CHAIR OF

COMPUTATIONAL

INTELLIGENCE

CIDA is a newly established unit at our chair. Here we work on computational intelligence methodologies for data analytics in bioinformatics and biology. Our goal is to develop new methodologies for behavioural, medical and veterinary data. 

Outlined below are additonal details of the various projects worked on within the group.

Behavioural Data:

Mice on a Treadmill Project

The learning behaviour of mice is is explored using a combination of a reward system and timing. The food-deprived mice run on a treadmill, and at a specific point a liquid treat is released. The mice must learn when the treat is available as well as how to be accurate when licking the treat.  Variables, such as lick accuracy, lick precision, success/lap and licks/lap, are collected, and this is used to explore the learning behaviour of the mice. 

Collaborators: 

Leibniz-Institut für Neurobiologie:

• Felix Kuhn
• Prof. Dr. Stefan Remy

Associated Papers:

• Sanaz Mostaghim, Qihao Shan, Christiane Desel, Alexander Duscha, Aiden Haghikia, Tobias Hegelmeier, Felix Kuhn, Stefan Remy
• Medical and Behavioral Knowledge Discovery using Multi-Objective Analysis

• 2023 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), Eindhoven, Netherlands, 2023, pp. 1-8, doi: 10.1109/CIBCB56990.2023.10264881.

Medical Data:

Parkinson's Disease Study

Parkinson's disease has several symtmpoms that may appear in some individuals with the disease, and may not show in others. Due to this, several criteria are used to diagnose Parkinson's. The problem is difficult to study using uni-dimensional quantification, and therefore it is likely that multi-objective data analysis is a good candidate for analysis of the data. This applicability is explored in this project.

Collaborators: 

Otto-von-Guericke-Universität Magdeburg Medizinische Fakultät Universitätsklinik für Neurologie:

• Dr. rer. nat. Christiane Desel
• Alexander Duscha
• Dr. med. Tobias Hegelmaier
• Univ.-Prof. Dr. med. Aiden Haghikia

Associated Papers:

• Sanaz Mostaghim, Qihao Shan, Christiane Desel, Alexander Duscha, Aiden Haghikia, Tobias Hegelmeier, Felix Kuhn, Stefan Remy
• Medical and Behavioral Knowledge Discovery using Multi-Objective Analysis

• 2023 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), Eindhoven, Netherlands, 2023, pp. 1-8, doi: 10.1109/CIBCB56990.2023.10264881.

• Sanaz Mostaghim, Qihao Shan, Christiane Desel, Alexander Duscha, Aiden Haghikia, Tobias Hegelmeier 
• Unfolding the Variability of Clinical Data in Parkinson Treatment using Multi-objective Analysis
• 2023 IEEE Conference on Artificial Intelligence (CAI), Santa Clara, CA, USA, 2023, pp. 120-121, doi: 10.1109/CAI54212.2023.00058

Aneurysm Project

Details coming soon.

Collaborators: 

Otto-von-Guericke-Universität Magdeburg Medizinische Fakultät:

• PD Dr. med. B. Neyazi

Associated Papers:

Long-Covid and Fatigability Project

This project aims to characterise fatigue after Covid-19 infections using multi-objective data analysis. 

Collaborators: 

Otto-von-Guericke-Universität Magdeburg Medizinische Fakultät Universitätsklinik für Neurologie:

• Prof. Dr. phil. Tino Zähle
• Magdalena Mischke

Associated Papers:

Veterinary Data:

Lameness in Dogs Study

This project aims to create a program that can diagnose lameness in dogs automatically using a combination of multi-objective data analysis and machine learning to replicate the clinical process used for a diagnoses. 

Collaborators:

Tierärztliche Hochschule Hannover:

• Sebastian Schmidt

Synthetic Data:

Delta Project

This project explores the effect of noise during the process of multi-objective data analysis and clustering, as well as explores the effect of loosening the dominance relation when assigning the ranks during non-dominated sorting to be better outside of a delta amount around the points being compared. 

Collaborators: 

• Internal project