Peter van der Meulen

I would like to welcome to you all to my MR Introduction Courses. As described, I will address many different aspects of the MR system, physics, technology, and applications. I do encourage questions and interactions, so I can provide additional explanation and information. I try to keep up to your pace, but I’m sure that you may sometimes think that I’m too fast; I know that some concepts take time to grasp. So do not hesitate during the courses to ask questions!

In order not to spend too much time on personal introduction during the courses, I share some insights about myself via this mail, so you know me a bit better, know what you can expect from me and understand what my technical areas of expertise are.

I studied Technical Physics at the TU Delft, started with MR in 1980 and first worked on Multiple Quantum Spectroscopy, which is a form of 2D spectroscopy. Although the related physics are quite different from MR imaging, the related acquisition and processing is surprisingly similar to 2D MR imaging. After spectroscopy I switched to MR imaging, very new at that time. Spine Echo was already known, but I made a predecessor of a fast sequence, MEMS (Multi Echo-Multi-Shot), nowadays known as TSE (Turbo Spin Echo), using the very slow shim gradients in a small bore (4 cm) 7T spectrometer: A very good start to learn some MRI basics and technological challenges.

In 1983 I started at Philips, that was just about to ship the first superconductive 0.5T systems. At that time, the only available scan technique was Spin Echo, that could make one image of only one slice in 2 to 4 minutes, with fixed echo time (50 ms). I first added more flexibility, added multi-echo, using composite RF pulses, and proposed/invented artefact reduction techniques (with phase cycling, subtraction techniques, additional gradients and optimized RF pulses). I also invented various fast preparation phases to calibrate the scanner (such as the PO-phase for Power optimization phase, the AS-phase for Automatic shimming and tune gradient to compensate eddy current effects) and made the first 3D Spin Echo version.

In 1984 I made the first product version of Multi Slice Spin Echo, including the mixed version with inversion recovery, closely together with an electronic engineer, who took care of the required accurate control of synthesizers of the scanner.

In 1985, working on sodium (23Na-) imaging, I invented FFE, a fast gradient echo technique, allowing imaging in seconds rather than minutes, which also opened a new range of scan techniques for different contrasts (T1-weighted, T2*-weighted) and dynamic studies, e.g. multi-phase cardiac scans. In parallel, I developed a course for the methods-development environment at that time, ppwfrf (pulse program, (gradient-) waveform, rf (-pulses)), that provided the input for the Nicolet sequencer and the waveform fifo’s.

In 1986-1988 the scanner software was completely innovated, because of the introduction of a new Data Acquisition System (the ADAS), and I created, with the help of a SW architect, a new environment to program MR pulse sequences, GOAL, Gyroscan’s Object Oriented Acquisition Language. This is an object oriented way for MR method developers and researchers to build pulse sequences, with as much abstraction from HW specific details as possible, but on the other hand always allowing to use the maximum capabilities of the available scanner hardware (e.g. gradient and RF power) for any scan technique.

I made the first versions of the PDF (Parameter Definition File), which is the part of the Methods software that runs on the host computer and is connected to the scanner user interface. Each
parameter modification by the user was (and is) directly translated to calculate the optimal pulse sequence, to validate the input and to show the effect of the change (e.g. on SNR, Te, TR, scan time, etc.). Since hardware characteristics are also captured in parameters, the software is completely portable to any system configuration. The run-time part of the methods software, the MPF (Measurement Program File) directs the run-time calculation of the sequences on the Scan Control Computer (initially the ADAS), to enable fast reaction of the scanner on events during the scan (e.g. because of behavior of or in the patient) and included a run-time interface with labels to the reconstructor. Over the years, the programming language and the interfaces to the main scanner software have been changed, but the foundation of this environment is still in place.

After that I more and more combined the work on pulse sequence optimizations (such as TSE) with system design, had the role of lead designer of the project responsible for Gyroscan-NT series, and became group leader System Design. From 1996 to 2004 I was the head of various R&D departments. First Systems (with System Design, System Verification, Test Systems and SW Management), then one of the two SW departments: Scan Software (including Methods, Recon, System SW (DCL), Scanner SW), followed by one of the Hardware Departments: Magnets & Mechanics (including magnets, gradient coils, patient supports, covers and cabling). In the role of SW manager, I also was SPI (SW Process Improvement) leader, and successfully reached CMM (Capability Maturity Model) level 3, that aims to have better organized and predicable SW development organization.

After being a line manager for many years, I wanted to spend more time on content and technology again and became system architect, leading the advanced development program for a while.
At the end 2006 I thought it was time for a change, left MR and went to Apptech (later named PInS, now known as PES, Philips Engineering Solutions), an internal engineering organization in Philips, providing services and support for development, both inside and outside of Philips. Main role was to investigate opportunities for the electronics and software group in Healthcare applications. Very interesting to work in a completely different organization and allowed to learn more about other healthcare applications and about more hybrid system combinations (like MR and therapy). At a business unit, like MR, there is always more work than can be staffed. Since Apptech’s activities determined by paying customers, projects could be short and continuity of work on specific topic is never guaranteed.

In 2010 I went back to MR, first in the Advanced Development group that we had at that time, and shortly after that became Principal Systems Engineer/System Architect. After being the lead designer of various projects, I was asked for the function of lead designer/architect of the newly founded MR Collaboration Office. A relatively small team, working on special projects, closely together with suppliers, start-ups, universities, and customers. Since end of 2022 the team is integrated with Clinical Science and named the Co-Creation team. We are responsible for the Philips 7T systems, for which we are currently making a large upgrade. In addition, we enable investigations on a 0.6T scanner, work with a start-up on a completely new scanner for neonatal patients, create an integrated MR-PET system, and work on combinations of MR and therapy and surgery. We are still extending the use of our general backend and even cover the magnetic
field range from 50 mT (so 2 MHz for 1H) to 14T (600 MHz).

Private situation: Living in Best with my wife, who is general practitioner. Two kids, boy (30 y) and girl (28 y), one dog and a cat. I like to play tennis, walking, rock climbing (via ferrata), travelling.

I’m looking very much forward to the courses, interacting face-to-face and supporting you in increasing your MR knowledge!

Peter van der Meulen