Skip to main content
. 2021 Mar 15;13(6):1294. doi: 10.3390/cancers13061294

Figure 2.

Figure 2

Patient-to-profile workflow/design of the study. Left panel corresponds with the training of machine learning. Denaturation profiles of plasma samples of healthy individuals and glioma patients generated by nanoDSF are added to a database (Atlas) along with their corresponding clinical status. Artificial intelligence (AI) algorithms are then trained using this Atlas to generate a model. Right panel corresponds with the use of the obtained model to identify whether the nanoDSF profile of a new sample tested corresponds with a glioma or not. Dotted line indicates that the same workflow can be applied to another cancer (Cancer X).