The limit of phase retrieval - Project B
The problem of recovering a signal from its Fourier transform magnitude arises in many areas in engineering and science, such as optics, X-ray crystallography, speech recognition and blind channel estimation. This problem is called phase retrieval and received considerable attention recently.
In this project we will investigate three widespread formulations of the phase retrieval problem:
1. Phase retrieval from short-time Fourier transform magnitude.
2. Phase retrieval under sparsity constraints.
3. Phase retrieval with Toeplitz matrices.
The aim of the project is to explore several algorithms for phase retrieval from both numerical and theoretical aspects.
Moshe Lichtenstein | עמית אלפסי
Sub-Sampling of Spectral CT
X-Ray Computer Tomography, known as CT, is one of the main tools Doctors use today to examine patients. In every main hospital, there are CT machines that work 24/7 to provide Doctors with in-depth views of the human body. CT Scans enable them to save lives on a daily basis.
Spectral CT is a new methodology, in which patients are irradiated by several (currently only two) X-Ray wavelengths. By combining the information obtained from the detectors for each wavelength, we can better characterize materials inside the human body, and create "colorful" CT images.
Applications for Spectral CT include better detection of cancerous cells, analysis of blood vessels and much more.
In this project the students will first study the basics of computer aided tomography, advanced mathematical imaging tools and cutting edge techniques in digital signal processing for sub-sampling. The students will then learn to model a spectral CT scan, and apply sub-sampling tools in order to hopefully reduce radiation and improve its performance.
• The project is performed in collaboration with researchers and radiologists from RAMBAM.
Gil Ben-ari & Rotem Zamir
Application of ECG
The use of ECG signal, which demonstrate the heart's electric activity, is common in many applications. In mobile applications such as pacemakers, there is a significant importance to save battery energy and hence to compress the signal and sub sampled it. In this project, we will continue the work that has been done in the lab for compressing the signal and sub sampling it, and we will realize different medical applications such as pain detector for a patient under the influence of anesthesia. The project is held with cooperation of Sheba hospital and with the assistance of the cardiologist Dr. Shai Tagmen Yarden. An ambitious and fun project, which focuses on introducing the ECG signal and the world of compressed sensing (CS), in the field of signal processing.
There is a possibility for a 2 semester project to those who will be interested.
For further details, please contact Gal Mazor: firstname.lastname@example.org