Using the example of two patients' asking radiologists to give the results of unfavourable examinations just carried out, I question whether they should speak frankly in these circumstances. From a utilitarian viewpoint I discuss veracity, obligations to speak candidly and prohibitions of deception, arguing that such moral rules are contingent, not absolute. I may justly withhold information to protect an innocent third party; it may not be wrong to do so when that person is the one from whom information is withheld. I discuss whether doctors have a particular obligation to be truthful with patients, concluding that their obligation is to act for the best; if tact is in the patient's interest, any obligation of veracity may be overridden. To trust someone means assuming they protect one's interests, not that they necessarily act in specific ways. I review research indicating that not all patients wish to be given bad news. I note that when patients have false beliefs which appears medically beneficial, or when one wishes to inculcate hope, people who advocate candour may (inconsistently) become more equivocal. In examine the relationships and responsibilities of the members of the health care team in such matters. Seeing this as an example of the tension between (paternalistic) exercise of beneficence and respect for autonomy, I argue that autonomy is simply one instrumental good and that others may take precedence should the patient's interests require it. I claim that, while it is desirable that people indicate their wishes, to accede to them when this significantly harmful or imprudent is morally irresponsible, particularly should the individual not have exercised ideal deliberation. I also argue that to insist upon a putative "right to know", when this is harmful, is irrational. The answers to my question should reflect the way we wish to live together.

No description available.

The aim of the study was to develop a scintigraphic technique to quantify the uptake of cisplatin in tumours. This will allow the relationship between tumour uptake and outcome to be assessed and in future may lead to a diagnostic test which can predict patient response to chemotherapy. Platinum enriched with ¹⁹⁴Ptwas irradiated for four days<i> </i>in NRG’s TIRO 1 reactor, to produce ^195mPt. Spectral analysis of the target was performed using a calibrated HpGe detector and its constituent radioisotopes identified as ^195mPt,¹⁹⁹Au and ¹⁹²Ir. Using the detector’s intrinsic efficiency calibration their activities were estimated to be 1049, 133 and 5.8 MBq respectively. The performance of the gamma camera was tested using quality control procedures recommended by IPEM and found to be satisfactory. A torso phantom was used to determine the MDL of ^195mPt in a 2 cm diameter tumour using SPECT acquisitions (32 steps, 60 s per step). The MDL was found to be 8 ppm assuming an administered patient dose of 50 MBq and a total cisplatin dose of 105 mg. Acquisitions were attenuation corrected using the Chang method. A broad beam attenuation coefficient of 0.18 cm^-1, estimated using ²⁰¹T1C1, was used. The measured system sensitivity was (20.5±0.5) cps/MBq. This technique consistently underestimated the activity in the organs of interest; this could be due to uncertainties in the activity calculated during spectral analysis or the technique itself. This work indicates that ^195mPt -cisplatin is suitable for clinical scintigraphy and has led to the development of a clinical protocol that has been approved for a pilot study.

The use of hyperthermia as an adjuvant treatment to existing cancer therapies such as radiation therapy or chemotherapy is well established. A problem common to all clinical hyperthermia applications is the lack of temperature information available during treatment. To lend understanding to these treatments, computational modelling of the electromagnetic interactions inducing heat within the tissues has been undertaken. The work carried out in this thesis develops predictive models for the power deposition from two hyperthermia techniques: an invasive method using interstitial microwave antenna arrays; and a non-invasive method employing surface current sheet applicators, incorporating the capabilities of existing electromagnetic models and the new approaches developed in this thesis. Such models not only provide understanding of the power deposition during treatment, and hence a measure of the heating produced, but also provide analysis for the improvement of antenna/applicator design, thus improving clinical treatment methodologies. The thesis initially considers the interstitial antenna arrays, developing a model capable of solving the power deposition from an array of symmetric/asymmetric insulated antennas with arbitrary orientations and positions. An improved efficiency solution method is derived and evaluated. The scattering and absorption effects produced by insertion of such arrays within an inhomogeneous biological environment is incorporated using a conjugate gradient FFT solver, based on a spectral iterative technique. Investigations into arbitrary array orientations are carried out and new methods of power deposition control are subsequently proposed. The non-invasive current sheet applicators are approached in the same manner producing a model capable of simulating the power deposition from an array of arbitrarily orientated applicators overlying a highly inhomogeneous treatment volume.

Most of this thesis is an account of the effort to develop new methods for biomagnetic data analysis. Variations of the heart rate reflect the neural heart control mechanisms which are performed via the electrical modulation of the sinoatrial node by the autonomic nervous system. This modulation involves the interaction of several physiological mechanisms that operate on differing time scales. Using SQUID (superconducting quantum interference device) instrumentation, the fetal cardiogram can be measured with great accuracy and a high temporal resolution, thereby providing the opportunity to assess the neural function in the fetus non-invasively by analysing heart rate variability (HRV). However, a quantitative analysis of HRV requires several other physiological parameters such as blood pressure, respiration etc. to be analysed simultaneously with HRV. These parameters are obviously inaccessible in the fetus although they are routinely recorded in premature neonates treated in the intensive care units. Using a time domain correlation method, the behaviour of different HRV components was quantitatively studied for both fetuses and premature neonates and a number of consistent features were found. The correlation between neonatal HRV, respiration and arterial blood pressure was studied with the ultimate goal of constructing a numerical model of HRV. It was also observed that different types of ventilation equipment used in neonatal intensive care cause different patterns of respiration/HRV correlation, which may be indicative of the efficacy of the ventilator. Investigation of the spontaneous activity of the human brain and in particular alpha rhythm is another area where SQUID-based biomagnetic techniques can make an important contribution. In the final chapter of this work the multichannel alpha magnetoencephalogram (MEG) is considered as a sequences of MEG maps. A neural-net based algorithm for segmentation of MEG records into words is presented. Using this method three recurring words were found in an eight-second magnetoecephalogram. This could be of value for active testing of the functional role of the cortex in neurological experiments.

The use of associated particle timing based on the D-D reaction has been demonstrated for elemental analysis of bulk samples such as human body. The neutron energy of 2.8 MeV eliminates the background from organic matrices. The nanosecond timing with HPGe detector renders it possible to identify the spatial origin of the measured gamma radiation limiting the sensitive area to a single pixel. By this technique the background could be reduced by a factor of ~ 1000, but the present set-up has achieved an effective factor only in the range 20 to 100, due to the losses in generation of timing signals. The very clean gamma-spectra obtained permit the use of high efficiency scintillation detectors. Sensitivities for measuring Al, Fe, Ti, Na, Cd and Cu, presented at an extrapolated dose of 10 mSv.

Dual energy x-ray absorptiometry is the current technique used for bone mineral density measurements and recently for soft tissue analysis. The technique of dual energy absorptiometry was investigated as a method of measuring heavy elements induced or found in human body, in particular platinum and iodine, by using two energies, one energy below the k-edge and the other above it. The accuracy of iodine measurement was predicted to be 3.2, which is comparable with other methods. The discrimination between different tissue types was investigated by building a phantom and scanning it at 180 different angles using the Hologic QDR-1000/W system. The raw data files were analysed and tomographically reconstructed to produce 6 different axial images from beams with different energy spectra, produced by a combination of two x-ray voltages and three beam filters. The final part of the thesis concerns a new method to solve the beam hardening problem. A new exponential beam hardening formula was derived first by simulation work and then tested on real phantom. The measured attenuations were combined in a polynomal (quadratic) expression, which was optimised using a least squares program, and separation between different tissue components was successfully carried out on the data from the Hologic system. Measuring the bone density with the new procedure can achieve an r.m.s error of 0.26% over a wide range of body composition, and a maximum error of 1.5%, which represents a significant improvement over previous published work.

The utilization of electrical energy in the treatment of orthopaedic disorders such as fracture non-union is becoming established, and the use of pulsed magnetic fields (PMF's) to induce current in the required region is increasingly popular since it is electrically 'safe' and surgically non-invasive. Paradoxically, to date, this technique remains the least researched on a biological basis. Consequently, the effects of PMF's on embryonic-chick bone rudiments and adult human fracture non-union callus in vitro were studied. Using 7-d embryonic tibiae cultured for 7-days, it was shown that PMF's (applied continuously), significantly reduced 3H-proline incorporation whilst h droxyproline (collagen) levels were unaltered as was the incorporation of 3'b5O. PMF's also reduced cAMP levels but the incorporation of H-thymidine was unaffected. It was concluded that experimental treatment selectively inhibited chondrogenesis, and electron microscope observations supported this view. Osteotomised 12-d embryonic-chick tibiae were also used to study osteogenesis and bone repair. It was found that continuous exposure to PMF's for 18 days in vitro, significantly increased the lengths of experimental rudiments. However, there were no marked differences in the rate or pattern of repair. Incorporation of labelled proline, and its conversion to collagen specific hydroxyproline in the chondrogenic epiphyses was unaffected. However, the experimental groups showed markedly reduced levels of hydrolytic enzyme activity and hence presumably, less catabolic label release into the medium. PMF's significantly stimulated non-specific protein synthesis in the osteogenic diaphyses and this was tentatively linked to the stimulation of mineralization which was found in 15-d embryonic-chick mandibular rami exposed to PMF's for 6 days in vitro. Cartilage from adult human fracture non-union callus was cultured successfully, and in vitro differentiation observed. It was also shown, histochemically, that PMF's enhanced calcification of both hypertrophic and fibrocartilaginous callus samples. The biological and clinical implications of the results, and, in particular, the chondrogenic suppression and stimulation of mineralization/calcification processes by PMF's, are discussed.

Magnetopneumography (MPG) is a non-invasive technique of measuring occupationally acquired pulmonary dust. Most industrially generated dusts and fumes contain ferro- or ferrimagnetic particles which, after being exposed to a magnetising field, exhibit some degree of remanent magnetisation. By magnetising the thorax of an occupationally exposed subject in an external magnetising field and measuring the thoracic remanent magnetic fields using a SQUID (Superconducting QUantum Interference Device) magnetometer, information on the total lung dust burden and lung dust distribution may be obtained. The work described here concentrates on subjects who have been occupationally exposed to asbestos over extended periods as laggers at a local power station. Initial calibration of the response of the SQUID system was carried out using phantom measurements. Qualitative information regarding a subject's pulmonary dust burden and distribution is described and a comparison of the measured dust distributions with diseased tissue locations as shown by established medical imaging techniques such as Computerised Tomography (CT) and x-radiography is performed. The phenomenon of relaxation, which describes the variation in thoracic remanent magnetic field with time is also investigated. A physical interpretation of the phenomenon is given and the characteristic parameters relating to the severity of disease, and to gross localised pathological tissue changes are discussed. A mathematical model of the problem is described and computational simulations of both the whole lung phantoms, and human subjects are carried out. In this manner, a direct comparison between measurement and theory can be made. Finally, using <i>post mortem</i> lung samples known to contain asbestos fibres, a direct comparison between information derived from electron microscopy and MPG is made in order to assess the sensitivity and reliability of measurements made using the magnetometric technique.

The Monte Carlo code MCNP-4B has been utilised to investigate the origins of radiation dose perturbation in the head and neck region, arising from natural heterogeneities, the changing external contours of the patient, boundaries between materials of different atomic number and prosthetic implants. MCNP-4B was used to develop a simple model of a linear accelerator treatment head which incorporated the electron target, primary collimator, beam flattening filter and the secondary collimators. The model was used to calculate the energy spectra and angular distribution of the x-ray beam from a 4 MV Philips SL 75/5 and a 10 MV SL 15, and then tested by using these data to compute the central and off-axis x-ray beam profiles for various field sizes in water. Monte Carlo simulations using the calculated spectra were used to assess the dose distribution of treatment plans obtained in a simple heterogeneous phantom by several commercially available treatment planning systems. Practical measurements were undertaken using film dosimetry. The dose distributions were calculated for a variety of irradiation conditions designed to show the effects of surface obliquity, inhomogeneities and missing tissue above tangential beams. The results show maximum differences of 47% between some planning algorithm and film. Overall, the dose distribution obtained from film was most faithfully reproduced by the MCNP. The dose due to backscatter and lack of forward scatter from a metal implant within a head and neck phantom has been shown vary with atomic number, field size, surface obliquity, implant thickness, variations in secondary electron transport and the presence of teeth.