A5025038394- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Breast Cancer Treatment Studies
- Brain Tumor Detection and Classification
- Cancer Genomics and Diagnostics
- HER2/EGFR in Cancer Research
- Meningioma and schwannoma management
- Brain Metastases and Treatment
- Cancer Immunotherapy and Biomarkers
- Colorectal Cancer Treatments and Studies
- Colorectal Cancer Screening and Detection
- Glioma Diagnosis and Treatment
- Bladder and Urothelial Cancer Treatments
- Gene expression and cancer classification
- Cell Image Analysis Techniques
- Bioinformatics and Genomic Networks
- Smart Systems and Machine Learning
Universitätsklinikum Erlangen
2023-2024
Comprehensive Cancer Center Erlangen
2023-2024
Friedrich-Alexander-Universität Erlangen-Nürnberg
2023-2024
Otto-von-Guericke University Magdeburg
2019-2020
Image analysis assistance with artificial intelligence (AI) has become one of the great promises over recent years in pathology, many scientific studies being published each year. Nonetheless, and perhaps surprisingly, only few image AI systems are already routine clinical use. A major reason for this is missing validation robustness systems: beyond a narrow context, large variability digital images due to differences preanalytical laboratory procedures, staining scanners can be challenging...
The tumor-stroma ratio (TSR) has been repeatedly shown to be a prognostic factor for survival prediction of different cancer types. However, an objective and reliable determination the remains challenging. We present easily adaptable deep learning model accurately segmenting tumor regions in hematoxylin eosin (H&E)-stained whole slide images (WSIs) colon patients into five distinct classes (tumor, stroma, necrosis, mucus, background). can determined presence necrotic or mucinous areas....
The count of mitotic figures (MFs) observed in hematoxylin and eosin (H&E)-stained slides is an important prognostic marker, as it a measure for tumor cell proliferation. However, the identification MFs has known low inter-rater agreement. In computer-aided setting, deep learning algorithms can help to mitigate this, but they require large amounts annotated data training validation. Furthermore, label noise introduced during annotation process may impede algorithms' performance. Unlike H&E,...
Assessment of immunohistochemical (IHC) Ki-67 expression plays a crucial role in breast cancer diagnostics for many therapy decisions. The International Breast Cancer Working Group (IKWG) has recently proposed new global scoring method on whole tissue sections to improve accuracy, involving assessment up four representative tumor areas with negligible, low, medium and strong proliferation levels. As precise manual is time-consuming, artificial intelligence (AI) support necessary make this...
516 Background: Assessment of immunohistochemical (IHC) HER2 expression plays a pivotal role in breast cancer diagnostics. In the era HER2-low and HER2-targeted antibody drug conjugates, accurate discrimination defined IHC scores is essential. At same time, scoring suffers from poor interobserver concordance. Artificial intelligence (AI) may optimize this regard to standardization, accuracy efficiency, but previous approaches fail show required consistency across samples different sites,...
Abstract Background: Immunotherapy against PD-L1 is used for treatment of several indications Urothelial Carcinoma (UC). Recent results indicate that it likely to become part the standard first-line regimen advanced UC in near future. Accurate readout Immunohistochemical (IHC) expression therefore essential inform decisions. However, manual PD-L1-scoring prone high inter- and intra-observer variability. Artificial intelligence (AI) may help improve standardization, accuracy efficiency IHC...
The count of mitotic figures (MFs) observed in hematoxylin and eosin (H&E)-stained slides is an important prognostic marker as it a measure for tumor cell proliferation. However, the identification MFs has known low inter-rater agreement. Deep learning algorithms can standardize this task, but they require large amounts annotated data training validation. Furthermore, label noise introduced during annotation process may impede algorithm's performance. Unlike H&E, mitosis-specific antibody...
<title>Abstract</title> The count of mitotic figures (MFs) observed in hematoxylin and eosin (H&E)-stained slides is an important prognostic marker, as it a measure for tumor cell proliferation. However, the identification MFs has known low inter-rater agreement. In computer-aided setting, deep learning algorithms can help to mitigate this, but they require large amounts annotated data training validation. Furthermore, label noise introduced during annotation process may impede...
Digital pathology can be thought of as a model composed 3 main elements; classification algorithm, Graphical User Interface (GUI) and the pathologists. Currently there is only one way interaction from algorithm to pathologist. This paper, proposes an additional backward path which new feedback-based method, aimed improve performance algorithms by utilizing feedback The GUI developed for this purpose, simple adaptive different algorithms. method showed significant improvement in applied...