, urology and a growing range of endoscopic applications, and we draw on sharp minds wherever we find them.
"Grok," to understand completely, comes from Heinlein's Stranger in a Strange Land.
Flexible hours, mostly remote, on medical R&D that reaches patients. We're always interested in researchers, developers and interns in computer vision and machine learning.
Whether you're scoping a new product, need imaging expertise on an existing program, or want a quality-minded development partner, we'd be glad to hear from you.
Email us