Most of the tissue substitute materials currently used inclinical radiation dosimetry are designed to simulate muscle or bone when irradiated with photons A few materials have been developed for neutron dosimetry, but substitutes speci fically designed for beams of high energy charged particles are not to be found in the literature. This thesis deals with the formulation and manufacture of, tissue substitutes for particulate radiations and the subsequent application of these substitutes in dlectron, pion, proton and neutron dosimetry. The method of "elemental equivalence" was used and over 80 solid, gel and liquid substitutes have been produced2 which simulate the most important tissues (adipose, blood, bone, muscle, etc), body organs (brain, lungj etc) and tissue components (fat, protein, water). Most of these materials are "tissue equivalent" and are useful for all types of radiations. The compilation of selected chemical compounds (compound library) used for the formulatign; and the computer programs written for the theoretical evaluation of the new materials are described and discussed. The experimental comparison of some selected substitutes with the corresponding real tissues, using fast neutrons, high energy protons, cobalt-60 gamma rays and 120 kVP X-rays., verified the high precision of the simulation procedures. The results of depth dose measurements in various tissue substitutes ý as well as water, using 7.5 MeV neutrons 150 MeV protons, 70 MeV negative pions , 10 MeV electrons and cobalt-60 gamma rays are presented. The effect of tissue heterogeneities on the dose distributions from thesý radiations was investigated. Isodose shift factors for air, lung, fat and bone were derived for all the above radiation modalities and detailed lung correction factors were measured for 7.5 MeV neutrons and cobalt-60 gamma rays. In view of the proposed use of the 160 MeV proton beam of the Atomic Energy Research Establishment (Harwell, JJ. K) for patient treatment, a complete series of pre-therapeutic measurements was performed with this proton beam facility using the new materials, and the results are presented and discussed in detail. Finally, the applications of the new substitutes in other practical clinical aspects are described and some examples of such applications given.

A 700 keV Van de Graaff accelerator was used to accelerate protons onto solid targets of different light elements to produce ultrasoft, characteristic X-rays (< 5 keV). The proton energies were calibrated using the (p, y) resonances at 633 keV in Aluminium and at 340 and 483 keV in Fluorine. The X-ray emission characteristics of Aluminium, Carbon, Gold, Silicon/Carbon, Silicon/Nitrogen and Titanium/Boron were studied as a function of incident proton energy, angle of inclination of the target (30° - 60° to the proton beam) and angle of detection of the X-rays (40° - 130° to the beam). Detection of the X-rays was achieved using a gas-flow proportional counter directly coupled to a low-noise pre-amplifier. The resulting spectra, recorded on a multichannel analyser, were well fitted by linear combinations of single Gaussian curves to give peak position (X-ray energy), width and area (X-ray intensity). Carbon contamination of the target surface was studied in detail for the Aluminium target. A number of low beam currents onto the target (10 - 70 nA) were used for total irradiation times of up to 17 hours in order to establish the degree of overall X-ray energy mixing. The information gained from the study of both the Carbon contamination and the X-ray emission characteristics was used to propose practical optimum conditions for the production of ultrasoft X-rays by proton bombardment in their application to biological and biochemical irradiations. A computer code, capable of following the electron track histories resulting from ultrasoft X-ray interactions has been used to compare the details of such energy deposition with the results of mammalian cell irradiations made at the M. R. C. Radiobiology Unit for a number, of different ultrasoft X-ray energies. Such a-comparison has been used to try to identify the mechanisms of radiation action. Included in this work is the application of the computer code to a variety of. characteristic X-ray photon energies, thus extending the available, calculated data.

A model based on Monte Carlo techniques is developed to transport ionising radiation through the radiation head of a 6MV linear accelerator fitted with multileaf collimators Major emphasis is given to the detailed geometrical descriptiqn of the multileaf collimator. The model produces dose distributions in water from photon beams defined by the jaws and the multileaf collimator. The model accounts for contaminant electrons in the photon beam, off-axis x-ray radiation originating at the collimator and the transmission and penumbra effects of the side planes and front face of the leaves in the multileaf collimator Dose distributions in water calculated by the model are compared with experiment using lonisation chambers, diodes and film and found to be within 1 5% The transmission and the penumbra of the multileaf collimator leaves calculated by the Monte Carlo model are compared with experiment and found to be in good agreement

Thesis (Ph.D.)--Massachusetts Institute of Technology, Dept. of Medical Engineering and Medical Physics, 1984. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND SCIENCE. / Includes bibliographical references. / by Bruce Robert Rosen. / Ph.D.

Thesis (Ph.D.)--Massachusetts Institute of Technology, Dept. of Medical Engineering and Medical Physics, 1984. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND SCIENCE. / Includes bibliographical references. / by Elazer Reuven Edelman. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Medical Engineering and Medical Physics, 1988. / Includes bibliographical references. / by Richard Glenn Stevens Spencer. / Ph.D.

Medical imaging plays an important role in clinical procedures related to cancer, such as diagnosis, treatment selection, and therapy response evaluation. Magnetic resonance imaging (MRI) is one of the most popular acquisition modalities which is widely used in brain tumour analysis and can be acquired with different acquisition protocols, e.g. conventional and advanced. Automated segmentation of brain tumours in MR images is a difficult task due to their high variation in size, shape and appearance. Although many studies have been conducted, it still remains a challenging task and improving accuracy of tumour segmentation is an ongoing field. The aim of this thesis is to develop a fully automated method for detection and segmentation of the abnormal tissue associated with brain tumour (tumour core and oedema) from multimodal MRI images. In this thesis, firstly, the whole brain tumour is segmented from fluid attenuated inversion recovery (FLAIR) MRI, which is commonly acquired in clinics. The segmentation is achieved using region-wise classification, in which regions are derived from superpixels. Several image features including intensity-based, Gabor textons, fractal analysis and curvatures are calculated from each superpixel within the entire brain area in FLAIR MRI to ensure a robust classification. Extremely randomised trees (ERT) classifies each superpixel into tumour and non-tumour. Secondly, the method is extended to 3D supervoxel based learning for segmentation and classification of tumour tissue subtypes in multimodal MRI brain images. Supervoxels are generated using the information across the multimodal MRI data set. This is then followed by a random forests (RF) classifier to classify each supervoxel into tumour core, oedema or healthy brain tissue. The information from the advanced protocols of diffusion tensor imaging (DTI), i.e. isotropic (p) and anisotropic (q) components is also incorporated to the conventional MRI to improve segmentation accuracy. Thirdly, to further improve the segmentation of tumour tissue subtypes, the machine-learned features from fully convolutional neural network (FCN) are investigated and combined with hand-designed texton features to encode global information and local dependencies into feature representation. The score map with pixel-wise predictions is used as a feature map which is learned from multimodal MRI training dataset using the FCN. The machine-learned features, along with hand-designed texton features are then applied to random forests to classify each MRI image voxel into normal brain tissues and different parts of tumour. The methods are evaluated on two datasets: 1) clinical dataset, and 2) publicly available Multimodal Brain Tumour Image Segmentation Benchmark (BRATS) 2013 and 2017 dataset. The experimental results demonstrate the high detection and segmentation performance of the III single modal (FLAIR) method. The average detection sensitivity, balanced error rate (BER) and the Dice overlap measure for the segmented tumour against the ground truth for the clinical data are 89.48%, 6% and 0.91, respectively; whilst, for the BRATS dataset, the corresponding evaluation results are 88.09%, 6% and 0.88, respectively. The corresponding results for the tumour (including tumour core and oedema) in the case of multimodal MRI method are 86%, 7%, 0.84, for the clinical dataset and 96%, 2% and 0.89 for the BRATS 2013 dataset. The results of the FCN based method show that the application of the RF classifier to multimodal MRI images using machine-learned features based on FCN and hand-designed features based on textons provides promising segmentations. The Dice overlap measure for automatic brain tumor segmentation against ground truth for the BRATS 2013 dataset is 0.88, 0.80 and 0.73 for complete tumor, core and enhancing tumor, respectively, which is competitive to the state-of-the-art methods. The corresponding results for BRATS 2017 dataset are 0.86, 0.78 and 0.66 respectively. The methods demonstrate promising results in the segmentation of brain tumours. This provides a close match to expert delineation across all grades of glioma, leading to a faster and more reproducible method of brain tumour detection and delineation to aid patient management. In the experiments, texton has demonstrated its advantages of providing significant information to distinguish various patterns in both 2D and 3D spaces. The segmentation accuracy has also been largely increased by fusing information from multimodal MRI images. Moreover, a unified framework is present which complementarily integrates hand-designed features with machine-learned features to produce more accurate segmentation. The hand-designed features from shallow network (with designable filters) encode the prior-knowledge and context while the machine-learned features from a deep network (with trainable filters) learn the intrinsic features. Both global and local information are combined using these two types of networks that improve the segmentation accuracy.

No description available.

Dosimetry is the methodology of determining the amount of radiation energy imparted in matter and volume. Although several techniques and devices are available for use in both laboratory and clinical settings, most rely on certain conditions, assumptions and approximations to convert the energy into radiation dose. The many uncertainties from current techniques are introduced due to the material differences between the sensitive detector volume and the phantom material, typically water. The aim of this thesis is to use the water sample itself to detect the amount of radiation energy that has been imparted upon it. Radiation energy absorbed by the sample is ultimately converted into heat, raising the temperature of the sample and changing the refractive index property. The refractive index change results in a shortening of the optical path length and as a result, light passing through the sample experiences a phase change. Phase information cannot be directly measured, as it is merely a property of light wave propagation, thus another technique must be used. Digital holographic interferometry was employed to capture snapshots of the sample’s changing state over time, and when compared with a reference, the interference phase information was extracted and used to calculate the refractive index change, which can then be related to radiation absorbed dose. The aim of this research was to design and build interferometry setups using holographic interferometry to determine the refractive index change induced by radiation and to explore the possibilities of using fibre optics. Experiments were conducted on the setups to determine the validity of the method and the accuracy of the system. With external heating sources in the forms of an open flame and infrared laser, we could see distinct heating patterns formed in the phase images. The phase allowed the calculation of the temperature and therefore energy from the change in refractive index, but was limited to phase differences within 2π between the images, due to wrapped phases. In the stability tests, we demonstrated the accuracy of the system and found it was heavily influenced by the amount of vibration in the vicinity. In the short term, a standard deviation of 0.015 degrees was recorded but a larger standard deviation of 0.078 degrees was measured in the longer term. We can be confident of the temperature measurements to within 0.1 of a degree, equal to hundreds of Grays in radiation dose, however this is not sufficiently accurate for dosimetry. Future work may include improving accuracy by reducing the vibration in the system.

No description available.