Interpretable Deep Learning as a mean for decrypting disease signature in Multiple Sclerosis

Federica Cruciani (1) , Lorenza Brusini (1) , Mauro Zucchelli (2, 3) , Gustavo Retuci Pinheiro (4) , Francesco Setti (1) , Ilaria Boscolo Galazzo (1) , Rachid Deriche​ (3) , Leticia Rittner (4) , Massimiliano Calabrese (1) , Gloria Menegaz (1)

1 UNIVR - Università degli studi di Verona = University of Verona
2 COMUE UCA - COMUE Université Côte d'Azur (2015-2019)
3 ATHENA - Computational Imaging of the Central Nervous System
4 UNICAMP - Universidade Estadual de Campinas = University of Campinas

Objective.The mechanisms driving multiple sclerosis (MS) are still largely unknown, calling for new methods allowing to detect and characterize tissue degeneration since the early stages of the disease. Our aim is to decrypt the microstructural signatures of the Primary Progressive versus the Relapsing-Remitting state of disease based on diffusion and structural magnetic resonance imaging data.Approach.A selection of microstructural descriptors, based on the 3D-Simple Harmonics Oscillator Based Reconstruction and Estimation and the set of new algebraically independent Rotation Invariant spherical harmonics Features, was considered and used to feed convolutional neural networks (CNNs) models. Classical measures derived from diffusion tensor imaging, that are fractional anisotropy and mean diffusivity, were used as benchmark for diffusion MRI (dMRI). Finally, T1-weighted images were also considered for the sake of comparison with the state-of-the-art. A CNN model was fit to each feature map and layerwise relevance propagation (LRP) heatmaps were generated for each model, target class and subject in the test set. Average heatmaps were calculated across correctly classified patients and size-corrected metrics were derived on a set of regions of interest to assess the LRP contrast between the two classes.Main results.Our results demonstrated that dMRI features extracted in grey matter tissues can help in disambiguating primary progressive multiple sclerosis from relapsing-remitting multiple sclerosis patients and, moreover, that LRP heatmaps highlight areas of high relevance which relate well with what is known from literature for MS disease.Significance.Within a patient stratification task, LRP allows detecting the input voxels that mostly contribute to the classification of the patients in either of the two classes for each feature, potentially bringing to light hidden data properties which might reveal peculiar disease-state factors.

Mots clés: convolutional neural network; diffusion MRI; explainable artificial intelligence; layer-wise relevance propagation; multiple sclerosis; rotation invariant spherical harmonics features.

Domaines: Imagerie médicale; Apprentissage [cs.LG]

Federica Cruciani, Lorenza Brusini, Mauro Zucchelli, Gustavo Retuci Pinheiro, Francesco Setti, et al.. Interpretable Deep Learning as a mean for decrypting disease signature in Multiple Sclerosis. Journal of Neural Engineering, 2021, ⟨10.1088/1741-2552/ac0f4b⟩. ⟨hal-02971374⟩