Anders Ahnesjö's projects in Medical Radiation Physics

Proton therapy

Track structure based biological effectiveness analysis and modelling

Fernanada Villegas-Navarro, Gloria Bäckström, Nina Tilly, Anders Ahnesjö

Different radiation modalities such as low energy photons, proton beams or carbon ion beams, have different variation in biological response per dose. We have used a Monte Carlo track structure code, LIonTrack, to simulate the energy deposition around particle tracks in an event-by-event mode for different radiation qualities such as photons, protons and other light ions.

Analyses of the cluster size patterns of the energy deposition sites on a nanometric, biomolecule scale shows that ionization pattern clusters corresponding to DNA substructure sizes correlates better to biological response effectiveness than the macroscopic quantity linear energy transfer (LET) commonly used as radiation quality descriptor. These findings continue with investigations of influence from cavity size used to determine spread of the spread of energy depositions to serve as input for detector design for use in e.g. the proton beam at Skandion.

Interplay effects of scanned proton beams with patient movement

David Boersma, Erik Almhagen, Hediye Acun, Anders Ahnesjö

Modern proton therapy facilities apply a technique where a narrow beam is scanned over the tumour volume to be treated. A risk factor with scanned proton radiation is that patient movements during irradiation may interact with the scanning movement of the beams. These interplay effects may result in that parts of the tumour receive less than the planned dose, or parts of a nearby organ at risk gets overdosed.

In this project, we develop a computer based simulation environment for detailed study of the processes to aid in quality assurance of proton treatments. We are now able to simulate the effects of various breathing patterns based on the radiation transport code packages Geant4 and GATE, and dose accumulation software using CT images for different phases of detailed studies of patient movements.

Strategies to manage patient movement during radiotherapy

Dose Painting - use of functional imaging for radiotherapy dose prescriptions

Eric Grönlund, Tufve Nyholm, Anders Ahnesjö
In collaboration with Silvia Johansson

In routine radiotherapy, the dose prescription is given as a certain dose level to be given for the entire tumour target volume usually defined on anatomical CT images. Functional imaging can potentially be used to prescribe a heterogeneous dose distribution, “dose painting”, tailored to achieve equal tumour control probability with a smaller amount of radiation.

We have developed a formalism for use of retrospective data to yield optimized dose painting prescriptions demonstrating a potential for raising the cure rate significantly. We now study how these processes can be adapted to ensure delivery robustness of such dose distributions.

Probabilistic evaluation and optimization of radiotherapy treatment plans

David Tilly, Anders Ahnesjö

Uncertainties in radiotherapy delivery are routinely handled by expanding the target volume with a standardized margin to ensure adequate dose coverage. An alternative is to employ a probabilistic based planning procedure where patient specific uncertainties are explicitly considered to find the best treatment plant. This can be very computational intensive, needing several hundred simulations per patient of the interplay between anatomy and the radiation beam to sample the involved uncertainties. In the project, we have developed efficient algorithms which make the approach possible for clinical applications.

MR-Linac – integrated MR imaging during radiotherapy for target control

Tufve Nyholm

The integration of MR imaging into a treatment linac offers new possibilities for target identification and control during irradiation which can reduce the geometrical margins used to ensure dose coverage of the tumour. In collaboration with Uppsala University Hospital, which has contracted new equipment for such treatments, and the centre for image analysis, a multidisciplinary research program is set up to develop ultrafast registration algorithms for positional feedback corrections, etc.

Application of optical body surface scanning in the thorax region

Kenneth Wikström, Ulf Isacsson, Anders Ahnesjö
In collaboration with Kristina Nilsson

A problem common for several radiotherapy scenarios is to establish the accuracy and precision with which practical motion indicators can be used for in-beam tumour positioning or out-of-beam protection of organ at risk. Photogrammetric methods using optical scanning of the body contour is a promising method, which is commercially available. We study the clinical applicability of such data in particular for two patient groups: left sided breast cancer where heart protection is crucial for prevention of heart complications later in life, and lung cancer as to precisely hit the tumour.

Rectal wall protection and in vivo dose determination with a rod retractor

Andreas Johansson, Ulf Isacsson, Anders Ahnesjö
In collaboration with Gunilla Ljung, Silvia Johansson and Kristina Nilsson

Due to the proximity of the prostate to the large bowel there is risk for rectal side effects in radiotherapy of prostate cancer. The distance between the prostate gland and the rectal wall can be increased by means of a rod retractor that pushes the rectum backwards during treatment. The rectal wall can then be moved out from the high dose region close to the prostate. Also, due to the induced tension of connective tissues it is hypothesized that the prostate gland is immobilized so that smaller margins can be used while aiming beams at the prostate as to further reduce healthy tissue dose burdens. The study is implemented as a collaboration between Mälarsjukhuset Eskilstuna and Uppsala University Hospital.

Radiation safety strategies in diagnostic radiology

Hans-Erik Källman, Anders Ahnesjö

Radiation used in diagnostic imaging is one of the largest dose contributors to humans from artificial sources. Image metadata is a systematic source of information containing useful exposure indicators.

In this project, image metadata has been proved useful for dose management. Retrospective metadata, together with images from computed tomography examinations in the county of Dalarna are now used as input to Monte Carlo simulations for a more detailed analysis of patient dose distributions and organ doses. This forms a basis for improvement of dose management strategies for reduction of dose exposures on a population level.