In April 2019, FDA released a draft guidance providing manufacturers and FDA staff with detailed recommendations on assessing the technical performance of quantitative imaging devices and how the documentation from those assessments should be provided in premarket submissions. From a big picture perspective, one should remember the overall goal is to “provide performance specifications for the quantitative imaging functions, supporting performance data to demonstrate that the quantitative imaging functions meet those performance specifications, and sufficient information for the end user to obtain, understand, and interpret the values provided by the quantitative imaging functions.”
The guidance document establishes consistency terminology to be used when describing quantitative imaging functions. We recommend that your documentation use this terminology to best communicate your product to FDA. If your documentation uses other terms, provide a “terminology map” right up front in the submission so that the reviewer will correctly understand your documentation.
The guidance provides a section on potential sources of measurement error that would be very useful for risk management. We would recommend that your hazard analysis activity consider all of the categories from the guidance and use it as a model for further hazard discovery. One should also review recalls related to similar devices to learn of possible hazards and failure modes. Our website provides a collection of software related recalls to aid in that activity. To see a listing of posts related to imaging recalls, just click this link: https://www.softwarecpr.com/?s=imaging
The guidance emphasizes the need for predefined performance specifications. Like all requirements and specifications, one should consider how an objective evaluation of the requirement or specification will be performed. From a software perspective, we recommend methods to have the software provide additional outputs, not normally provided to an operator, but necessary for proper evaluation. For example, the software might output interim calculations and components of the larger calculation that can be evaluated separately, or perhaps statistical parameters for a set of data. This type of output can help support the argument for software verification required by the guidance.
Download the guidance here: FDA Quantitative Imaging Guidance draft – April 2019