Initial evaluation of a continuous speech recognition program for radiology

The aims of this work were to measure the accuracy of one continuous speech recognition product and dependence on the speaker's gender and status as a native or nonnative English speaker, and evaluate the product's potential for routine use in transcribing radiology reports. IBM MedSpeak/Radiology software, version 1.1 was evaluated by 6 speakers. Two were nonnative English speakers, and 3 were men. Each speaker dictated a set of 12 reports. The reports included neurologic and body imaging examinations performed with 6 different modalities. The dictated and original report texts were compared, and error rates for overall, significant, and subtle significant errors were computed. Error rate dependence on modality, native English speaker status, and gender were evaluated by performing ttests. The overall error rate was 10.3 +/- 3.3%. No difference in accuracy between men and women was found; however, significant differences were seen for overall and significant errors when comparing native and nonnative English speakers (P = .009 and P = .008, respectively). The speech recognition software is approximately 90% accurate, and while practical implementation issues (rather than accuracy) currently limit routine use of this product throughout a radiology practice, application in niche areas such as the emergency room currently is being pursued. This methodology provides a convenient way to compare the initial accuracy of different speech recognition products, and changes in accuracy over time, in a detailed and sensitive manner.     

Arginine vasopressin metabolism in dogs. I. Evidence for a receptor-mediated mechanism.

The plasma clearance rates (PCR) of arginine vasopressin (AVP), and iodinated AVP (125I-AVP) were determined after pulse injection in conscious water-loaded dogs. Both the PCR and the apparent initial volume of distribution were significantly greater for AVP than for the biologically inactive iodinated AVP 37.4 +/- 4.8 ml/kg per min vs. 6.7 +/- 0.8 ml/kg per min (P less than 0.001) and 12.7 +/- 0.9% body wt vs. 7.1 +/- 0.4% body wt (P less than 0.001). AVP clearance was then determined by the constant-infusion technique at doses that produced equilibrium AVP concentrations within and above the physiological range. AVP-PCR was 37.4 +/- 7.1 ml/kg per min at 34 microU/kg per min, which was comparable to that after pulse injection (P less than 0.9). AVP clearance fell progressively, and urine osmolality progressively increased with increasing AVP infusion rates to plateau values at 136 microU/kg per min; a strong negative correlation was observed between mean AVP-PCR and urine osmolality (r = -0.993). The data suggest a relationship between the biological activity of AVP and its clearance. It is proposed that plasma membrane receptors may mediate a portion of the metabolic clearance of AVP.     

Disability-Adjusted Life Years Averted Versus Quality-Adjusted Life Years Gained: A Model Analysis for Breast Cancer Screening

ObjectivesTo quantify the impact of mammography-based screening on the quality of life, disability-adjusted life years (DALYs) averted or quality-adjusted life years (QALYs) gained can be used. We aimed to assess whether the use of DALYs averted or QALYs gained will lead to different cost-effective screening strategies.MethodsUsing the microsimulation model MISCAN, we simulated different breast cancer screening strategies varying in starting age (starting at 45, 47, and 50 years), stopping age (stopping at 69, 72, and 74 years), and frequency (annual [A], biennial [B], combination of both [A + B], and triennial [T]). In total, we defined 24 different breast cancer screening strategies, including no screening as a reference strategy. We calculated incremental cost-effectiveness ratios (ICERs) and compared which strategies were on the efficiency frontiers for DALYs and QALYs.ResultsBreast cancer screening averted between 46.00 and 105.58 DALYs and gained between 28.69 and 64.50 QALYs per 1000 women. For DALYs there were 5 strategies on the efficiency frontier (T50-69, T50-74, T45-74, B45-74, and A45-74). The same strategies plus one (B45-72) were on the efficiency frontier for QALYs.ConclusionsUsing DALYs averted instead of QALYs gained to assess the effects on quality of life from breast cancer screening in the Dutch population yields differences in ICERs, but almost the same strategies were on the efficiency frontiers. Whether the choice in outcome measure leads to a difference in optimal policy depends on the cost-effectiveness threshold.     

Factors influencing birth weight in newborns of diabetic and non-diabetic women a population based study

Maternal diabetes is known to be related to an increase in birth weight of the offspring. However, the mechanism of the association is not entirely clear. In addition, the contribution of the demographic, obstetric and metabolic factors to birth weight in diabetic mothers is not well defined. All the diabetic women (68 requiring insulin-treatment and 403 on diet alone) and a random sample of 1 in 12 of all non-diabetic women (893 women) who delivered in one regional hospital between March 1987 and June 1988 inclusive, were included in the study. Tests for gestational diabetes are routinely performed in our pregnant women population, thus, the study is a population based one. The mean birth weight of infants of diabetic mothers adjusted for gestational age was higher than in those of non-diabetic mothers. However, no relationship was found between maternal glycosylated hemoglobin measured at delivery and the infants birth weight. Furthermore, at each week of gestation, infants born to diabetic mothers were heavier than the infants of non-diabetic mothers (for weeks 37 to 40, p < 0.05), while no differences were found in glycosylated hemoglobin levels between the two groups at any time. In a multivariate model we showed that after controlling for gestational age, the only factors which independently and significantly affected birth weight in our population were diabetes, ethnic origin, and the parity of the mother. Our findings support the possibility that substances which induce hyperinsulinemia, other than glucose, may be related to the higher birth weight of infants of diabetic mothers.     

POSSIBLE INVOLVEMENT OF INOSITOL-LIPID METABOLISM IN MALIGNANT HYPERTHERMIA

-Adrenoceptor stimulation may induce malignant hyperthermia(MH) in vivo. Consequently, we have investigated the effectsof the -adrenoceptor agonist phenylephrine and, for comparison,the effects of the ß-adrenoceptor agonist isoproterenolon inositol-lipid metabolism of malignant hyperthermia susceptible(MHS) and healthy control (MHN) swine. The experiments wereperformed on electrically stimulated (frequency 0.2 Hz) trabeculaeisolated from the right ventricles of the hearts of MHS andMHN animals. After labelling with [³H] inositol for 6 h, differentinositol phosphates were measured by high pressure liquid chromatography,including inositol 1 - phosphate, inositol 1,4-bisphosphate,inositol 1,3,4-trisphosphate, inositol 1,4,5-trisphosphate (1,4,5-IP₃)and inositol 1,3,4,5 - tetrakisphosphate. After stimulationwith isoproterenol, the inositol phosphate content did not increaseor vary between muscle from MHS and MHN animals. In contrast,all inositol phosphates increased after stimulation with phenylephrinein both muscle types, the effects being greater in MHS thanin MHN, especially as regards 1,4,5-IP₃ content. As 1,4,5-IP₃,a presumed second messenger, has been shown to mobilize intracellularcalcium, it is concluded that an enhanced -adrenergic responseis involved in the development of MH. ^*Address for correspondence: Abteilung für Anästhesiologie,Universitäts-Krankenhaus Eppendorf, Martinistrasse52, D-2000Hamburg 20, Germany. Presented in part at the 1989 Meeting of the European Academyof Anaesthesiology in Bonn.     

Randomised controlled trial of zinc supplementation in malnourished Bangladeshi children with acute diarrhoea

OBJECTIVE: To evaluate the impact of zinc supplementation on the clinical course, stool weight, duration of diarrhoea, changes in serum zinc, and body weight gain of children with acute diarrhoea. DESIGN: Randomised double blind controlled trial. Children were assigned to receive zinc (20 mg elemental zinc per day) containing multivitamins or control group (zinc-free multivitamins) daily in three divided doses for two weeks. SETTING: A diarrhoeal disease hospital in Dhaka, Bangladesh. PATIENTS: 111 children, 3 to 24 months old, below 76% median weight for age of the National Center for Health Statistics standard with acute diarrhoea. Children with severe infection and/or oedema were excluded. MAIN OUTCOME MEASURES: Total diarrhoeal stool output, duration of diarrhoea, rate of weight gain, and changes in serum zinc levels after supplementation. RESULTS: Stool output was 28% less and duration 14% shorter in the zinc supplemented group than placebo (p = 0.06). There were reductions in median total diarrhoeal stool output among zinc supplemented subjects who were shorter (less than 95% height for age), 239 v 326 g/kg (p < 0.04), and who had a lower initial serum zinc (< 14 mmol/l), 279 v 329 g/kg (p < 0.05); a shortening of mean time to recovery occurred (4.7 v 6.2 days, p < 0.04) in those with lower serum zinc. There was an increase in mean serum zinc in the zinc supplemented group (+2.4 v -0.3 mumol/l, p < 0.001) during two weeks of supplementation, and better mean weight gain (120 v 30 g, p < 0.03) at the time of discharge from hospital. CONCLUSIONS: Zinc supplementation is a simple, acceptable, and affordable strategy which should be considered in the management of acute diarrhoea and in prevention of growth faltering in children specially those who are malnourished.