Daniel A. Wolf

Abstract: Your other Left! Vision-language Models Fail to Understand Relative Positions in Medical Images

Daniel A. Wolf, Heiko Hillenhagen, Billurvan Taskin, Alex Bäuerle et autres

de (code pays fourni par la source)

Semi-quantitative software evaluation of COVID-19 CT examinations—correlation with clinical parameters

Nicoleta Trif, Christopher Kloth, Susanne Martina Büttner, N Egenrieder et autres

Background: Software-guided semi-quantitative analysis of coronavirus disease 2019 (COVID-19) pneumonia in lung computed tomography (CT) datasets for severity assessment. Further to correlate imaging findings with the need of intensive care medicine and clinical parameters. Methods: This single-center retrospective study analyzed 66 consecutive …

de (code pays fourni par la source)

Your other Left! Vision-Language Models Fail to Identify Relative Positions in Medical Images

Daniel A. Wolf, Heiko Hillenhagen, Alex Bäuerle, Meinrad Beer et autres

de, us (code pays fourni par la source)

Your other Left! Vision-Language Models Fail to Identify Relative Positions in Medical Images

Daniel A. Wolf, Heiko Hillenhagen, Alex Bäuerle, Meinrad Beer et autres

Clinical decision-making relies heavily on understanding relative positions of anatomical structures and anomalies. Therefore, for Vision-Language Models (VLMs) to be applicable in clinical practice, the ability to accurately determine relative positions on medical images is a fundamental prerequisite. Despite its importance, this …

Abstract: Selective Reduction of CT Data for Self-supervised Pre-training Improves Downstream Classification Performance

Daniel A. Wolf, Tristan Payer, Catharina Silvia Lisson, Christoph Gerhard Lisson et autres

de (code pays fourni par la source)

Less is More: Selective reduction of CT data for self-supervised pre-training of deep learning models with contrastive learning improves downstream classification performance

Daniel A. Wolf, Tristan Payer, Catharina Silvia Lisson, Christoph Gerhard Lisson et autres

BACKGROUND: Self-supervised pre-training of deep learning models with contrastive learning is a widely used technique in image analysis. Current findings indicate a strong potential for contrastive pre-training on medical images. However, further research is necessary to incorporate the particular characteristics of these …

de (code pays fourni par la source)

Forced LMX1A expression induces dorsal neural fates and disrupts patterning of human embryonic stem cells into ventral midbrain dopaminergic neurons

Pedro Rifes, Janko Kajtez, Josefine Rågård Christiansen, Alrik L. Schörling et autres

The differentiation of human pluripotent stem cells into ventral mesencephalic dopaminergic (DA) fate is relevant for the treatment of Parkinson's disease. Shortcuts to obtaining DA cells through direct reprogramming often include forced expression of the transcription factor LMX1A. Although reprogramming with LMX1A …

dk, se (code pays fourni par la source)

Applicability of the CT Radiomics of Skeletal Muscle and Machine Learning for the Detection of Sarcopenia and Prognostic Assessment of Disease Progression in Patients with Gastric and Esophageal Tumors

Daniel Vogele, Teresa M. Mueller, Daniel A. Wolf, Stephanie Otto et autres

Purpose: Sarcopenia is considered a negative prognostic factor in patients with malignant tumors. Among other diagnostic options, computed tomography (CT), which is repeatedly performed on tumor patients, can be of further benefit. The present study aims to establish a framework for classifying …

de (code pays fourni par la source)

Abstract: Self-supervised Pre-training for Dealing with Small Datasets in Deep Learning for Medical Imaging

Daniel A. Wolf, Tristan Payer, Catharina Silvia Lisson, Christoph Gerhard Lisson et autres

de (code pays fourni par la source)

Radiomics and Clinicopathological Characteristics for Predicting Lymph Node Metastasis in Testicular Cancer

Catharina Silvia Lisson, Sabitha Manoj, Daniel A. Wolf, Stefan Andreas Schmidt et autres

Accurate prediction of lymph node metastasis (LNM) in patients with testicular cancer is highly relevant for treatment decision-making and prognostic evaluation. Our study aimed to develop and validate clinical radiomics models for individual preoperative prediction of LNM in patients with testicular cancer. …

de (code pays fourni par la source)

Self-supervised pre-training with contrastive and masked autoencoder methods for dealing with small datasets in deep learning for medical imaging

Daniel A. Wolf, Tristan Payer, Catharina Silvia Lisson, Christoph Gerhard Lisson et autres

Deep learning in medical imaging has the potential to minimize the risk of diagnostic errors, reduce radiologist workload, and accelerate diagnosis. Training such deep learning models requires large and accurate datasets, with annotations for all training samples. However, in the medical imaging …

de (code pays fourni par la source)

Self-Supervised Pre-Training with Contrastive and Masked Autoencoder Methods for Dealing with Small Datasets in Deep Learning for Medical Imaging

Daniel A. Wolf, Tristan Payer, Catharina Silvia Lisson, Christoph Gerhard Lisson et autres

Deep learning in medical imaging has the potential to minimize the risk of diagnostic errors, reduce radiologist workload, and accelerate diagnosis. Training such deep learning models requires large and accurate datasets, with annotations for all training samples. However, in the medical imaging …

de (code pays fourni par la source)