Tassilo Wald
YOU?
Author Swipe
View article: Comprehensive language-image pre-training for 3D medical image understanding
Comprehensive language-image pre-training for 3D medical image understanding Open
Vision-language pre-training, i.e., aligning images with paired text, is a powerful paradigm to create encoders that can be directly used for tasks such as classification and retrieval, and for downstream tasks such as segmentation and rep…
View article: The Missing Piece: A Case for Pre-training in 3D Medical Object Detection
The Missing Piece: A Case for Pre-training in 3D Medical Object Detection Open
View article: Large Scale Supervised Pretraining For Traumatic Brain Injury Segmentation
Large Scale Supervised Pretraining For Traumatic Brain Injury Segmentation Open
The segmentation of lesions in Moderate to Severe Traumatic Brain Injury (msTBI) presents a significant challenge in neuroimaging due to the diverse characteristics of these lesions, which vary in size, shape, and distribution across brain…
View article: ReSi Benchmark Models (Vision - Part 4)
ReSi Benchmark Models (Vision - Part 4) Open
ReSi Benchmark Models (Vision - Part 4)Vision Default Models trained on CIFAR100
View article: nnInteractive: Redefining 3D Promptable Segmentation
nnInteractive: Redefining 3D Promptable Segmentation Open
Accurate and efficient 3D segmentation is essential for both clinical and research applications. While foundation models like SAM have revolutionized interactive segmentation, their 2D design and domain shift limitations make them ill-suit…
View article: Primus: Enforcing Attention Usage for 3D Medical Image Segmentation
Primus: Enforcing Attention Usage for 3D Medical Image Segmentation Open
Transformers have achieved remarkable success across multiple fields, yet their impact on 3D medical image segmentation remains limited with convolutional networks still dominating major benchmarks. In this work, we a) analyze current Tran…
View article: Enhancing deep learning methods for brain metastasis detection through cross-technique annotations on SPACE MRI
Enhancing deep learning methods for brain metastasis detection through cross-technique annotations on SPACE MRI Open
View article: LNQ 2023 challenge: Benchmark of weakly-supervised techniques for mediastinal lymph node quantification
LNQ 2023 challenge: Benchmark of weakly-supervised techniques for mediastinal lymph node quantification Open
Accurate assessment of lymph node size in 3D CT scans is crucial for cancer staging, therapeutic management, and monitoring treatment response. Existing state-of-the-art segmentation frameworks in medical imaging often rely on fully annota…
View article: A Unified Framework for Foreground and Anonymization Area Segmentation in CT and MRI Data
A Unified Framework for Foreground and Anonymization Area Segmentation in CT and MRI Data Open
This study presents an open-source toolkit to address critical challenges in preprocessing data for self-supervised learning (SSL) for 3D medical imaging, focusing on data privacy and computational efficiency. The toolkit comprises two mai…
View article: Revisiting MAE pre-training for 3D medical image segmentation
Revisiting MAE pre-training for 3D medical image segmentation Open
Self-Supervised Learning (SSL) presents an exciting opportunity to unlock the potential of vast, untapped clinical datasets, for various downstream applications that suffer from the scarcity of labeled data. While SSL has revolutionized fi…
View article: Decoupling Semantic Similarity from Spatial Alignment for Neural Networks
Decoupling Semantic Similarity from Spatial Alignment for Neural Networks Open
What representation do deep neural networks learn? How similar are images to each other for neural networks? Despite the overwhelming success of deep learning methods key questions about their internal workings still remain largely unanswe…
View article: Longitudinal Segmentation of MS Lesions via Temporal Difference Weighting
Longitudinal Segmentation of MS Lesions via Temporal Difference Weighting Open
Accurate segmentation of Multiple Sclerosis (MS) lesions in longitudinal MRI scans is crucial for monitoring disease progression and treatment efficacy. Although changes across time are taken into account when assessing images in clinical …
View article: How do deep-learning models generalize across populations? Cross-ethnicity generalization of COPD detection
How do deep-learning models generalize across populations? Cross-ethnicity generalization of COPD detection Open
Objectives To evaluate the performance and potential biases of deep-learning models in detecting chronic obstructive pulmonary disease (COPD) on chest CT scans across different ethnic groups, specifically non-Hispanic White (NHW) and Afric…
View article: Mitigating False Predictions In Unreasonable Body Regions
Mitigating False Predictions In Unreasonable Body Regions Open
Despite considerable strides in developing deep learning models for 3D medical image segmentation, the challenge of effectively generalizing across diverse image distributions persists. While domain generalization is acknowledged as vital …
View article: nnU-Net Revisited: A Call for Rigorous Validation in 3D Medical Image Segmentation
nnU-Net Revisited: A Call for Rigorous Validation in 3D Medical Image Segmentation Open
The release of nnU-Net marked a paradigm shift in 3D medical image segmentation, demonstrating that a properly configured U-Net architecture could still achieve state-of-the-art results. Despite this, the pursuit of novel architectures, an…
View article: Skeleton Recall Loss for Connectivity Conserving and Resource Efficient Segmentation of Thin Tubular Structures
Skeleton Recall Loss for Connectivity Conserving and Resource Efficient Segmentation of Thin Tubular Structures Open
Accurately segmenting thin tubular structures, such as vessels, nerves, roads or concrete cracks, is a crucial task in computer vision. Standard deep learning-based segmentation loss functions, such as Dice or Cross-Entropy, focus on volum…
View article: Capturing COPD heterogeneity: anomaly detection and parametric response mapping comparison for phenotyping on chest computed tomography
Capturing COPD heterogeneity: anomaly detection and parametric response mapping comparison for phenotyping on chest computed tomography Open
Background Chronic obstructive pulmonary disease (COPD) poses a substantial global health burden, demanding advanced diagnostic tools for early detection and accurate phenotyping. In this line, this study seeks to enhance COPD characteriza…
View article: Abstract: Reformulating COPD Classification on Chest CT Scans as Anomaly Detection using Contrastive Representations
Abstract: Reformulating COPD Classification on Chest CT Scans as Anomaly Detection using Contrastive Representations Open
View article: Abstract: RecycleNet
Abstract: RecycleNet Open
View article: Abstract: Multi-dataset Approach to Medical Image Segmentation
Abstract: Multi-dataset Approach to Medical Image Segmentation Open
View article: Prediction of disease severity in COPD: a deep learning approach for anomaly-based quantitative assessment of chest CT
Prediction of disease severity in COPD: a deep learning approach for anomaly-based quantitative assessment of chest CT Open
Objectives To quantify regional manifestations related to COPD as anomalies from a modeled distribution of normal-appearing lung on chest CT using a deep learning (DL) approach, and to assess its potential to predict disease severity. Mate…
View article: RecycleNet: Latent Feature Recycling Leads to Iterative Decision Refinement
RecycleNet: Latent Feature Recycling Leads to Iterative Decision Refinement Open
Despite the remarkable success of deep learning systems over the last decade, a key difference still remains between neural network and human decision-making: As humans, we cannot only form a decision on the spot, but also ponder, revisiti…
View article: cOOpD: Reformulating COPD classification on chest CT scans as anomaly detection using contrastive representations
cOOpD: Reformulating COPD classification on chest CT scans as anomaly detection using contrastive representations Open
Classification of heterogeneous diseases is challenging due to their complexity, variability of symptoms and imaging findings. Chronic Obstructive Pulmonary Disease (COPD) is a prime example, being underdiagnosed despite being the third le…
View article: Exploring new ways: Enforcing representational dissimilarity to learn new features and reduce error consistency
Exploring new ways: Enforcing representational dissimilarity to learn new features and reduce error consistency Open
Independently trained machine learning models tend to learn similar features. Given an ensemble of independently trained models, this results in correlated predictions and common failure modes. Previous attempts focusing on decorrelation o…
View article: SAM.MD: Zero-shot medical image segmentation capabilities of the Segment Anything Model
SAM.MD: Zero-shot medical image segmentation capabilities of the Segment Anything Model Open
Foundation models have taken over natural language processing and image generation domains due to the flexibility of prompting. With the recent introduction of the Segment Anything Model (SAM), this prompt-driven paradigm has entered image…
View article: MultiTalent: A Multi-Dataset Approach to Medical Image Segmentation
MultiTalent: A Multi-Dataset Approach to Medical Image Segmentation Open
The medical imaging community generates a wealth of datasets, many of which are openly accessible and annotated for specific diseases and tasks such as multi-organ or lesion segmentation. Current practices continue to limit model training …
View article: Extending nnU-Net is all you need
Extending nnU-Net is all you need Open
Semantic segmentation is one of the most popular research areas in medical image computing. Perhaps surprisingly, despite its conceptualization dating back to 2018, nnU-Net continues to provide competitive out-of-the-box solutions for a br…
View article: Temporal Feature Networks for CNN based Object Detection
Temporal Feature Networks for CNN based Object Detection Open
For reliable environment perception, the use of temporal information is essential in some situations. Especially for object detection, sometimes a situation can only be understood in the right perspective through temporal information. Sinc…