Index

Unsupervised detection of small hyperreflective features in ultrahigh resolution optical coherence tomography

Alzheimer’s Disease and Depression: A Bias Analysis and Machine Learning Investigation

Alzheimer’s disease is one of the most common neurodegenerative disorders that greatly impact individual and societal levels. These patients not only suffer from dementia but also from depression which can lead to more decline in cognitive abilities. However, both AD and depression have some common symptoms that make the detection of depression in Alzheimer’s extremely challenging. But several studies have used subsets of the DementiaBank database and employed different audio embeddings to detect depressive AD patients. Nevertheless, such embeddings can be biased for non-clinical factors.

Super-short Scans in Bone XRM Acquisitions

Sparse-angle CT Super Resolution using Known Operators

Matrix Operations for Applications in Quantum Annealing

Motivation:
Quantum annealing is a promising technology for quantum
computing to solve quadratic problems. D Wave makes
quantum annealers and provides an open source Python
interface: Ocean [1]. Ocean’s hybrid models do not yet
support matrix problem formulations. Current approaches are
based on SymPy [ 2]–> slow for matrix problems

Goal:
•
Speed up matrix operations for problem formulations

References:

[1] https://docs.ocean.dwavesys.com/
[2] https://www.sympy.org/

Development of a comprehensive SPECT phantom dataset using Monte Carlo Simulation

Background

Single Photon Emission Computed Tomography (SPECT) [1] is a medical imaging technique used
to study the biological function and detection of various diseases in humans and animals. Due
to the low amount of radioactivity typically used in SPECT scans, we have a lot of noise in our
SPECT acquired images, and because it is an inverse problem we do not have an exact ground truth.
For this reason we simulate objects with numerical ground truth, that will be used to create our
simulated dataset. The created dataset can then be used to train a Neural Network, analyze noise,
test multiple reconstruction techniques or evaluate the effects of acquisition geometry.
The objective of this research laboratory is to generate a large dataset of SPECT images, that will
be useful in the applications of deep learning in medical image processing.

Methods

We simulate 100 phantoms with different shapes and properties e.g. attenuation and activity maps.
Simulating simple geometric phantoms such as spheres, cubes and cylinders is the first step of
this research laboratory. In the following step we generate alphabetic letters phantoms. Last we
simulate more realistic physical phantoms like the Shepp-Logan or XCAT phantoms. To simulate
measurements of these phantoms, we use SIMIND, a Monte Carlo based simulation program [2].
SIMIND can describe different scintillation cameras, that can be used to obtain sets of projection
images of the simulated phantom. SIMIND allows the adjustment of different acquisition parameters
e.g. photon energy, number of projections, detector size, energy resolution, allowing the creation of a
comprehensive database of SPECT acquisitions in terms of geometry, and acquisition configuration.
After postprocessing the projection data, we obtain the reconstructed 3D images from the data by
applying iterative reconstruction techniques like Ordered Subset Expectation Maximization (OSEM)
and Ordered Subset Conjugate Gradient Minimization (OSCGM).

Expected Results

At the end of this research laboratory, the student shall have a deeper knowledge of Monte Carlo
Simulation (MCS) and reconstruction for SPECT imaging. Further, the student shall have created
a dataset that will be available for future projects, including denoising, reconstruction and other
image processing related tasks. Additionally, the student shall summarize their findings in a short
report and write a documentation about the database and how to use it.

References

[1] Miles N Wernick and John N Aarsvold. Emission tomography: the fundamentals of PET and
SPECT. Elsevier, 2004.
[2] Michael Ljungberg and Sven-Erik Strand. A monte carlo program for the simulation of scintillation
camera characteristics. Computer methods and programs in biomedicine, 29(4):257–272,
1989.

A Review of Diagnosis Rheumatoid Arthritis, with Evaluating Parameters of Micro-CT Scanner and Laboratory Measurements

Simulation of Spike Artifact Obstructed MR Images for Machine Learning Methods

Represent senor data of district heating network using contrastive learning

According to the European Commissions, heating and cooling in buildings and industry take up half of the total EU energy consumption [1]. One of the state-of-the-art thermal energy supply infrastructures is the district heating network, which is widespread in middle and north Europe. Thus, a robust and efficient management of energy supply trough out district heating networks has significant impact on reducing the greenhouse gas emission. In the last decades, smart sensors, such as pressure or consumption sensors, have been utilized to monitor the condition of the networks, therefore providing a solid basis for further data-driven approaches addressing various tasks, e.g., heat load prediction, automated anomaly detection etc.

With the latest advances of machine learning and deep learning, data-driven approaches have gained more importance and applicability for improving the efficiency of district heating networks. Although many works have been proposed for heat load prediction [2, 3] or anomaly detection [4] to date, a fundamental challenge remains the learning of universal representation for time series. Recently, contrastive learning paradigm has been adapted for the learning tasks with time series [5]. However, the proposed works haven´t been validated with real-world sensor data of district heating network. In this work, we are aiming at finding appropriate representation using contrastive learning for real-world sensor data acquired from district heating network. The thesis should consist of the following aspects:

Literature review of contrastive learning for time series
Analysis and understanding of real-world sensor data
Develop and implement contrastive learning framework for district heating network
Evaluate the proposed method(s)

[1] European Commission, Heating and cooling. 2018 https://ec.europa.eu/energy/topics/energy-efficiency/heating-and-cooling_en.

[2] Xue et al., District Heating Load Prediction Algorithm Based on Feature Fusion LSTM Model. Energies 2019, 12, 2122. https://www.mdpi.com/1996-1073/12/11/2122

[3] Chatterjee et al., Prediction of Household-level Heat-Consumption using PSO enhanced SVR Model, NeurIPS 2021 Workshop on Tackling Climate Change with Machine Learning, Dec. 2021. https://www.climatechange.ai/papers/neurips2021/42

[4] F. Zhang and H. Fleyeh, “Anomaly Detection of Heat Energy Usage in District Heating Substations Using LSTM based Variational Autoencoder Combined with Physical Model,” 2020 15th IEEE Conference on Industrial Electronics and Applications (ICIEA), 2020, pp. 153-158, doi: 10.1109/ICIEA48937.2020.9248108.

[5] Yue et al., TS2Vec: Towards Universal Representation of Time Series. arXiv:2106.10466. Feb.2022. https://arxiv.org/abs/2106.10466