Assessment of Severe Apnoea through Voice Analysis, Automatic Speech, and Speaker Recognition Techniques

This study is part of an ongoing collaborative effort between the medical and the signal processing communities to promote research on applying standard Automatic Speech Recognition (ASR) techniques for the automatic diagnosis of patients with severe obstructive sleep apnoea (OSA). Early detection of severe apnoea cases is important so that patients can receive early treatment. Effective ASR-based detection could dramatically cut medical testing time. Working with a carefully designed speech database of healthy and apnoea subjects, we describe an acoustic search for distinctive apnoea voice characteristics. We also study abnormal nasalization in OSA patients by modelling vowels in nasal and nonnasal phonetic contexts using Gaussian Mixture Model (GMM) pattern recognition on speech spectra. Finally, we present experimental findings regarding the discriminative power of GMMs applied to severe apnoea detection. We have achieved an 81% correct classification rate, which is very promising and underpins the interest in this line of inquiry.

Antigenic proteins involved in occupational rhinitis and asthma caused by obeche wood (Triplochiton scleroxylon).

BACKGROUND Obeche wood dust is a known cause of occupational asthma where an IgE-mediated mechanism has been demonstrated. OBJECTIVE To characterize the allergenic profile of obeche wood dust and evaluate the reactivity of the proteins by in vitro, ex vivo and in vivo assays in carpenters with confirmed rhinitis and/or asthma MATERIALS AND METHODS An in-house obeche extract was obtained, and two IgE binding bands were purified (24 and 12 kDa) and sequenced by N-terminal identity. Specific IgE and IgG, basophil activation tests and skin prick tests (SPTs) were performed with whole extract and purified proteins. CCD binding was analyzed by ELISA inhibition studies. RESULTS Sixty-two subjects participated: 12 with confirmed occupational asthma/rhinitis (ORA+), 40 asymptomatic exposed (ORA-), and 10 controls. Of the confirmed subjects, 83% had a positive SPT to obeche. There was a 100% recognition by ELISA in symptomatic subjects vs. 30% and 10% in asymptomatic exposed subjects and controls respectively (p<0.05). Two new proteins were purified, a 24 kDa protein identified as a putative thaumatin-like protein and a 12 kDa gamma-expansin. Both showed allergenic activity in vitro, with the putative thaumatin being the most active, with 92% recognition by ELISA and 100% by basophil activation test in ORA+ subjects. Cross-reactivity due to CCD was ruled out in 82% of cases. CONCLUSIONS Two proteins of obeche wood were identified and were recognized by a high percentage of symptomatic subjects and by a small proportion of asymptomatic exposed subjects. Further studies are required to evaluate cross reactivity with other plant allergens.