About DeepMedic:

DeepMedic is software for 3D image segmention, based on a multi-scale 3D Deep Convolutional Neural Network, from the BioMedIA Group of Imperial College London. The system has been shown to achieve excellent performance on brain lesion segmentation for various tasks, including brain injuries, brain tumors, and ischemic stroke lesions. This method is part of the systems that won the ISLES 2015 and BRATS 2017 competitions.

What we are making:

3D Deep Learning Suite

A complete set of versatile and easy to use tools to train, evaluate, and make use of 3D image segmentation models in a simple and intuitive manner for non-technical users.

Tried and Tested Performance

DeepMedic is the winning algorithm in ISLES 2015 competition, part of the winning system in BraTS 2017, and widely used for variety of segmentation tasks in 3D medical imaging.

Integrated Image Pre-Processor

The framework includes automated data harmonisation, and informative dataset metadata checks, made easily and intuitively customisable via its Graphical User Interface.

Graphical Interface

Graphical User Interface enables easy training and testing of Deep Learning models.

Easy To Configure And Customize

Easy to customize and tune for your data and needs to achieve higher performance levels.

Pre-Trained Model Zoo

Distributing pre-trained segmentation models for common medical tasks.

Testimonials:

DeepMedic was successfully adopted in the clinical workflow [for segmenting organs-at-risk in prostate radiotherapy] maintaining in the clinical setting the accuracy obtained in the feasibility study.

Savenije et al., Radiation Oncology, 2020. UMC Utrecht

no statistically significant differences were found between the performance metrics [between DeepMedic and human reader for stroke lesion segmentation in DW MRI]

Wu et al., Stroke, 2019 Massachusetts General Hospital / Harvard University

the system [DeepMedic] learned to outperform its “human teachers” and produced output that was better, on average, than its training data.

Mitchell et al., Journal of Medical Imaging, 2020 Mayo Clinic

the Dice score of the algorithm results were comparable with the Dice scores between human readers in our subcohort [for stroke lesion segmentation in DW MRI]

Winzeck et al., American Journal of Neuroradiology, 2019 MGH, Harvard University, Cambridge University

a whole training process within a single day utilizing a NVIDIA GeForce GTX 580 graphics board which makes this library also very interesting for research purposes on low-end GPU hardware.

Lippert et al. SPIE 2018 Fraunhofer MEVIS

Various independent studies found that DeepMedic out-performed other state-of-the-art deep-learning models in different tasks, such as for liver segmentation in CT or brain vessel segmentation from cerebrovascular images.

Kavur et al. arxiv 2020 Dokuz Eylul University

Patel et al. SPIE 2020 University at Buffalo

DeepMedic

Deep Learning Suite for 3D Image Segmentation