Brent Huisman

The world-wide home of Brent Huisman. Enjoy your stay!

Particle Physics, Proton Therapy, Treatment verification.

C++, Python, Numpy, Dicom, Qt.

Get in touch.

Projects

All this and more at Github and Sourcehut.

par2deep

par2deep: code – Analogous to the various *deep commands (md5deep, hashdeep…) this tool serves to create, verify and repair parity files in a file tree. I use par2deep to secure my photos and music across drives, machines and operating systems, and I intend to keep securing my data this way in the decades to come. First experience with GUI programming (first tkinter, now Qt), threading, and user request ;)

Postdoc at the Dutch Cancer Institute in Amsterdam, Netherlands

2017-2020

Dosia code – A viewer for Dicom plans/doses/CTs, and with a component to calculate dose independently. First experience with foreign function interfaces (calling a C lib from first Pascal and now Python). First serious experience with Qt for building GUIs.

medimage code – Wrote a library that represents a few image file formats common in medical research as Python numpy arrays. Code broken out of my evolving analyses over the years into a separate library. Available on PyPI, and used by other researchers.

cpp_headers code – First prototype of Dosia was written in C++11, my first experience writing a C++ codebase from scratch. Led to a collection of useful math and string functions, and some parsing experience.

postdoc_tools code – This package includes a number of libraries and tools created to perform my work as a postdoc, hence postdoc_tools. It’s a continuation of my phd_tools repo, but updated for Python 3 and cruft removed, new stuff added.

PhD at Creatis and IPNL in Lyon, France

2013-2017

Thesis (2017): Accelerated Clinical Prompt Gamma simulations for Proton Therapy, defended May 15, 2017 in Lyon, France. Download PDF.

Article (PMB 2016): Accelerated Prompt Gamma estimation for clinical Proton Therapy simulations, published October 14, 2016 in Physics in Medicine and Biology.

Poster (PTCOG 2016, ICTR-PHE 2016): Accelerated Prompt Gamma estimation for clinical Proton Therapy simulations

Poster (PTCOG 2017): Performance of Prompt Gamma fall-off detection in clinical simulations

Simulation code: Gate – contributed code to the open source Monte Carlo tool Gate, used for the simulation of photon and particle therapy and imaging. My main contribution is the PromptGammaTLE actor, which speeds up the convergence of the prompt gamma signal by a factor of 1000.

Simulation code: Gate Example.

Analysis code: A suite of tools to prepare, read and process Gate inputs and outputs: phd_tools. Since used by other students in their own analyses.

Proposal (2013): Towards real-time treatment control in protontherapy using prompt-radiation imaging: simulation and system optimization.

bsc

Protontherapy is an emerging cancer treatment method that consists in irradiating tumours with proton beams. Although the proton ballistics, thanks to the Bragg peak, allows delivering high dose to the tumours while limiting the energy deposited in the healthy surrounding tissues, uncertainties remain in the proton range and clinicians generally avoid direct exposure of organs at risk behind the Bragg peak. Recently, prompt gamma ray (PG) monitoring is being studied to overcome these limitations. PG are photons created by nuclear fragmentation of the target nuclei. Contrary to the gamma photons used in positron emission tomography (PET), PG are emitted almost instantaneously and cover a broad energy spectrum (up to more than 10 MeV). The works lead by the Lyon Nuclear Physics Institute (IPNL) and the IBA company (with which we collaborate) has shown that the PG depth profile can be measured with dedicated collimated gamma cameras and give information on the position of the dose distal fall off with an accuracy in the millimetre range, on a spot by spot basis. The goal of the thesis is (i) to investigate how to take full advantage of the information given by prompt radiation cameras and (ii) to optimize the camera design and acquisition protocol in clinical conditions. It will allow to provide recommendations for clinical usage of prompt radiation monitoring systems.

MSc at Nikhef in Amsterdam, Netherlands

2011-2012

Thesis: Proton Imaging with Gridpix Detectors

bsc

Hadron therapy is an upcoming cancer treatment modality with a potentially better spatial accuracy than X-ray irradiation. While an X-ray beam displays a exponential decay in intensity as it traverses matter, a hadron beam has depth profile with a sharp peak, at which point most particles will be stopped. The practical consequence of this behavior is that hadrons allow dose to be deposited more accurately than X-rays. Hadron therapy is therefore well suited to the treatment of tumors located close to sensitive organs. Imaging of the tumor and its surrounding tissue still heavily relies on X-ray CT. To construct a treatment plan for hadron therapy, the X-ray radiodensity map must be converted to a stopping power map for particle beams. This conversion introduces an inherent uncertainty of up to 3%, compromising the improved accuracy that hadron therapy potentially provides. Imaging in the same modality as the treatment would remove this conversion error. This thesis describes the Nikhef/KVI Proton Imager, a device built to measure the hadron radiodensity directly.

A prototype Python driver for a DAQ-device developed in the course of the project was the inspiration for the PySPARC project, a Python implementation for data acquisition with a HiSPARC DAQ. Helped with performance profiling and wrote the prototype from scratch.

Internship: Dicom plan converter

After obtaining my degree, I did a brief internship at the Amsterdam Medical Center, where I wrote my first Java project from scratch. This tool is a viewer for dicom RTplans, and converts plans between Agility and MLCi80 collimators. Has been used clinically for patient treatment.

BSc at the Van der Waals-Zeeman Institute in Amsterdam, Netherlands

2010

Thesis: Photoluminescence of Silicon Nanocrystals: The Origin of Different Components

bsc

In this project the photoluminescence (PL) of silicon nanocrystals is studied. It is found that the PL spectrum consists of a band-to-band component and a defect related component. A linear dependence between the excitation energy and the intensity of the defect related component is found. The fraction of excited crystals was kept constant by altering the laser beam intensity. The defect related energy level is probably introduced by oxygen bonds at the edge of the nanocrystal, and possibly lies inside the conduction band. A systematic method of measurement is proposed for future characterization of silicon nanocrystals.