PhD candidate to develop Deep Learning models for neuro-endocrine tumours of the lung (PNET-AID...

The Department Pulmonary Medicine in close collaboration with the department of Pathology and Clinical Bioinformatics, is seeking a PhD student with experience in the application of Machine and Deep learning models as part of our PNET-AID project. This is an excellent opportunity to develop deep learning technology to have direct impact on cancer research, with the aim to improve personalized treatment for patients.

The Department of Pulmonary Medicine is a large tertiary referral center for rare disease among which neuro-endocriene tumors of the lung and has a track record in research on this topic. The Department of Pathology and Clinical Bioinformatics is implementing deep learning methods to support the diagnostic and research work of pathologists and clinicians, including computer-aided diagnosis in a routine clinical setting. Our PatHology Artificial iNtelligence plaTfOrM (PHANTOM) is an ongoing multidisciplinary cross-department project to deliver Deep Learning (DL) methodologies for diagnostic pathology, translational research and personalized medicine at Erasmus MC. Our DL approaches have been applied to determine diagnostic, prognostic and predictive biomarkers in collaborative projects, including oncology (e.g. pancreatic, lung), nephrology and dermatology. DL research application to histopathology and multi-omics have been developed by Dr. Li, Dr. Stubbs and Dr. von der Thüsen, driven by the requirements of our pathologists and our research laboratories (e.g. Tumour Immune Pathology Lab) to deliver personalized medical treatment decision support.

Your role is to develop and implement DL algorithms in the field of pulmonary neuroendocrine tumours (PNETs) in the Hanarth Fonds-funded PNET-AID research program and in close collaboration with Maastricht University Medical Center (MUMC+). You will apply artificial intelligence techniques to analyze digitalized (immuno)histopathological slides and also to combine more data types in one predictive model where necessary. These methods are aimed to assist decision making by pathologists and clinicians during tissue-based diagnostics and personalized treatment of PNET patients. This is an excellent opportunity to develop and apply cutting-edge DL technology to have direct impact on cancer research, thus supporting increasingly personalized treatment for our patients.