TY - GEN
T1 - Interpretable deep model for predicting gene-addicted non-small-cell lung cancer in ct scans
AU - Pino, C.
AU - Palazzo, S.
AU - Trenta, F.
AU - Cordero, F.
AU - Bagci, U.
AU - Rundo, F.
AU - Battiato, S.
AU - Giordano, D.
AU - Aldinucci, M.
AU - Spampinato, C.
N1 - Funding Information:
This work has been partially funded by: a) the H2020 Deep-Health project: Deep-Learning and HPC to Boost Biomedical Applications for Health (G.A. 825111) and b) the University of Catania under the “Piano per la Ricerca 2016/2018”.
Publisher Copyright:
© 2021 IEEE.
PY - 2021/4/13
Y1 - 2021/4/13
N2 - Genetic profiling and characterization of lung cancers have recently emerged as a new technique for targeted therapeutic treatment based on immunotherapy or molecular drugs. However, the most effective way to discover specific gene mutations through tissue biopsy has several limitations, from invasiveness to being a risky procedure. Recently, quantitative assessment of visual features from CT data has been demonstrated to be a valid alternative to biopsy for the diagnosis of gene-addicted tumors. In this paper, we present a deep model for automated lesion segmentation and classification as gene-addicted or not. The segmentation approach extends the 2D Tiramisu architecture for 3D segmentation through dense blocks and squeeze-and-excitation layers, while a multi-scale 3D CNN is used for lesion classification. We also train our model with adversarial samples, and show that this approach acts as a gradient regularizer and enhances model interpretability. We also built a dataset, the first of its nature, consisting of 73 CT scans annotated with the presence of a specific genomics profile. We test our approach on this dataset achieving a segmentation accuracy of 93.11% (Dice score) and a classification accuracy in identifying oncogene-addicted lung tumors of 82.00%.
AB - Genetic profiling and characterization of lung cancers have recently emerged as a new technique for targeted therapeutic treatment based on immunotherapy or molecular drugs. However, the most effective way to discover specific gene mutations through tissue biopsy has several limitations, from invasiveness to being a risky procedure. Recently, quantitative assessment of visual features from CT data has been demonstrated to be a valid alternative to biopsy for the diagnosis of gene-addicted tumors. In this paper, we present a deep model for automated lesion segmentation and classification as gene-addicted or not. The segmentation approach extends the 2D Tiramisu architecture for 3D segmentation through dense blocks and squeeze-and-excitation layers, while a multi-scale 3D CNN is used for lesion classification. We also train our model with adversarial samples, and show that this approach acts as a gradient regularizer and enhances model interpretability. We also built a dataset, the first of its nature, consisting of 73 CT scans annotated with the presence of a specific genomics profile. We test our approach on this dataset achieving a segmentation accuracy of 93.11% (Dice score) and a classification accuracy in identifying oncogene-addicted lung tumors of 82.00%.
KW - 3D Tiramisu
KW - Lesion Classification
KW - XAI
UR - http://www.scopus.com/inward/record.url?scp=85107191324&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85107191324&partnerID=8YFLogxK
U2 - 10.1109/ISBI48211.2021.9433832
DO - 10.1109/ISBI48211.2021.9433832
M3 - Conference contribution
AN - SCOPUS:85107191324
T3 - Proceedings - International Symposium on Biomedical Imaging
SP - 891
EP - 894
BT - 2021 IEEE 18th International Symposium on Biomedical Imaging, ISBI 2021
PB - IEEE Computer Society
T2 - 18th IEEE International Symposium on Biomedical Imaging, ISBI 2021
Y2 - 13 April 2021 through 16 April 2021
ER -
