Fast interactive medical image segmentation with weakly supervised deep learning method

Kibrom Berihu Girum (1, 2, 3) , Gilles Créhange (1, 2, 3) , Raabid Hussain (1, 2) , Alain Lalande (1, 3)

1 Equipe IFTIM [ImViA - EA7535]
2 ImViA - Imagerie et Vision Artificielle [Dijon]
3 Service de RMN Spectroscopie, Médecine Nucléaire (CHU de Dijon)

To achieve accurate image segmentation, which is the first critical step in medical image analysis and interventions, using deep neural networks seems a promising approach provided sufficiently large and diverse annotated data from experts. However, annotated datasets are often limited because it is prone to variations in acquisition parameters and require high-level expert's knowledge, and manually labeling targets by tracing their contour is often laborious. Developing fast, interactive, and weakly supervised deep learning methods is thus highly desirable.

Mots clés: Brachytherapy; CNN; Domain adaptation; Echocardiography; Generative model; Weakly supervised segmentation; Brachytherapy

Domaines: Imagerie médicale

Kibrom Berihu Girum, Gilles Créhange, Raabid Hussain, Alain Lalande. Fast interactive medical image segmentation with weakly supervised deep learning method. International Journal of Computer Assisted Radiology and Surgery, 2020, 15 (9), pp.1437-1444. ⟨10.1007/s11548-020-02223-x⟩. ⟨hal-03119211⟩