This article summarizes the main takeaways of the CHI's whitepaper "Machine Learning and Medical Devices Connecting practice to policy (and back again)".
The statistical significance of your model evaluation depends on the size of the test set as well as the required model performance. In this article, we show how to water-proof your model evaluation.
Reuse is a wonderful concept to reduce time, effort, and cost. In this article, we discuss the regulatory requirements for pre-trained deep neural networks.
"I'm thrilled about the potential of machine learning for next-generation patient care and eager to help to bring it to the medical devices market."
We are a small but mighty, owner run business with profound hands-on expertise in the areas of machine learning, software validation, and regulatory affairs. Our passion is to help make latest machine learning technology usable in a commercial environment with all its market and regulatory requirements.
The primary interest and focus of our customers – ranging from cutting edge start-ups in the medical device industry to leading technology providers and global pharmaceutical company – is to develop innovative products with state-of-the-art benefit for the patients. Our contribution is to help bring these products into the market fast, safely, and smoothly.
We know that relying on someone’s support in a such critical area requires trust, and that trust must grow over time. The best way to get to know us, our mode of operation, and the value we can bring to the table, is to get in touch with us. Together, we will figure out if our machine learning validation solutions are a good fit to your business needs. We will be happy to get back to you if you drop us a short note via our contact form.
We work very hard to make our customers happy - read what they have to say below.
"I took part in the one-day workshop Developing AI applications in compliance with the regulatory requirements and was delighted. Despite the online format, it was a very lively event with a dynamic discussion and active involvement of the participants. Prof. Haase understood how to present even very abstract contexts in a comprehensible way and how to break them down to regulatory tasks. It was noticeable that he was really interested in answering our questions and pointing out solutions to problems."
"A very nice team, which flexibly adapts to customer requirements and has convinced with high expertise and practical experience. Through their expert support we are ideally prepared for the audit regarding the validation of Machine Learning."