Search Results for author: Nils D. Forkert

Found 10 papers, 1 papers with code

Towards objective and systematic evaluation of bias in medical imaging AI

1 code implementation3 Nov 2023 Emma A. M. Stanley, Raissa Souza, Anthony Winder, Vedant Gulve, Kimberly Amador, Matthias Wilms, Nils D. Forkert

In this article, we introduce a novel analysis framework for systematically and objectively investigating the impact of biases in medical images on AI models.

counterfactual

Weakly Supervised Medical Image Segmentation With Soft Labels and Noise Robust Loss

no code implementations16 Sep 2022 Banafshe Felfeliyan, Abhilash Hareendranathan, Gregor Kuntze, Stephanie Wichuk, Nils D. Forkert, Jacob L. Jaremko, Janet L. Ronsky

The aim of this paper was to develop and evaluate a method to generate probabilistic labels based on multi-rater annotations and anatomical knowledge of the lesion features in MRI and a method to train segmentation models using probabilistic labels using normalized active-passive loss as a "noise-tolerant loss" function.

Image Segmentation Medical Diagnosis +3

Self-Supervised-RCNN for Medical Image Segmentation with Limited Data Annotation

no code implementations17 Jul 2022 Banafshe Felfeliyan, Abhilash Hareendranathan, Gregor Kuntze, David Cornell, Nils D. Forkert, Jacob L. Jaremko, Janet L. Ronsky

The effectiveness of the proposed method for segmentation tasks in different pre-training and fine-tuning scenarios is evaluated based on the Osteoarthritis Initiative dataset.

Anomaly Detection Image Segmentation +4

Bone Adaptation as a Geometric Flow

no code implementations9 Nov 2021 Bryce A. Besler, Tannis D. Kemp, Nils D. Forkert, Steven K. Boyd

Interestingly, the parameters of the model can be defined in such a way that the flow is volume-preserving.

Constructing High-Order Signed Distance Maps from Computed Tomography Data with Application to Bone Morphometry

no code implementations2 Nov 2021 Bryce A. Besler, Tannis D. Kemp, Nils D. Forkert, Steven K. Boyd

The narrowband is solved using a closest point algorithm extended for implicit embeddings that are not a signed distance field.

Quantization

High-Order Signed Distance Transform of Sampled Signals

no code implementations26 Oct 2021 Bryce A. Besler, Tannis D. Kemp, Nils D. Forkert, Steven K. Boyd

Such a transform is an improvement to the classic notion of an exact signed distance transform because it does not exhibit artifacts of quantization.

Quantization Unity +1

Local Morphometry of Closed, Implicit Surfaces

no code implementations29 Jul 2021 Bryce A Besler, Tannis D. Kemp, Andrew S. Michalski, Nils D. Forkert, Steven K. Boyd

The proposed method enables an improved local and global evaluation of curvature for purposes of morphometry on closed, implicit surfaces.

Hippocampus Quantization

Bidirectional Modeling and Analysis of Brain Aging with Normalizing Flows

no code implementations26 Nov 2020 Matthias Wilms, Jordan J. Bannister, Pauline Mouches, M. Ethan MacDonald, Deepthi Rajashekar, Sönke Langner, Nils D. Forkert

Brain aging is a widely studied longitudinal process throughout which the brain undergoes considerable morphological changes and various machine learning approaches have been proposed to analyze it.

Automatic segmentation of stroke lesions in non-contrast computed tomography with convolutional neural networks

no code implementations MIDL 2019 Anup Tuladhar, Serena Schimert, Deepthi Rajashekar, Helge C. Kniep, Jens Fiehler, Nils D. Forkert

A total of 291 multi-center clinical NCCT datasets were used: 204 for CNN training, 48 for validation and developing post-processing methods, and 39 for testing.

Lesion Segmentation

DeepVesselNet: Vessel Segmentation, Centerline Prediction, and Bifurcation Detection in 3-D Angiographic Volumes

no code implementations25 Mar 2018 Giles Tetteh, Velizar Efremov, Nils D. Forkert, Matthias Schneider, Jan Kirschke, Bruno Weber, Claus Zimmer, Marie Piraud, Bjoern H. Menze

Our experiments show that, by replacing 3-D filters with cross-hair filters in our network, we achieve over 23% improvement in speed, lower memory footprint, lower network complexity which prevents overfitting and comparable accuracy (with a Cox-Wilcoxon paired sample significance test p-value of 0. 07 when compared to full 3-D filters).

Transfer Learning

Cannot find the paper you are looking for? You can Submit a new open access paper.