Background: Radiological imaging has existed for about 100 years and there have been significant advances in computer technology during the last 25 years. The ideal investigation should be non-invasive, repeatable, and have a sensitivity (detection of the lesion) and specificity (ability to predict the absence of disease) of 100%.
Techniques: Recent advances in ultrasonography, computed tomography (CT) scanning and magnetic resonance imaging (MRI) have enabled a more accurate demonstration of anatomical structures and better spatial resolution. This has led to the detection of smaller lesions and faster scan times, and thus new or recurrent disease is demonstrated at an earlier stage and motion artefacts are reduced. Advances in imaging techniques have also enabled percutaneous manoeuvres, ranging from diagnostic biopsy of suspicious lesions to therapeutic stenting of malignant strictures, to be performed. Magnetic resonance spectroscopy (MRS) and positron emission tomography (PET) represent a different concept. They assess in vivo tissue metabolism and provide a physiological approach which allows comparison of normal and abnormal tissue, such as tumour. The measurement of certain metabolites or isotope tracers correlates with tumour metabolism, and the response to treatment can be predicted by quantitative changes. Thus, PET scanning and MRS will probably be valuable in assessing new chemotherapeutic agents in animals and patients in vivo. Patients likely to respond to a certain drug can be selected and further studies may help determine ideal dosing regimens. Lack of fine anatomical detail on PET images can be overcome by 'fusing' the 'slice' with the corresponding CT or MRI slice, allowing accurate anatomical and physiological information to be displayed.
Conclusion: As more specific indicators of disease and the response to therapy become available, the combination of anatomical and functional imaging modalities will enable treatment to be undertaken at an earlier stage with the potential for increased survival.