Deep Learning for Multi-Task Medical Image Segmentation in Multiple Modalities

By Pim Moeskops, Jelmer M. Wolterink, Bas H. M. van der Velden, Kenneth G. A. Gilhuijs, Tim Leiner, Max A. Viergever, Ivana Isgum. International Conference on Medical Image Computing and Computer-Assisted Intervention, 2016

339 citations recorded by the source corpus.

Automatic segmentation of medical images is an important task for many clinical applications. In practice, a wide range of anatomical structures are visualised using different imaging modalities. In this paper, we investigate whether a single convolutional neural network (CNN) can be trained to perform different segmentation tasks.

Subfields: Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/c08a2cc3b4953646078ed75ed7f827375a8309b5

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. P. Heng
2. Qi-Yan Dou
3. Lequan Yu
4. Ronald M. Summers
5. Jing Qin
6. Hao Chen
7. Daniel Rueckert
8. B. van Ginneken

In Radiology, Nuclear Medicine and Imaging

1. P. Heng
2. Hao Chen
3. Lequan Yu
4. Qi-Yan Dou
5. Ronald M. Summers
6. Jing Qin
7. Xiaosong Wang
8. Daniel Rueckert

Citations in the corpus

• Deep Learning Techniques for Medical Image Segmentation: Achievements and Challenges (2019, background): “5D approaches [54, 60, 65] are inspired by the fact that 2.”
• ChestX-Ray8: Hospital-Scale Chest X-Ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases (2017, methodology): “ction in CT images [40]; automated pancreas segmentation [36]; cell image segmentation and tracking [35], predicting spinal radiological scores [21] and extensions of multi-modal imaging segmentation [30,16]. The main limitation is that all proposed methods are evaluated on some small-to-middle scal”
• 3D deeply supervised network for automated segmentation of volumetric medical images (2017, methodology): “…features, have evolutionized the natural image processing ( Krizhevsky et al., 012; He et al., 2016; Long et al., 2015 ), and also witnessed sucessful applications in medical image analysis domain ( Albarqouni t al., 2016; Chen et al., 2017; Moeskops et al., 2016a; Setio et al., 016 ).”
• CANet: Cross-Disease Attention Network for Joint Diabetic Retinopathy and Diabetic Macular Edema Grading (2019, background): “Since jointly grading DR and DME diseases is related to the multi-task learning, we also review related works in medical imaging domain [9]–[11], [40], [41] and most of them are designed for image classification or regression tasks.”
• Deep Learning for Cardiac Image Segmentation: A Review (2019, methodology): “End-to-end CNNs With respect to the lumen or vessel wall segmentation, most deep learning based approaches use an end-to-end CNN segmentation scheme to predict dense segmentation probability maps (38, 122, 126, 205).”
• Medical Image Analysis using Convolutional Neural Networks: A Review (2017, background): “Some recent studies have shown that deep learning algorithms are successfully used for medical imaging applications [78-81].”
• Unpaired Multi-Modal Segmentation via Knowledge Distillation (2020, background): “[39] build a single network for different segmentation tasks in different modalities.”
• A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promises (2020, methodology): “[182]: Here, a single CNN is trained to perform three different segmentation tasks, including coronary artery”
• Medical Image Segmentation Using Deep Learning: A Survey (2020, methodology): “hen forwarded to the shared convolutional and independent normalization layers, and finally the logarithms that can be used to calculate knowledge distillation losses will be obtained. Moeskops et al. [141] investigated a question whether it is possible to train a single convolutional neural network ”
• A Survey on Active Learning and Human-in-the-Loop Deep Learning for Medical Image Analysis (2019, background): “Many works aim to train models or acquire training data for several tasks at once, it is argued that this can save on cost as complementary information may result in higher performance over multiple di ff erent tasks (Moeskops et al., 2016).”
• Evaluation of a deep learning approach for the segmentation of brain tissues and white matter hyperintensities of presumed vascular origin in MRI (2017, neutral): “This paper has presented the evaluation of an automatic segmentation method for brain tissues and WMH in MRI using a convolutional neural network.”