no code implementations • 16 May 2024 • Milda Pocevičiūtė, Gabriel Eilertsen, Stina Garvin, Claes Lundström
Our contributions include showing that MIL for digital pathology is affected by clinically realistic differences in data, evaluating which features from a MIL model are most suitable for detecting changes in performance, and proposing an unsupervised metric named Fr\'echet Domain Distance (FDD) for quantification of domain shifts.
no code implementations • 1 May 2023 • Axel Gödrich, Daniel König, Gabriel Eilertsen, Michael Teutsch
In this paper, tone mapping algorithms for thermal infrared images with 16 bpp are investigated that can preserve this information.
1 code implementation • ACM SIGGRAPH Conference Proceedings 2022 • Param Hanji, Rafał K. Mantiuk, Gabriel Eilertsen, Saghi Hajisharif, Jonas Unger
As the problem of reconstructing high dynamic range (HDR) images from a single exposure has attracted much research effort, it is essential to provide a robust protocol and clear guidelines on how to evaluate and compare new methods.
no code implementations • 27 May 2022 • Rym Jaroudi, Lukáš Malý, Gabriel Eilertsen, B. Tomas Johansson, Jonas Unger, George Baravdish
This paper presents the Standalone Neural ODE (sNODE), a continuous-depth neural ODE model capable of describing a full deep neural network.
no code implementations • 11 Feb 2022 • George Baravdish, Gabriel Eilertsen, Rym Jaroudi, B. Tomas Johansson, Lukáš Malý, Jonas Unger
The inverse problem of supervised reconstruction of depth-variable (time-dependent) parameters in a neural ordinary differential equation (NODE) is considered, that means finding the weights of a residual network with time continuous layers.
no code implementations • 17 Dec 2021 • Milda Pocevičiūtė, Gabriel Eilertsen, Sofia Jarkman, Claes Lundström
In this work we evaluate if adding uncertainty estimates for DL predictions in digital pathology could result in increased value for the clinical applications, by boosting the general predictive performance or by detecting mispredictions.
1 code implementation • 10 Dec 2021 • Karin Stacke, Jonas Unger, Claes Lundström, Gabriel Eilertsen
We bring forward a number of considerations, such as view generation for the contrastive objective and hyper-parameter tuning.
no code implementations • 17 Sep 2021 • Apostolia Tsirikoglou, Karin Stacke, Gabriel Eilertsen, Jonas Unger
The scarcity of labeled data is a major bottleneck for developing accurate and robust deep learning-based models for histopathology applications.
1 code implementation • 19 Aug 2021 • Gabriel Eilertsen, Saghi Hajisharif, Param Hanji, Apostolia Tsirikoglou, Rafal K. Mantiuk, Jonas Unger
Here, we reproduce a typical evaluation using existing as well as simulated SI-HDR methods to demonstrate how different aspects of the problem affect objective quality metrics.
no code implementations • 23 Apr 2021 • Gabriel Eilertsen, Apostolia Tsirikoglou, Claes Lundström, Jonas Unger
This work investigates the use of synthetic images, created by generative adversarial networks (GANs), as the only source of training data.
no code implementations • 16 Mar 2021 • Milda Pocevičiūtė, Gabriel Eilertsen, Claes Lundström
Machine learning (ML) algorithms are optimized for the distribution represented by the training data.
no code implementations • 14 Aug 2020 • Milda Pocevičiūtė, Gabriel Eilertsen, Claes Lundström
We present a survey on XAI within digital pathology, a medical imaging sub-discipline with particular characteristics and needs.
Explainable artificial intelligence
Explainable Artificial Intelligence (XAI)
+1
no code implementations • 20 May 2020 • Apostolia Tsirikoglou, Karin Stacke, Gabriel Eilertsen, Martin Lindvall, Jonas Unger
One such scenario relates to detecting tumor metastasis in lymph node tissue, where the low ratio of tumor to non-tumor cells makes the diagnostic task hard and time-consuming.
1 code implementation • 13 Feb 2020 • Gabriel Eilertsen, Daniel Jönsson, Timo Ropinski, Jonas Unger, Anders Ynnerman
of neural network classifiers, and train a large number of models to represent the weight space.
1 code implementation • 25 Sep 2019 • Karin Stacke, Gabriel Eilertsen, Jonas Unger, Claes Lundström
Most centrally, we present a novel measure for evaluating the distance between domains in the context of the learned representation of a particular model.
no code implementations • CVPR 2019 • Gabriel Eilertsen, Rafał K. Mantiuk, Jonas Unger
The regularization is formulated to account for different types of motion that can occur between frames, so that temporally stable CNNs can be trained without the need for video material or expensive motion estimation.
2 code implementations • 20 Oct 2017 • Gabriel Eilertsen, Joel Kronander, Gyorgy Denes, Rafał K. Mantiuk, Jonas Unger
We demonstrate that our approach can reconstruct high-resolution visually convincing HDR results in a wide range of situations, and that it generalizes well to reconstruction of images captured with arbitrary and low-end cameras that use unknown camera response functions and post-processing.