A5016512818- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Artificial Intelligence in Healthcare and Education
- Osteoarthritis Treatment and Mechanisms
- Knee injuries and reconstruction techniques
- Hepatocellular Carcinoma Treatment and Prognosis
- COVID-19 diagnosis using AI
- Machine Learning in Healthcare
- Topic Modeling
- Total Knee Arthroplasty Outcomes
- MRI in cancer diagnosis
- Lower Extremity Biomechanics and Pathologies
- Medical Imaging Techniques and Applications
- Generative Adversarial Networks and Image Synthesis
- Advanced X-ray and CT Imaging
- Advanced MRI Techniques and Applications
- Lung Cancer Diagnosis and Treatment
- Radiology practices and education
- Colorectal Cancer Treatments and Studies
- Colorectal Cancer Screening and Detection
- Medical Imaging and Analysis
- Cancer Genomics and Diagnostics
- Cancer-related molecular mechanisms research
- Digital Imaging for Blood Diseases
- Natural Language Processing Techniques
Universitätsklinikum Aachen
2016-2025
RWTH Aachen University
2012-2025
Technical University of Munich
2022-2024
Klinikum rechts der Isar
2022-2024
Charité - Universitätsmedizin Berlin
2022-2024
Heidelberg University
2023-2024
Friedrich-Alexander-Universität Erlangen-Nürnberg
2024
Deutsches Herzzentrum München
2024
Humboldt-Universität zu Berlin
2023-2024
Freie Universität Berlin
2023-2024
Purpose To compare the diagnostic performance of radiomic analysis (RA) and a convolutional neural network (CNN) to radiologists for classification contrast agent-enhancing lesions as benign or malignant at multiparametric breast MRI. Materials Methods Between August 2011 2015, 447 patients with 1294 enhancing (787 malignant, 507 benign; median size, 15 mm ± 20) were evaluated. Lesions manually segmented by one radiologist. RA was performed using L1 regularization principal component...
Abstract Artificial intelligence (AI) can predict the presence of molecular alterations directly from routine histopathology slides. However, training robust AI systems requires large datasets for which data collection faces practical, ethical and legal obstacles. These obstacles could be overcome with swarm learning (SL), in partners jointly train models while avoiding transfer monopolistic governance. Here, we demonstrate successful use SL large, multicentric gigapixel images over 5,000...
Abstract Recent advances in computer vision have shown promising results image generation. Diffusion probabilistic models generated realistic images from textual input, as demonstrated by DALL-E 2, Imagen, and Stable Diffusion. However, their use medicine, where imaging data typically comprises three-dimensional volumes, has not been systematically evaluated. Synthetic may play a crucial role privacy-preserving artificial intelligence can also be used to augment small datasets. We show that...
Deep learning (DL) can accelerate the prediction of prognostic biomarkers from routine pathology slides in colorectal cancer (CRC). However, current approaches rely on convolutional neural networks (CNNs) and have mostly been validated small patient cohorts. Here, we develop a new transformer-based pipeline for end-to-end biomarker by combining pre-trained transformer encoder with network patch aggregation. Our approach substantially improves performance, generalizability, data efficiency,...
Deep learning (DL) can predict microsatellite instability (MSI) from routine histopathology slides of colorectal cancer (CRC). However, it is unclear whether DL also other biomarkers with high performance and predictions generalize to external patient populations. Here, we acquire CRC tissue samples two large multi-centric studies. We systematically compare six different state-of-the-art architectures pathology slides, including MSI mutations in BRAF, KRAS, NRAS, PIK3CA. Using a validation...
Generative models, such as DALL-E 2 (OpenAI), could represent promising future tools for image generation, augmentation, and manipulation artificial intelligence research in radiology, provided that these models have sufficient medical domain knowledge. Herein, we show has learned relevant representations of x-ray images, with capabilities terms zero-shot text-to-image generation new the continuation an beyond its original boundaries, removal elements; however, images pathological...
As large language models (LLMs) like OpenAI's GPT series continue to make strides, we witness the emergence of artificial intelligence applications in an ever-expanding range fields. In medicine, these LLMs hold considerable promise for improving medical workflows, diagnostics, patient care, and education. Yet, there is urgent need open-source that can be deployed on-premises safeguard privacy. our work, present innovative dataset consisting over 160,000 entries, specifically crafted...
Abstract Although generative adversarial networks (GANs) can produce large datasets, their limited diversity and fidelity have been recently addressed by denoising diffusion probabilistic models, which demonstrated superiority in natural image synthesis. In this study, we introduce Medfusion, a conditional latent DDPM designed for medical generation, evaluate its performance against GANs, currently represent the state-of-the-art. Medfusion was trained compared with StyleGAN-3 using...
Background Deep learning (DL) models can potentially improve prognostication of rectal cancer but have not been systematically assessed. Purpose To develop and validate an MRI DL model for predicting survival in patients with based on segmented tumor volumes from pretreatment T2-weighted scans. Materials Methods were trained validated retrospectively collected scans diagnosed between August 2003 April 2021 at two centers. Patients excluded the study if there concurrent malignant neoplasms,...
Abstract Large language models (LLMs) have shown potential in various applications, including clinical practice. However, their accuracy and utility providing treatment recommendations for orthopedic conditions remain to be investigated. Thus, this pilot study aims evaluate the validity of generated by GPT-4 common knee shoulder using anonymized MRI reports. A retrospective analysis was conducted 20 reports, with varying severity complexity. Treatment were elicited from evaluated two...
A knowledge gap persists between machine learning (ML) developers (e.g., data scientists) and practitioners clinicians), hampering the full utilization of ML for clinical analysis. We investigated potential ChatGPT Advanced Data Analysis (ADA), an extension GPT-4, to bridge this perform analyses efficiently. Real-world datasets study details from large trials across various medical specialties were presented ADA without specific guidance. autonomously developed state-of-the-art models based...
This study compares 2 large language models and their performance vs that of competing open-source models.
Oncologists face increasingly complex clinical decision-making processes as new cancer therapies are approved and treatment guidelines revised at an unprecedented rate. With the aim of improving oncologists' efficiency supporting their adherence to most recent recommendations, we evaluated use large language model generative pretrained transformer 4 (GPT-4) interpret from American Society Clinical Oncology European for Medical Oncology. The ability GPT-4 answer clinically relevant questions...
Abstract The technological progress in artificial intelligence (AI) has massively accelerated since 2022, with far-reaching implications for oncology and cancer research. Large language models (LLMs) now perform at human-level competency text processing. Notably, both image processing networks are increasingly based on transformer neural networks. This convergence enables the development of multimodal AI that take diverse types data as an input simultaneously, marking a qualitative shift...
Abstract The introduction of large language models (LLMs) into clinical practice promises to improve patient education and empowerment, thereby personalizing medical care broadening access knowledge. Despite the popularity LLMs, there is a significant gap in systematized information on their use care. Therefore, this systematic review aims synthesize current applications limitations LLMs using data-driven convergent synthesis approach. We searched 5 databases for qualitative, quantitative,...
The European Union's recently adopted Artificial Intelligence (AI) Act is the first comprehensive legal framework specifically on AI. This particularly important for healthcare domain, as other existing harmonisation legislation, such Medical Device Regulation, do not explicitly cover medical AI applications. Given far-reaching impact of this regulation sector, commentary provides an overview key elements Act, with easy-to-follow references to relevant chapters.
Root phenotyping is a challenging task, mainly because of the hidden nature this organ. Only recently, imaging technologies have become available that allow us to elucidate dynamic establishment root structure and function in soil. In tips, optical analysis relative elemental growth rates expansion zones hydroponically-grown plants revealed it maximum intensity cellular processes rather than length zone control acclimation changes temperature. Acclimation entire systems was studied at high...
Abstract Identifying image features that are robust with respect to segmentation variability is a tough challenge in radiomics. So far, this problem has mainly been tackled test–retest analyses. In work we analyse radiomics feature reproducibility two phases: first manual segmentations provided by four expert readers and second probabilistic automated using recently developed neural network (PHiseg). We test on three publicly available datasets of lung, kidney liver lesions. find consistent...
Deep learning is a powerful tool in computational pathology: it can be used for tumor detection and predicting genetic alterations based on histopathology images alone. Conventionally, prediction of are two separate workflows. Newer methods have combined them, but require complex, manually engineered pipelines, restricting reproducibility robustness. To address these issues, we present new method simultaneous alterations: The Slide-Level Assessment Model (SLAM) uses single off-the-shelf...
Abstract Artificial Intelligence (AI) can support diagnostic workflows in oncology by aiding diagnosis and providing biomarkers directly from routine pathology slides. However, AI applications are vulnerable to adversarial attacks. Hence, it is essential quantify mitigate this risk before widespread clinical use. Here, we show that convolutional neural networks (CNNs) highly susceptible white- black-box attacks clinically relevant weakly-supervised classification tasks. Adversarially robust...
The text-guided diffusion model GLIDE (Guided Language to Image Diffusion for Generation and Editing) is the state of art in text-to-image generative artificial intelligence (AI). has rich representations, but medical applications this have not been systematically explored. If had useful knowledge, it could be used image analysis tasks, a domain which AI systems are still highly engineered towards single use-case. Here we show that publicly available reasonably strong representations key...