VisualDx Augmented Intelligence Project

Industry Project

VisualDX is the most widely used diagnostic support system in the world. VisualDX combines machine learning and vision science with structured clinical data and clinical question sets to let non-specialist healthcare providers (1) enter patient specific findings, (2) to build custom differentials, and then (3) view images, treatment recommendations and management information. The current project would create a stand-alone Android version of VisualDX that will allow for use of the platform with no required internet connection and expand its content set. The goal is to empower non-specialist providers to quickly and accurately answer clinical questions to deliver the best care.

For NASA, this project brings a medical decision support tool that can function on deep-spaceflight missions or any other environment without access to the Internet. The VisualDx module developed for this project includes over 100 conditions on the Exploration Medical Condition List (MCL), and the infrastructure built can support the addition of more conditions and more modules in the future, including an ultrasound module currently being developed as part of another Translational Research Institute for Space Health (TRISH)-funded project (IND0009--"VisualDx Clinical Decision Support for Ultrasound"). The machine learning model and AI algorithm have been trained on spaceflight-specific dermatologic conditions that are the most common in spaceflight. As part of this project, VisualDx also improved upon NASA's MCL by expanding upon the listed diagnosis concepts to include more specific conditions. For example, while the MCL included a singular concept for "rash," VisualDx contributors were able to split that concept into over 40 specific dermatologic conditions relevant to spaceflight. For example, on the original MCL the heat rash diagnosis was not differentiated from eczema, yeast infection, or abscess. VisualDx recognized the opportunity to make this improvement early on in the grant period and was able to include much more granular concepts in the end product as a result. Further, the development of the stand-alone, space-specific version of VisualDx brings value to NASA by: - providing medical decision support in the absence of timely communication between Earth and crew members. - providing such medical decision support in a software package that is likely to run on existing computing equipment, thus requiring no additional physical storage space. - increased support and confidence in medical decision making by non-medical professionals staffing deep spaceflight missions, augmenting existing medical training protocols.

The completion of this project has significant applications on Earth as well. The development of the stand-alone version enables deployment of the application in remote locations or settings with poor Internet infrastructure, making clinical decision support available to large populations previously unable to access such support. Additionally, the development to support non-specialist terminology and synonyms for medical terms makes the product more accessible to new user populations, instead of supporting only skilled clinical providers.

This is early/mid-stage (TRL 4) — the most realistic path in is NASA SBIR/STTR, which funds small businesses and research institutions to develop technology aligned with NASA's needs (equity-free, phased funding). Check whether a current SBIR/STTR solicitation topic overlaps with this project's technology area, or contact the project directly (above) to ask.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.