The current applications of neural networks to in vivo medical imaging and signal processing are reviewed. As is evident from the literature neural networks have already been used for a wide variety of tasks within medicine. As this trend is expected to continue this review contains a description of recent studies to provide an appreciation of the problems associated with implementing neural networks for medical imaging and signal processing.
While EIT images can produce clinically useful qualitative information, the extraction of quantitative data is essential in clinical monitoring. In the case of imaging of the thorax the parameters available relate to cardiac activity and pulmonary perfusion. Imaging the relatively small changes in the resistivity of the lungs due to pulmonary perfusion in the presence of noise and the larger ventilation component is difficult. Suggested solutions involve multiple time averaging of cardiac gated data or reconstructed images. The required number of data frames for this type of processing is large (at least 100 cardiac cycles). Because the ventilation and perfusion components of the resistivity signals are well separated in the frequency domain, they can be differentiated by filtering. We report the results of this analysis which requires a data collection period of typically 15 s.
(ICA). These studies have shown a rapid decline in mean blood flow velocity by up to 60-75% of pretreatment levels within five minutes of the completion of the injection. A decrease in both the peak systolic and diastolic velocities accounted for the fall in mean blood flow velocity. The findings of one study,7 using 'blind' isonation of the ACA with continuous wave Doppler, suggested that infusion of indomethacin over 20 minutes avoids the fall in cerebral blood flow velocity (CBFV) seen with more rapid administration. However, a more recent study using near infrared spectroscopy8 did not confirm these findings and failed to show a difference in the fall in total cerebral blood flow when injection of indomethacin over five minutes was compared with slow infusion (20-30 minutes).To date there have been no studies of the effects of indomethacin on different regional areas of the cerebral circulation of human infants. In the following report both the anterior and the middle cerebral arteries were studied using duplex scanning, to determine whether prolonged infusion (over 30 minutes) of indomethacin was associated with changes in CBFV and if so whether regional differences in CBFV could be demonstrated.
The electrocardiographic and anticholinergic effects of trazodone (150 mg) and imipramine (75 mg) were investigated in 8 healthy volunteers. Both agents increased the QTc interval and decreased T wave height, but the effects occurred earlier with trazodone (from 30 min onwards) than with imipramine (150 and 180 min after dosing). Both drugs decreased heart rate, imipramine at 30 and 60 min and trazodone at 90 min. After 120 min, heart rate began to increase with imipramine an effect which was not seen with trazodone. Salivary volume was significantly decreased by imipramine at 120 and 180 min whereas trazodone did not influence salivary volume. Plasma levels of trazodone and imipramine were significantly related to the decrease in T wave amplitude. The increase in QTc interval correlated significantly with the plasma level of imipramine. These results suggest that trazodone, like the tricyclic antidepressants prolongs ventricular repolarization; but, in contrast to imipramine, it does not have anticholinergic activity.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.