Gustavo Carneiro

Associate Professor

Australian Centre for Visual Technologies

The University of Adelaide

Dataset Humboldt Fellowship Project Main References	Dataset There are two datasets available: 1. Cross validation sets for the training/testing of the microvessel pixel classification 2. Cross validation sets for the training/testing of the flexible and latent structured output model. They are available as two mat files, and we also provide an m-file to read those sets. The structure of the datasets is self-explanatory, but do not hesitate to email Gustavo Carneiro in case you have questions. Download from HERE. If you use this dataset, please cite the following work: Gustavo Carneiro, Tingying Peng, Christine Bayer and Nassir Navab. Weakly-supervised Structured Output Learning with Flexible and Latent Graphs using High-order Loss Functions. In Proceedings of the International Conference on Computer Vision (ICCV) 2015. Humboldt Fellowship Project We introduce two new structured output models that use a latent graph, which is flexible in terms of the number of nodes and structure, where the training process minimises a high-order loss function using a weakly annotated training set. These models are developed in the context of microscopy imaging of malignant tumours, where the estimation of the number and proportion of classes of microcirculatory supply units (MCSU) is important in the assessment of the efficacy of common cancer treatments (an MCSU is a region of the tumour tissue supplied by a microvessel). The proposed methodologies take as input multimodal microscopy images of a tumour, and estimate the number and proportion of MCSU classes. This estimation is facilitated by the use of an underlying latent graph (not present in the manual annotations), where each MCSU is represented by a node in this graph, labelled with the MCSU class and image location. The training process uses the manual weak annotations available, consisting of the number of MCSU classes per training image, where the training objective is the minimisation of a high-order loss function based on the norm of the error between the manual and estimated annotations. One of the models proposed is based on a new flexible latent structure support vector machine (FLSSVM) and the other is based on a deep convolutional neural network (DCNN) model. Using a dataset of 89 weakly annotated pairs of multimodal images from eight tumours, we show that the quantitative results from DCNN are superior, but the qualitative results from FLSSVM are better and both display high correlation values regarding the number and proportion of MCSU classes compared to the manual annotations. Main References Gustavo Carneiro, Tingying Peng, Christine Bayer and Nassir Navab. Weakly-supervised Structured Output Learning with Flexible and Latent Graphs using High-order Loss Functions. In Proceedings of the International Conference on Computer Vision (ICCV) 2015. Gustavo Carneiro, Tingying Peng, Christine Bayer and Nassir Navab. Flexible and Latent Structured Output Learning: Application to Histology. In MICCAI Workshop Machine Learning in Medical Imaging (MLMI) 2015. Gustavo Carneiro, Tingying Peng, Christine Bayer and Nassir Navab. Automatic Detection of Necrosis, Normoxia and Hypoxia in Tumors from Multimodal Cytological Images. In Proceedings of the International Conference on Image Processing (ICIP) 2015.

Dataset

Humboldt Fellowship Project

Main References

Dataset

There are two datasets available:
1. Cross validation sets for the training/testing of the microvessel pixel classification
2. Cross validation sets for the training/testing of the flexible and latent structured output model.
They are available as two mat files, and we also provide an m-file to read those sets. The structure of the datasets is self-explanatory, but do not hesitate to email Gustavo Carneiro in case you have questions.

Download from HERE.
If you use this dataset, please cite the following work:
Gustavo Carneiro, Tingying Peng, Christine Bayer and Nassir Navab. Weakly-supervised Structured Output Learning with Flexible and Latent Graphs using High-order Loss Functions. In Proceedings of the International Conference on Computer Vision (ICCV) 2015.

Humboldt Fellowship Project

We introduce two new structured output models that use a latent graph, which is flexible in terms of the number of nodes and structure, where the training process minimises a high-order loss function using a weakly annotated training set. These models are developed in the context of microscopy imaging of malignant tumours, where the estimation of the number and proportion of classes of microcirculatory supply units (MCSU) is important in the assessment of the efficacy of common cancer treatments (an MCSU is a region of the tumour tissue supplied by a microvessel). The proposed methodologies take as input multimodal microscopy images of a tumour, and estimate the number and proportion of MCSU classes. This estimation is facilitated by the use of an underlying latent graph (not present in the manual annotations), where each MCSU is represented by a node in this graph, labelled with the MCSU class and image location. The training process uses the manual weak annotations available, consisting of the number of MCSU classes per training image, where the training objective is the minimisation of a high-order loss function based on the norm of the error between the manual and estimated annotations. One of the models proposed is based on a new flexible latent structure support vector machine (FLSSVM) and the other is based on a deep convolutional neural network (DCNN) model. Using a dataset of 89 weakly annotated pairs of multimodal images from eight tumours, we show that the quantitative results from DCNN are superior, but the qualitative results from FLSSVM are better and both display high correlation values regarding the number and proportion of MCSU classes compared to the manual annotations.

Main References

Gustavo Carneiro, Tingying Peng, Christine Bayer and Nassir Navab. Weakly-supervised Structured Output Learning with Flexible and Latent Graphs using High-order Loss Functions. In Proceedings of the International Conference on Computer Vision (ICCV) 2015.

Gustavo Carneiro, Tingying Peng, Christine Bayer and Nassir Navab. Flexible and Latent Structured Output Learning: Application to Histology. In MICCAI Workshop Machine Learning in Medical Imaging (MLMI) 2015.

Gustavo Carneiro, Tingying Peng, Christine Bayer and Nassir Navab. Automatic Detection of Necrosis, Normoxia and Hypoxia in Tumors from Multimodal Cytological Images. In Proceedings of the International Conference on Image Processing (ICIP) 2015.

Gustavo Carneiro

Associate Professor

School of Computer Science

Australian Centre for Visual Technologies

The University of Adelaide