Unimore logo AImageLab
Gianpaolo Bontempo

Gianpaolo Bontempo

Position at AImageLab:

PhD Student
Dipartimento di Ingegneria "Enzo Ferrari"

Email:

gianpaolo_DOT_bontempo_AT_unimore_DOT_it

Gianpaolo Bontempo

​Research Topic:

Computer vision has achieved astonishing results in analysing histopathological whole slide images (WSI). Thanks to their pyramid structure and size, they allow an analysis of the tissue on different scales. However, considering their usual gigapixel dimension, annotating regions of interest for supervised tasks requires considerable effort. Then, it is common practice to have a unique label characterising the entire WSI, while the WSI image analysis implies dividing it into patches that inherit the WSI's label. However, the patches do not all forcefully correlate with the WSI's label. In this contest, multi-instance learning (MIL) approaches have been considered. In particular, they consider each slide as a bag of patches, and the task of MIL is to predict the label of the bag having the entire set of patches and the bag label. Two elements characterise the typical architecture: 1) an embedder able to extract representative embeddings from patches; 2) an aggregator able to output the bag's label considering the entire set of embeddings.
The latter is fundamental to predicting the bag label and explaining which patches are more important than the others. 
 

Publications:

  • Lovino, M., Bontempo, G., Cirrincione, G., Ficarra, E. (2020). Multi-omics Classification on Kidney Samples Exploiting Uncertainty-Aware Models. In: Huang, DS., Jo, KH. (eds) Intelligent Computing Theories and Application. ICIC 2020. Lecture Notes in Computer Science(), vol 12464. Springer, Cham. https://doi.org/10.1007/978-3-030-60802-6_4

     

  • Citarrella, F., Bontempo, G., Lovino, M., Ficarra, E. (2022). FusionFlow: An Integrated System Workflow for Gene Fusion Detection in Genomic Samples. In: , et al. New Trends in Database and Information Systems. ADBIS 2022. Communications in Computer and Information Science, vol 1652. Springer, Cham. https://doi.org/10.1007/978-3-031-15743-1_8

     

  1. Marconato, E., Bontempo, G., Teso, S., Ficarra, E., Calderara, S., Passerini, A. (2022). Catastrophic Forgetting in Continual Concept Bottleneck Models. In: Mazzeo, P.L., Frontoni, E., Sclaroff, S., Distante, C. (eds) Image Analysis and Processing. ICIAP 2022 Workshops. ICIAP 2022. Lecture Notes in Computer Science, vol 13374. Springer, Cham. https://doi.org/10.1007/978-3-031-13324-4_46

     

 

 

Research Projects

Research Activities