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.     

The ileoanal J pouch: radiographic evaluation

Endorectal ileoanal pull-through offers an attractive alternative to proctocolectomy and ileostomy for patients with ulcerative colitis, Gardner syndrome, and familial polyposis. To our knowledge, a careful radiographic analysis of the ileum, ileal pouch, and ileoanal anastomosis after ileoanal pull-through has not been reported. Thirty-two patients with ulcerative colitis, Gardner syndrome, and familial polyposis underwent colectomy, mucosal proctectomy, and endorectal ileoanal pull-through of a 15-cm ileal "J" pouch and loop ileostomy. Twenty-five (78%) of 32 of all the pouches radiographically demonstrated spiral folds extending from the middle of the pouch to the pectinate line. Other radiographic features included a mesenteric mass effect, pseudopolyps, and a central lucency that indicated intrapouch sutures. Radiographs provide useful information in the postoperative management of the ileal pull-through.     

The influence of reported paternal attitudes on the decision to breast-feed

Objective: To identify factors that influence a woman's decision to breast-feed.
Methodology: Five hundred and fifty-six women were recruited from the maternity wards of two Perth hospitals. Data were collected from a self-administered questionnaire completed by participants prior to discharge. Logistic regression analysis was used to determine factors influencing the initiation of breast-feeding.
Results: At discharge from hospital 83.8% of women were breast-feeding, including 6% who were giving complementary formula feeds. After controlling for potentially confounding demographic and biomedical factors, the father's reported preference for breast-feeding was found to be the most important factor influencing a woman's decision to breast-feed (OR 10.18).
Conclusion: Fathers participate in and influence the choice of infant feeding method and should be included in breast-feeding discussions.     

Bacterial and mammalian DNA alkyltransferases sensitize Escherichia coli to the lethal and mutagenic effects of dibromoalkanes

Here we confirm and extend our previous studies demonstrating that the mutagenic potency of 1,2-dibromoethane (DBE) and dibromomethane (DBM) is markedly enhanced (not prevented) in bacteria expressing the O6- alkylguanine-DNA alkyltransferase (ATase) encoded by the Escherichia coli ogt gene. We demonstrate that, in close parallel with mutagenesis, the Ogt ATase sensitizes the bacteria to the lethal effects of these carcinogens, suggesting that one or more of the potentially mutagenic lesions induced by DBE and DBM in the presence of Ogt has additional lethal capacity. We further demonstrate that the sensitization to both lethality and mutagenesis by DBE and DBM is a property shared by other DNA alkyltransferases. This objective was accomplished by quantifying the induction of mutations and lethal events in ogt- ada- E. coli expressing an exogenous bacterial or mammalian ATase from a multicopy plasmid. Mammalian recombinant ATases enhanced the lethal and mutagenic actions of DBE and suppressed the lack of sensitivity of the vector- transformed bacteria to DBM. In most cases the order of effectiveness of the ATases ranked: murine > human > Ogt > rat. Further comparisons included the full-length Ada ATase from E. coli and a truncated Ada version (T-ada) that retains the O6-methylguanine binding domain of the protein. The full-length Ada ATase was effective in enhancing the lethality but not the mutagenicity induced by DBE and DBM. The T-ada ATase provided less sensitization than Ada to lethality by DBE, but of the three bacterial ATases T-ada yielded the highest sensitization to mutagenesis by this compound. T-ada and Ada ATases were in general less effective than the mammalian versions, with the exception of the rat recombinant ATase. The effectiveness of the different mammalian and bacterial ATases in promoting the deleterious actions of dibromoalkanes was compared with the effectiveness of these proteins in suppressing the lethal and mutagenic effects induced by N-nitroso-N-methylurea. The ability to sensitize E. coli to the lethal and mutagenic effects of DBE and DBM seems restricted to DNA alkyltransferase, since overexpression of thioredoxin (Trx) or glutaredoxin (Grx1) in ogt- ada- cells showed no effect, in spite of the reported potential of bioactive dihaloethane- derived species to alkylate Trx.      

Assessment of the mutagenicity of dichloroacetic acid in lacI transgenic B6C3F1 mouse liver

Dichloroacetic acid (DCA) is a chlorination byproduct found in finished drinking water. When administered in drinking water this chemical has been shown to produce hepatocellular adenomas and carcinomas in B6C3F1 mice over the animal's lifetime. In this study, we investigated whether mutant frequencies were increased in mouse liver using treatment protocols that yielded significant tumor induction. DCA was administered continuously at either 1.0 or 3.5 g/l in drinking water to male transgenic B6C3F1 mice harboring the bacterial lacI gene. Groups of five or six animals were killed at 4, 10 or 60 weeks and livers removed. At both 4 and 10 weeks of treatment, there was no significant difference in mutant frequency between the treated and control animals at either dose level. At 60 weeks, mice treated with 1.0 g/l DCA showed a 1.3-fold increase in mutant frequency over concurrent controls (P = 0.05). Mice treated with 3.5 g/l DCA for 60 weeks had a 2.3-fold increase in mutant frequency over the concurrent controls (P = 0.002). The mutation spectrum recovered from mice treated with 3.5 g/l DCA for 60 weeks contained G:C-->A:T transitions (32.79%) and G:C-->T:A transversions (21.31%). In contrast, G:C-->A:T transitions comprised 53.19% of the recovered mutants among control animals. Although only 19.15% of mutations among the controls were at T:A sites, 32.79% of the mutations from DCA-treated animals were at T:A sites. This is consistent with the previous observation that the proportion of mutations at T:A sites in codon 61 of the H-ras gene was increased in DCA-induced liver tumors in B6C3F1 mice. The present study demonstrates DCA-associated mutagenicity in the mouse liver under conditions in which DCA produces hepatic tumors.