A5004997134- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- Colorectal Cancer Screening and Detection
- Generative Adversarial Networks and Image Synthesis
- Cancer Immunotherapy and Biomarkers
- Sarcoma Diagnosis and Treatment
- Music and Audio Processing
- Cervical Cancer and HPV Research
- Lymphoma Diagnosis and Treatment
- Breast Cancer Treatment Studies
- Cancer-related molecular mechanisms research
- Social Robot Interaction and HRI
- Esophageal Cancer Research and Treatment
- Colorectal Cancer Surgical Treatments
- AI in Service Interactions
- Cell Image Analysis Techniques
- Digital Marketing and Social Media
- Cancer Genomics and Diagnostics
- Aesthetic Perception and Analysis
University of Amsterdam
2020-2024
The Netherlands Cancer Institute
2022-2024
Fresenius (Germany)
2023-2024
Amsterdam University of the Arts
2020
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...
The level of tumour-infiltrating lymphocytes (TILs) is a prognostic factor for patients with (triple-negative) breast cancer (BC). Computational TIL assessment (CTA) has the potential to assist pathologists in this labour-intensive task, but current CTA models rely heavily on many detailed annotations. We propose and validate fundamentally simpler deep learning based that can be trained only ten minutes hundredfold fewer pathologist collected whole slide images (WSIs) TILs scores clinical...
Latent diffusion models (LDMs) have emerged as a state-of-the-art image generation method, outperforming previous Generative Adversarial Networks (GANs) in terms of training stability and quality. In computational pathology, generative are valuable for data sharing augmentation. However, the impact LDM-generated images on histopathology tasks compared to traditional GANs has not been systematically studied. We trained three LDMs styleGAN2 model histology tiles from nine colorectal cancer...
To give product and brand recommendations, marketers make use of conversational agents which increasingly communicate via voice rather than text. Existing research comparing the persuasiveness text showed mixed results. The quality speech synthesis employed may strongly influence consumers' responses. This study investigates to what extent a agent with pragmatically aligned prosody is more persuasive (i.e. yields positive attitude) an standard or text, whether perceived human-likeness...
We present WeakSTIL, an interpretable two-stage weak label deep learning pipeline for scoring the percentage of stromal tumor infiltrating lymphocytes (sTIL%) in H&E-stained whole-slide images (WSIs) breast cancer tissue. The sTIL% score is a prognostic and predictive biomarker many solid types. However, due to high labeling efforts intra- interobserver variability within between expert annotators, this currently not used routine clinical decision making. WeakSTIL compresses tiles WSI using...
We present TindART - a comprehensive visual arts recommender system. leverages real time user input to build user-centric preference model based on content and demographic features. Our system is coupled with analytics controls that allow users gain deeper understanding of their art taste further refine personal recommendation model. The features in are extracted using multi-task learning deep neural network which accounts for link between multiple descriptive attributes the they represent....
We present WeakSTIL, an interpretable two-stage weak label deep learning pipeline for scoring the percentage of stromal tumor infiltrating lymphocytes (sTIL%) in H&E-stained whole-slide images (WSIs) breast cancer tissue. The sTIL% score is a prognostic and predictive biomarker many solid types. However, due to high labeling efforts intra- interobserver variability within between expert annotators, this currently not used routine clinical decision making. WeakSTIL compresses tiles WSI using...