Department of Computer Science at UH

University of Houston

Department of Computer Science

In Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy

Son M. Hoang

Will defend his PhD dissertation proposal

Pattern Recognition and Machine Learning for Tracks of Heavy-ion Particles in Timepix Detectors


One fundamental question in the search to develop a Space Radiation Dosimeter using the Timepix chip from the Medipix2 Collaboration is how to calculate Dose-equivalent and identify the nature of the source of radiation from the raw Timepix outputs. Dose is defined as the energy deposited by a source of radiation per unit mass of traversed matter. Dose-equivalent is the most common form to express the biological effects of the dose caused by a particular type of particle for radiation protection purposes. The raw Timepix output is generated by the Medipix2 Timepix chip that is attached to an overlying Si sensor layer using the bump-bonding technique. The resulting detector is a hybrid semiconductor CMOS-based pixel detector made of 256 x 256 square pixels with the readout electronics for each pixel embedded within the footprint of each 55 μm square pixel. Based on Timepix detectors, this research introduces a method for calculating Linear Energy Transfer (LET), identifying the source and velocity of ionizing radiation by using pattern recognition and machine learning.

Our analysis makes use of data taken in beams of heavy ions at HIMAC (Heavy Ion Medical Accelerator Facility) in Chiba, Japan. The results will be tested with both experimental data at HIMAC and real data coming from NASA-International Space Station (ISS). Our main challenge lies in extracting relevant features that can facilitate Dose-Equivalent calculation and discriminating among different sources and energy of radiation.


Date: Wednesday, November 7, 2012
Time: 3:00 PM
Place: 550-PGH

Faculty, students, and the general public are invited.
Advisor: Dr. Ricardo Vilalta