-
http://www.diagnijmegen.nl/
- Nijmegen, The Netherlands
- https://www.linkedin.com/in/anindo-saha/
- @anindox8
Block or Report
Block or report anindox8
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abuse-
website-content Public
Forked from DIAGNijmegen/website-contentThis repository stores all the content for the diag websites.
TeX MIT License UpdatedDec 5, 2022 -
picai_baseline-1 Public
Forked from DIAGNijmegen/picai_baselineBaseline AI models for 3D csPCa detection/diagnosis in bpMRI
Python Apache License 2.0 UpdatedJun 18, 2022 -
DeepLearningExamples Public
Forked from NVIDIA/DeepLearningExamplesDeep Learning Examples
Jupyter Notebook UpdatedDec 9, 2021 -
-
Fully supervised binary classification of skin lesions from dermatoscopic images using an ensemble of diverse CNN architectures (EfficientNet-B6, Inception-V3, SEResNeXt-101, SENet-154, DenseNet-16…
-
Weakly supervised 3D classification of multi-disease chest CT scans using multi-resolution deep segmentation features via dual-stage CNN architecture (DenseVNet, 3D Residual U-Net).
-
Fully supervised, healthy/malignant prostate detection in multi-parametric MRI (T2W, DWI, ADC), using a modified 2D RetinaNet model for medical object detection, built upon a shallow SEResNet backb…
-
Fully supervised, multi-class 3D brain segmentation in T1 MRI, using atlas-based segmentation algorithms (label propagation, tissue models, Expectation-Maximization algorithm).
-
Fully supervised, multi-class 3D brain segmentation in T1 MRI using an ensemble of diverse CNN architectures (3D FCN, 3D U-Net) with multi-scale input.
-
Unsupervised region proposal and supervised patch extraction algorithms for extracting candidate 2D ROIs to train SVM/CNN classifiers, for mass detection in mammograms.
-
Fully supervised binary classification of skin lesions from dermatoscopic images using multi-color space moments/texture features and Support Vector Machines/Random Forests.