Screening for albuminuria in diabetes:the need to do more

Aims: To reassess the utilisation rate of urinary albumin to creatinine ratio (ACR) screening in our centre; and the rate of repeat testing, where appropriate. To look at risk factors for albuminuria in our outpatient population. Methods: All patients attending one of our two weekly diabetes outpatient clinics in 2011–2012 were enrolled in this study. Demographic and relevant clinical data were extracted from electronic care records and analysed using SPSS 21. Results: Our study cohort comprised 998 people (51.4% men;59.6% White, 30.5% Southeast Asian, 9.9% Afro-Caribbean),most of whom had Type 2 diabetes (82.6%). The ACR testing rate in our centre was 62.8% (2012–2013 data; previously 62.4%). The incidence of initial albuminuria was 32.2% in women vs42.8% in men. Just 48.7% of patients (44.4% of women, 51.8% of men) with initial albuminuria were retested: 36.4% of women and 19.7% of men with initial albuminuria had no evidence of this on follow-up. Logistic regression modelling confirmed an association of high systolic blood pressure with albuminuria [odds ratio1.92 (1.01–3.70 in women, 1.08–3.57 in men)]. Treatment with anangiotens in converting enzyme inhibitor (ACEi) or angiotens in 2 receptor blocker (A2RB) was negatively associated with albuminuria in men [odds ratio 0.42 (0.20–0.89)], but not in women. Conclusions: A relatively high, albeit suboptimal, albuminuria screening rate in our outpatient population has been sustained.High systolic blood pressure was confirmed as a risk factor foralbuminuria. The incidence of albuminuria was higher in men, who had a lower rate of negative repeat testing and appeared to benefit more from ACEi/A2RB therapy. More rigorous screening for albuminuria is warranted to identify at-risk individuals.

Speaker identification using higher order spectral phase features and their effectiveness vis-a-vis Mel-Cepstral features

The effectiveness of higher-order spectral (HOS) phase features in speaker recognition is investigated by comparison with Mel Cepstral features on the same speech data. HOS phase features retain phase information from the Fourier spectrum unlikeMel–frequency Cepstral coefficients (MFCC). Gaussian mixture models are constructed from Mel– Cepstral features and HOS features, respectively, for the same data from various speakers in the Switchboard telephone Speech Corpus. Feature clusters, model parameters and classification performance are analyzed. HOS phase features on their own provide a correct identification rate of about 97% on the chosen subset of the corpus. This is the same level of accuracy as provided by MFCCs. Cluster plots and model parameters are compared to show that HOS phase features can provide complementary information to better discriminate between speakers.