Jörg Vorndran published his new AI-based FCD classifier (funded by the CRC 1540)
A new deep-learning-based algorithm to histopathologically classify Focal Cortical Dysplasia is accepted for publication in the EPILEPSIA journal
The histopathology diagnosis of Focal Cortical Dysplasia (FCD) remains challenging in our daily histopathology practise. Herein, we tested the usefulness of a new AI-based classifier to recognize the characteristic histopathology patterns of major FCD subtypes as defined by the International League against Epilepsy classification scheme. Interestingly, the classifier recognized the histopathology diagnosis in all cases, including those carrying a known pathogenic gene variant. However, 86.2% of all cases revealed more than one diagnostic category. Microscopic review confirmed that vertically oriented architectural abnormalities of the neocortex are more common than previously anticipated by the ILAE classification scheme. In conclusion, AI-based diagnostic classifier will become a helpful tool in our future histopathology laboratory. We also provided an open access web application allowing the histopathologist to virtually review digital tiles obtained from epilepsy surgery in order to corroborate their final diagnosis.
We congratulate Jörg Vorndran for this great achievement, which is also part of his PhD thesis at the FAU.
