Long-term follow-up of a randomized post-induction therapy trial in acute myelogenous leukemia (a Southeastern Cancer Study Group trial)

A phase III clinical trial was designed to determine if more intensive induction and consolidation therapy for acute myeloblastic leukemia increases the remission rate and prolongs survival. A minor objective was to determine if the use of non-cross resistant drugs was more effective than the same drugs used for induction. Patients with untreated leukemia between the ages of 15 and 50 were given daunorubicin 45 mg/m2 for the first 3 days of a 10-day continuous infusion of cytosine arabinoside, initially at a dose of 2000 mg/m2 but reduced to 100 mg/m2 because of toxicity. Those under 36 achieving a complete remission and with an histocompatible donor were assigned to a transplant arm. The rest were randomized to receive one of three consolidation arms: A, cytosine arabinoside, 200 mg/m2 daily for 7 days and daunorubicin 45 mg/m2 daily for 3 days for three courses; B, one course as in Arm A followed by amsacrine, 120 mg/m2 daily for 5 days followed by a 5-day continuous infusion of azacytidine, 150 mg/m2/day; C, thioguanine and cytosine arabinoside, 100 mg/m2 every 12 h and daunorubicin 10 mg/m2 daily for 5 days for three courses followed by four maintenance courses of cytosine arabinoside, 100 mg/m2 daily for 5 days and daunorubicin, 45 mg/m2 for 2 days every 13 weeks. From 1981 to 1986, 398 eligible patients were enrolled and 219 achieved a complete remission. The initial induction dose of cytosine arabinoside was reduced after five of 29 patients exhibited fatal gastrointestinal toxicity. Only 11 patients were assigned to the transplant arm. There were no significant differences in the consolidation arms. The 5 year disease-free survivals were 38, 31 and 27% in arms A, B, and C respectively. Intensive consolidation therapy with the same or different drugs used in induction was as effective as lower dose consolidation followed by maintenance therapy.     

A real-time neural system for color constancy

A neural network approach to the problem of color constancy is presented. Various algorithms based on Land's retinex theory are discussed with respect to neurobiological parallels, computational efficiency, and suitability for VLSI implementation. The efficiency of one algorithm is improved by the application of resistive grids and is tested in computer simulations; the simulations make clear the strengths and weaknesses of the algorithm. A novel extension to the algorithm is developed to address its weaknesses. An electronic system that is based on the original algorithm and that operates at video rates was built using subthreshold analog CMOS VLSI resistive grids. The system displays color constancy abilities and qualitatively mimics aspects of human color perception.     

Intelligent identification and control for autonomous guided vehicles using adaptive fuzzy-based algorithms

Fuzzy logic was first suggested as the mechanism by which humans drive cars. This paper addresses the use of fuzzy logic and algorithms towards the intelligent autonomous motion control of land vehicles. To cope with vehicle complexities, internal parametric changes, and with unpredictable environmental effects, the controllers that are presented, whilst heuristic in nature, are self-organizing or self-learning in that they generate automatically by observation an experiential rule base that models the vehicle, and via an appropriate performance index an optimal control rule base that is robust to large parametric changes. The methodology presented is applicable to any complex process which is too difficult to model or control using conventional methods, or which has relied on the experience of a human operator. An overview of fuzzy logic and static fuzzy logic control (akin to expert systems) is provided, together with illustrative examples.     

Knowledge elicitation using more than one expert to cover the same domain

Knowledge elicitation is accepted as being one of the most problematic areas in the creation of a knowledge-based system.A large amount of research has already concentrated on finding more efficient and effective techniques for eliciting knowledge from an individual expert. However, little attention has been given to the involvement of more than one source of expertise in knowledge-based system development.This paper is based on the authors' practical experience gained when developing a knowledge-based system for the conceptual design of bridges. It shows that the use of more than one expert throughout the knowledge elicitation process can improve both the efficiency of the approach and the quality of the knowledge acquired.     

Missed opportunities for immunisation at hospitals in the western Cape--a reappraisal

Immunisation practices were examined at 6 hospitals in the western Cape during the latter half of 1992 to determine whether these practices had improved subsequent to the February 1991 resolution of the Health Matters Committee (HMC) on immunisation in hospitals, and since a similar study was undertaken in 1990. Exit interviews were conducted with the escorts of all children aged 3-59 months who attended the study hospitals on the days designated for the study. In the second study, 88 of the 311 children studied (28.3%) were in need of immunisation on arrival, but only 12 of the 88 (13.6%) were immunised during the hospital visit. There was no evidence of an increase in requests to see children's Road-to-Health cards (37.1% compared with 35.2% previously). The incidence of missed opportunities for measles immunisation in children aged 6-59 months remained unacceptably high (51.4% compared with 63.7% previously, when a strict definition was used; and 15.7% compared with 18.1% previously, when a lenient definition was used). Health authorities at all levels need to take urgent action to address the problem of missed opportunities for immunisation at hospitals.     

Cerebral protection by intermittent reperfusion during temporary focal ischemia in the rat

Temporary arterial occlusion has been routinely used as an adjunct in intracranial aneurysm surgery. This has commonly been performed using a protocol of multiple short periods of occlusion alternating with periods of restoration of normal circulation. Recently, the logical basis of this method has come under scrutiny. There is extensive experimental evidence to suggest that repetitive, brief periods of global ischemia may cause more severe cerebral injury than an equivalent single period of global ischemia. Only recently has this issue begun to be addressed with regard to focal ischemia. Hence, despite the common use of temporary clipping, little experimental data are available regarding the ischemic consequences of temporary arterial occlusion with periods of reperfusion versus uninterrupted temporary occlusion. To investigate this issue, a protocol of occlusion/reperfusion that simulates the temporal profile that occurs during surgery was performed in a rat model of focal ischemia. Sixteen anesthetized Sprague-Dawley rats were divided into two groups. The animals in Group I underwent 60 minutes of uninterrupted middle cerebral artery occlusion and the animals in Group II were subjected to six separate 10-minute occlusion periods with 5 minutes of reperfusion between occlusions. Histopathological analysis was performed 72 hours postischemia. Group I had significantly increased mean infarction volumes (50.0 +/- 12.1 mm3) compared to Group II (8.7 +/- 3.1 mm3) (p = 0.008). Injuries in Group I occurred in both the cortex and striatum, whereas Group II showed only striatal injuries. Furthermore, the extent of the injuries in Group II was less severe, characterized by ischemic neuronal injury rather than frank infarction. The results indicate that intermittent reperfusion is neuroprotective during temporary focal ischemia and support the hypothesis that intermittent reperfusion is beneficial if temporary clipping is required during aneurysm repair.     

Value of transcranial Doppler indices in predicting raised ICP in infantile hydrocephalus. A study with review of the literature

Cerebral hemodynamic changes in infants with progressive hydrocephalus have been studied with the transcranial Doppler (TCD) technique. Several authors have referred to the correlation between the hemodynamic changes and increased intracranial pressure (ICP). Despite conflicting conclusions on the value of pulsatility index (PI) and resistance index (RI) measurements for monitoring infantile hydrocephalus, these pulsatility indices are the most commonly used for this purpose. Although clinical signs of raised ICP are highly variable and unreliable in infants, assumptions have been made in most of the studies about the presence of elevated ICP on the basis of the patient's clinical state. Few studies have reported on actual ICP values, however, and a direct relationship between ICP and TCD changes has never been adequately demonstrated. In the present study, this relationship was investigated in long-term simultaneous TCD/ICP measurements, in an attempt to develop a noninvasive method of monitoring the effect of ICP on intracranial hemodynamics. Two groups of data sets were established. Group I consisted of pre- and postoperative (shunt implantation) TCD/ICP measurements. Group II were long-term simultaneous TCD/ICP recordings showing significant ICP variations. In most of the postoperative measurements there was a decrease in the average PI and RI values. The correlation between PI or RI and ICP in the long-term simultaneous recordings, however, was generally poor. The risk of obtaining false positive or false negative PI or RI values in short-term measurements was also demonstrated. It can be concluded from our results, besides the wide range of reference values for the Doppler indices and extracranial influences upon them, that the present Doppler indices are inadequate for monitoring the complex intracranial dynamic responses in patients with raised ICP.