Al momento stai visualizzando Deep Learning for Radiological Image Segmentation

Deep Learning for Radiological Image Segmentation

Background

Artificial intelligence (AI) has shown promises in the development of radiomics signature, based on radiological images, that can predict patient’s response to therapy, thus allowing more personalized treatments. Despite the promising preliminary results, the translation of this approaches into clinical practice is still limited by many reasons, including the lack of automatic segmentation methods. Indeed, both manual and semi-automatic segmentations methods have two main drawbacks: they are a time-consuming task, that has to be regarded as prohibitive when very large databases are evaluated;  they may lead to a high inter-reader variability that can strongly impact on the performance of predictive tools. Therefore, developing automatic segmentation methods is of key importance to realize robust tools that can be effectively used in the clinical practice. 

Aim

The aim of this project is to develop new and innovative Deep Learning (DL) algorithms to automatically segment anatomical structures on radiological images. The main research branches are: 

  • Automatic segmentation of rectal cancer on Magnetic Resonance Imaging (MRI)
  • Automatic segmentation of liver metastases on Computed Tomography (CT) images
  • Automatic segmentation of prostate cancer on MRI

Funding

  1. Alleanza contro il Cancro (ACC) – WG RADIOMICS
  2. Alleanza contro il Cancro (ACC) – WG LUNG RATIONALE
  3. FONDAZIONE AIRC under 5 per mille 2018—ID. 21091 program—P.I. Bardelli Alberto, GLs Daniele Regge, Silvia Marsoni, and Salvatore Siena
  4. Fondazione AIRC under IG2017 – ID. 20398 project – P.I. Daniele Regge.

Publications

  1. Panic J, Rosati S, Defeudis A, Mazzetti S, Giannetto G, Micilotta M, Vassallo L, Gatti M, Regge D, Balestra G, Giannini V. 
    A Fully Automatic Deep Learning Algorithm to Segment Rectal Cancer on MR Images: A Multi-Center Study. In: IEEE Engineering in Medicine and Biology Conference, EMBC 2022 – Conference Proceedings. ; 2022.
  2. Defeudis A, Panic J, Vassallo L,  Regge D, Giannini V. A Deep Learning model to segment liver metastases on CT images acquired at different time-points during chemotherapy. In: IEEE Medical Measurements and Applications, MeMeA 2022- Conference Proceedings. ; 2022.
  3. Panic J, Giannini V, Defeudis A, Regge D, Balestra G, and Rosati S.  Impact of network parameters on a U-Net based system for rectal cancer segmentation on MR images. In: IEEE Medical Measurements and Applications, MeMeA 2022- Conference Proceedings. ; 2022.
  4. Barra D, Nicoletti G, Defeudis A, et al. Deep learning model for automatic prostate segmentation on bicentric T2w images with and without endorectal coil. In: Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. ; 2021:3370-3373. doi:10.1109/EMBC46164.2021.9630792
  5. Panic J, Defeudis A, Mazzetti S, et al. A Convolutional Neural Network based system for Colorectal cancer segmentation on MRI images. In: Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. Vol 2020-July. ; 2020:1675-1678. doi:10.1109/EMBC44109.2020.9175804
  6. Giannini V, Defeudis A, Rosati S, et al. Deep learning to segment liver metastases on CT images: Impact on a radiomics method to predict response to chemotherapy. In: IEEE Medical Measurements and Applications, MeMeA 2020 – Conference Proceedings. ; 2020. doi:10.1109/MeMeA49120.2020.9137150