Deep artificial intelligent systems for biomedical image processing

DAISYlabs is a joint effort of PhD, MSc and MD students of the University Medical Center Hamburg-Eppendorf (UKE) and the Hamburg University of Technology (TUHH) to establish a deep learning platform for biomedical image processing that allows efficient collaboration and knowledge exchange between UKE and TUHH working groups in the field of biomedical imaging and image processing. It is funded by Forschungszentrum Medizintechnik Hamburg ( fmthh ).

MEET THE TEAM

Founder

René Werner

Co-Founder / Head of scientific working group

Ivo Matteo Baltruschat

Co-Founder / Research Scientist

Core

Frederic Madesta

PhD Student

Rüdiger Schmitz

PhD Student

Nils Gessert

Research Scientist

Thilo Sentker

Research Scientist

Tobias Knopp

Professor

Maximilian Nielsen

Student

Helge Kniep

PhD Student

Alumni

Mohsin Shaikh

MSc Student

PROJECTS

We work on developing AI systems for medical image processing problems.

ISIC 2019: Skin Lesion Analysis Towards Melanoma Detection - Task 2

1^st rank of 16 unique teams in Task 2: Lesion diagnosis with images and metadata

ISIC 2019: Skin Lesion Analysis Towards Melanoma Detection - Task 1

1^st rank of 64 unique teams in Task 1: Lesion diagnosis with images only

ISIC 2018: Skin Lesion Analysis Towards Melanoma Detection - Task 3

2^nd rank of 77 unique teams in Task 3: Lesion Diagnosis

PAIP 2019 Challenge - Task 2

7^th rank of 17 unique teams in Task 2: Viable Tumor Burden Estimation

PAIP 2019 Challenge - Task 1

7^th rank of 27 unique teams in Task 1: Liver Cancer Segmentation

StructSeg 2019 - Task 4

7^th rank of 17 unique teams in Task 4: Gross Target Volume segmentation of lung cancer

MR Brain Segmentation 2018 - MRBrainS18

7^th rank of 40 unique teams in Task 2: Three Label Segmentation

StructSeg 2019 - Task 3

8^th rank of 17 unique teams in Task 3: Organ-at-risk segmentation from chest CT scans

ISIC 2018: Skin Lesion Analysis Towards Melanoma Detection - Task 1

11^th rank of 77 unique teams in Task 1: Lesion Boundary Segmentation

SIIM-ACR Pneumothorax Segmentation

Top 20% of 1475 unique teams in identifing Pneumothorax disease in chest x-rays

PUBLICATIONS

Multi-scale fully convolutional neural networks for histopathology image segmentation: from nuclear aberrations to the global tissue architecture

Rüdiger Schmitz, Frederic Madesta, Maximilian Nielsen, René Werner, Thomas Rösch

PDF Project Project

Skin Lesion Classification Using CNNs with Patch-Based Attention and Diagnosis-Guided Loss Weighting

Nils Gessert, Thilo Sentker, Frederic Madesta, Rüdiger Schmitz, Helge Kniep, Ivo Matteo Baltruschat, René Werner, Alexander Schlaefer

PDF Project

Ensemble Building of State-of-the-art Models for Skin Lesion Boundary Segmentation

Ivo Matteo Baltruschat, Thilo Sentker, Frederic Madesta, Rüdiger Schmitz, Helge Kniep, Nils Gessert, René Werner, Tobias Knopp

Project

Skin Lesion Diagnosis using Ensembles, Unscaled Multi-Crop Evaluation and Loss Weighting

Nils Gessert, Thilo Sentker, Frederic Madesta, Rüdiger Schmitz, Helge Kniep, Ivo Matteo Baltruschat, René Werner, Alexander Schlaefer

PDF Project