A5062356615- Renal and Vascular Pathologies
- Renal Transplantation Outcomes and Treatments
- Digital Imaging for Blood Diseases
- Artificial Intelligence in Healthcare
- Mycobacterium research and diagnosis
- AI in cancer detection
- Artificial Intelligence in Healthcare and Education
- Molecular Biology Techniques and Applications
University of Amsterdam
2020-2024
Amsterdam University Medical Centers
2020-2024
BackgroundHistopathological assessment of transplant biopsies is currently the standard method to diagnose allograft rejection and can help guide patient management, but it one most challenging areas pathology, requiring considerable expertise, time, effort. We aimed analyse utility deep learning preclassify histology kidney into three main broad categories (ie, normal, rejection, other diseases) as a potential biopsy triage system focusing on rejection.MethodsWe performed retrospective,...
Abstract Accurate pathological assessment of tissue samples is key for diagnosis and optimal treatment decisions. Traditional pathology techniques suffer from subjectivity resulting in inter-observer variability, limitations identifying subtle molecular changes. Omics approaches provide both evidence unbiased classification, which increases the quality reliability final assessment. Here, we focus on mass spectrometry (MS)-based proteomics as a method to reveal biopsy differences. For MS data...
DEEPGRAFT consortium. Introduction: The gold standard and currently only means to specifically classify diseases of renal allografts is a histopathological assessment transplant biopsies according the latest Banff criteria. Some disagreements, however, exist among pathologists on certain lesions even diagnoses, which might potentially result in differences treatment subsequent graft outcome. Deep learning using convolutional neural networks (CNNs), highly effective computational approach,...