Georg Wölflein
YOU?
Author Swipe
View article: 74P Self-hosted open-source large language models for autonomous clinical agents
74P Self-hosted open-source large language models for autonomous clinical agents Open
View article: Leveraging knowledge for explainable AI in personalized cancer treatment: challenges and future directions
Leveraging knowledge for explainable AI in personalized cancer treatment: challenges and future directions Open
Integrating multi-modal patient data to support personalized medicine has gained a lot of interest across different health domains over the past decade. Addressing this challenge requires the development and implementation of an informed, …
View article: Development and validation of an autonomous artificial intelligence agent for clinical decision-making in oncology
Development and validation of an autonomous artificial intelligence agent for clinical decision-making in oncology Open
Clinical decision-making in oncology is complex, requiring the integration of multimodal data and multidomain expertise. We developed and evaluated an autonomous clinical artificial intelligence (AI) agent leveraging GPT-4 with multimodal …
View article: LLM Agents Making Agent Tools
LLM Agents Making Agent Tools Open
Tool use has turned large language models (LLMs) into powerful agents that can perform complex multi-step tasks by dynamically utilising external software components. However, these tools must be implemented in advance by human developers,…
View article: LLM Agents Making Agent Tools
LLM Agents Making Agent Tools Open
View article: Abnormality-Driven Representation Learning for Radiology Imaging
Abnormality-Driven Representation Learning for Radiology Imaging Open
To date, the most common approach for radiology deep learning pipelines is the use of end-to-end 3D networks based on models pre-trained on other tasks, followed by fine-tuning on the task at hand. In contrast, adjacent medical fields such…
View article: In-context learning enables multimodal large language models to classify cancer pathology images
In-context learning enables multimodal large language models to classify cancer pathology images Open
Medical image classification requires labeled, task-specific datasets which are used to train deep learning networks de novo, or to fine-tune foundation models. However, this process is computationally and technically demanding. In languag…
View article: Unsupervised Foundation Model-Agnostic Slide-Level Representation Learning
Unsupervised Foundation Model-Agnostic Slide-Level Representation Learning Open
Representation learning of pathology whole-slide images (WSIs) has primarily relied on weak supervision with Multiple Instance Learning (MIL). This approach leads to slide representations highly tailored to a specific clinical task. Self-s…
View article: End-To-End Clinical Trial Matching with Large Language Models
End-To-End Clinical Trial Matching with Large Language Models Open
Matching cancer patients to clinical trials is essential for advancing treatment and patient care. However, the inconsistent format of medical free text documents and complex trial eligibility criteria make this process extremely challengi…
View article: Autonomous Artificial Intelligence Agents for Clinical Decision Making in Oncology
Autonomous Artificial Intelligence Agents for Clinical Decision Making in Oncology Open
Multimodal artificial intelligence (AI) systems have the potential to enhance clinical decision-making by interpreting various types of medical data. However, the effectiveness of these models across all medical fields is uncertain. Each d…
View article: In-context learning enables multimodal large language models to classify cancer pathology images
In-context learning enables multimodal large language models to classify cancer pathology images Open
Medical image classification requires labeled, task-specific datasets which are used to train deep learning networks de novo, or to fine-tune foundation models. However, this process is computationally and technically demanding. In languag…
View article: Joint multi-task learning improves weakly-supervised biomarker prediction in computational pathology
Joint multi-task learning improves weakly-supervised biomarker prediction in computational pathology Open
Deep Learning (DL) can predict biomarkers directly from digitized cancer histology in a weakly-supervised setting. Recently, the prediction of continuous biomarkers through regression-based DL has seen an increasing interest. Nonetheless, …
View article: From Whole-slide Image to Biomarker Prediction: A Protocol for End-to-End Deep Learning in Computational Pathology
From Whole-slide Image to Biomarker Prediction: A Protocol for End-to-End Deep Learning in Computational Pathology Open
Hematoxylin- and eosin (H&E) stained whole-slide images (WSIs) are the foundation of diagnosis of cancer. In recent years, development of deep learning-based methods in computational pathology enabled the prediction of biomarkers directly …
View article: Benchmarking Pathology Feature Extractors for Whole Slide Image Classification
Benchmarking Pathology Feature Extractors for Whole Slide Image Classification Open
Weakly supervised whole slide image classification is a key task in computational pathology, which involves predicting a slide-level label from a set of image patches constituting the slide. Constructing models to solve this task involves …
View article: Deep Multiple Instance Learning with Distance-Aware Self-Attention
Deep Multiple Instance Learning with Distance-Aware Self-Attention Open
Traditional supervised learning tasks require a label for every instance in the training set, but in many real-world applications, labels are only available for collections (bags) of instances. This problem setting, known as multiple insta…
View article: Whole-Slide Images and Patches of Clear Cell Renal Cell Carcinoma Tissue Sections Counterstained with Hoechst 33342, CD3, and CD8 Using Multiple Immunofluorescence
Whole-Slide Images and Patches of Clear Cell Renal Cell Carcinoma Tissue Sections Counterstained with Hoechst 33342, CD3, and CD8 Using Multiple Immunofluorescence Open
In recent years, there has been an increased effort to digitise whole-slide images of cancer tissue. This effort has opened up a range of new avenues for the application of deep learning in oncology. One such avenue is virtual staining, wh…
View article: Use of High-Plex Data Reveals Novel Insights into the Tumour Microenvironment of Clear Cell Renal Cell Carcinoma
Use of High-Plex Data Reveals Novel Insights into the Tumour Microenvironment of Clear Cell Renal Cell Carcinoma Open
Although immune checkpoint inhibitors (ICIs) have significantly improved the oncological outcomes, about one-third of patients affected by clear cell renal cell carcinoma (ccRCC) still experience recurrence. Current prognostic algorithms, …
View article: Use of high-plex data reveals novel insights into the tumour microenvironment of clear cell renal cell carcinoma
Use of high-plex data reveals novel insights into the tumour microenvironment of clear cell renal cell carcinoma Open
Although Immune Checkpoint Inhibitors (ICIs) have significantly improved Clear Cell Renal Cell Carcinoma (ccRCC) prognosis, about one third of patients experience recurrence. Current prognostic algorithms like the Leibovich Score (LS) rely…
View article: HoechstGAN: Virtual Lymphocyte Staining Using Generative Adversarial Networks
HoechstGAN: Virtual Lymphocyte Staining Using Generative Adversarial Networks Open
The presence and density of specific types of immune cells are important to understand a patient's immune response to cancer. However, immunofluorescence staining required to identify T cell subtypes is expensive, time-consuming, and rarel…
View article: Dataset of Rendered Chess Game State Images
Dataset of Rendered Chess Game State Images Open
This dataset contains 4,888 synthetic images of chess game states that occurred in games played by Magnus Carlsen. The images were rendered in Blender at different angles and lighting conditions.