PhD thesis topics in collaboration with the Department of Pathology:
Explainable AI in the classification of tumor cells from histological images:
Utilizing advanced image processing techniques and machine learning algorithms, healthcare professionals can accurately identify and classify tumor cells in histological images, providing valuable insights for diagnosis and treatment planning. This project aims to improve patient outcomes in oncology by improving the efficiency and accuracy of immunohistochemistry analysis.
Mögliche Themen für Bachelor-/Masterarbeiten oder Projekte:
Probabilistische Algorithmen für das Set Covering Problem:
Das Mengenüberdeckungsproblem (Set Covering Problem) ist ein NP-vollständiges Problem. Dabei wird aus einem System von Teilmengen S eines Universums U eine möglichst kleine Teilmenge von S gesucht, die U vollständig abdeckt. Eine einfache Heuristik zur approximativen Lösung des Problems basiert auf einem Greedy-Ansatz. Diese approximativen Lösungen unterscheiden sich um einen Faktor von maximal ln|U| von der exakten (aber nicht effizient berechenbaren) Lösung. In diesem Projekt soll theoretisch und experimentell untersucht werden, ob und wie die Lösung der Greedy-Heuristik durch zufällige Änderungen der Algorithmen-Eingabe beeinflusst werden kann. Dabei soll zunächst eine Methode entwickelt werden, um zufällige kleine Teilmengensysteme zu erstellen und solche zu identifizieren, für die die Greedy-Heuristik stark von einer exakten Lösung abweicht. Solche worst-case-Eingaben sollen dann im Sinne einer "Smoothed Analysis" untersucht werden, d.h es soll ausgewertet werden, inwiefern bereits eine kleine Änderung der Eingabe zu einem deutlich besseren Ergebnis führt. Erkenntnisse hieraus sollen dann in die Entwicklung einer neuen probabilistischen Heuristik zur Lösung des Mengenüberdeckungsproblems einfließen.
Bei Interesse wenden Sie sich bitte an Prof. Dr. Hans A. Kestler. Geeignete eigene Themenvorschläge können ebenfalls berücksichtigt werden.
Wissenschaftlicher Mitarbeiter (m/w/d)
Our paper "Identification of ordinal relations and alternative suborders within high-dimensional molecular data" has been published in Frontiers in Bioinformatics.
Our paper "Dealing with Class Overlap Through Cluster-BasedSample Weighting" has been published in Computers.
Our paper "A novel quantum algorithm for efficient attractor search in gene regulatory networks" has been published in Patterns.
Our paper "CLL to Richter syndrome: Integrating network strategies with experiments elucidating disease drivers and personalized therapies" has been published in Science Advances.
"Deep Learning Predicts Postoperative Mobility, Activities of Daily Living, and Discharge Destination in Older Adults from Sensor Data" has been published in Sensors.
Our paper "Robust signalling entropy estimation for biological process characterization" has been published in Briefings in Bioinformatics.
"Proposal for Using AI to Assess Clinical Data Integrity and Generate Metadata: Algorithm Development and Validation" has been published in JMIR Medical Informatics.
Our paper "Sparse keypoint segmentation of lung fissures: efficient geometric deep learning for abstracting volumetric images" has been published in the International Journal of Computer Assisted Radiology and Surgery. Please find electronical supplementary material here.
Our paper "Boolean network modeling and its integration with experimental read-outs: An interdiscipliary presentation using a leukemia model" has been published online first in Pathologie.
"Combined analysis of a serum mRNA/miRNA marker signature and CA 19-9 for timely and accurate diagnosis of recurrence after resection of pancreatic ductal adenocarcinoma: A prospective multicenter cohort study" has been published online first in the United European Gastroenterology Journal.
"Identifications of Similarity Metrics for Patients With Cancer: Protocol for a Scoping Review" has been published in JMIR Research Protocols.
"Recent Trends and Future Challenges in Learning from Data" has been published with Springer.
Our paper "Permutation-invariant linear classifiers" has been published in Machine Learning.
Our paper "Prediction of resistance to bevacizumab plus FOLFOX in metastatic colorectal cancer-Results of the prospective multicenter PERMAD trial" has been published in PLoS One.
Our paper "Segmentation-based cardiomegaly detection based on semi-supervised estimation of cardiothoracic ratio" has been published in Scientific Reports.
"Prospective study validating a multidimensional treatment decision score predicting the 24-month outcome in untreated patients with clinically isolated syndrome and early relapsing–remitting multiple sclerosis, the ProVal-MS study" has been published in Neurological Research and Practice.
Our paper "GatekeepR: an R shiny application for the identification of nodes with high dynamic impact in boolean networks" has been published online first in Bioinformatics.
Our paper "The Necessity of Interoperability to Uncover the Full Potential of Digital Health Devices" has been published in JMIR Medical Informatics.
"Multicentric pilot study to standardize clinical whole exome sequencing (WES) for cancer patients" has been published in npj Precision Oncology.
Our paper "AMBAR-interactive alteration annotations for molecular tumor boards" has been published in Computer Methods and Programs in Biomedicine.
"A protocol for the use of cloud-based quantum computers for logical network analysis of biological systems" has been published in STAR Protocols.
Our first quantum computing paper "Leveraging quantum computing for dynamic analyses of logical networks in systems biology" has been published in Patterns.