Human Dx is committed to providing access to relevant datasets to any researcher employed by an accredited academic institution for the express purpose of independently verifying and reproducing the results presented in our prior research.

This is to safeguard against benchmark contamination for evaluating AI and human-AI collaborative systems (see Balloccu et al., 2024; Sun et al., 2025), as well as due to privacy considerations associated with clinical case vignettes potentially representing patients seen in clinical practice as well as performance data of medical professionals and trainees providing diagnoses for those vignettes.

Thus, researchers must agree to use the dataset solely for validation purposes and also agree not to input the data into public versions of commercial artificial intelligence models that use such input for training purposes.

If you would like to gain access to the data for the scenario outlined above, or if you are instead looking to partner with us for new research efforts, please fill out the contact form below.