Long term peripheral nerve regeneration using a novel PCL nerve conduit.

The gold standard in surgical management of a peripheral nerve gap is currently autologous nerve grafting. This confers patient morbidity and increases surgical time therefore innovative experimental strategies towards engineering a synthetic nerve conduit are welcome. We have developed a novel synthetic conduit made of poly ε-caprolactone (PCL) that has demonstrated promising peripheral nerve regeneration in short-term studies. This material has been engineered to permit translation into clinical practice and here we demonstrate that histological outcomes in a long-term in vivo experiment are comparable with that of autologous nerve grafting. A 1cm nerve gap in a rat sciatic nerve injury model was repaired with a PCL nerve conduit or an autologous nerve graft. At 18 weeks post surgical repair, there was a similar volume of regenerating axons within the nerve autograft and PCL conduit repair groups, and similar numbers of myelinated axons in the distal stump of both groups. Furthermore, there was evidence of comparable re-innervation of end organ muscle and skin with the only significant difference the lower wet weight of the muscle from the PCL conduit nerve repair group. This study stimulates further work on the potential use of this synthetic biodegradable PCL nerve conduit in a clinical setting.

Short-term cytotoxic and inflammatory responses of human monocytes to stainless steel fibre networks

The aim of the current work was to examine the human monocyte response to 444 ferritic stainless steel fibre networks. 316L austenitic fibre networks, of the same fibre volume fraction, were used as control surfaces. Fluorescence and scanning electron microscopies suggest that the cells exhibited a good degree of attachment and penetration throughout both networks. Lactate Dehydrogenase (LDH) and TNF-α releases were used as indicators of cytotoxicity and inflammatory responses respectively. LDH release indicated similar levels of monocyte viability when in contact with the 444 and 316L fibre networks. Both networks elicited a low level secretion of TNF-α, which was significantly lower than that of the positive control wells containing zymosan. Collectively, the results suggest that 444 ferritic and 316L austenitic networks induced similar cytotoxic and inflammatory responses from human monocytes. © 2012 Materials Research Society.

System combination and score normalization for spoken term detection

Spoken content in languages of emerging importance needs to be searchable to provide access to the underlying information. In this paper, we investigate the problem of extending data fusion methodologies from Information Retrieval for Spoken Term Detection on low-resource languages in the framework of the IARPA Babel program. We describe a number of alternative methods improving keyword search performance. We apply these methods to Cantonese, a language that presents some new issues in terms of reduced resources and shorter query lengths. First, we show score normalization methodology that improves in average by 20% keyword search performance. Second, we show that properly combining the outputs of diverse ASR systems performs 14% better than the best normalized ASR system. © 2013 IEEE.

Investigation of multilingual deep neural networks for spoken term detection

The development of high-performance speech processing systems for low-resource languages is a challenging area. One approach to address the lack of resources is to make use of data from multiple languages. A popular direction in recent years is to use bottleneck features, or hybrid systems, trained on multilingual data for speech-to-text (STT) systems. This paper presents an investigation into the application of these multilingual approaches to spoken term detection. Experiments were run using the IARPA Babel limited language pack corpora (∼10 hours/language) with 4 languages for initial multilingual system development and an additional held-out target language. STT gains achieved through using multilingual bottleneck features in a Tandem configuration are shown to also apply to keyword search (KWS). Further improvements in both STT and KWS were observed by incorporating language questions into the Tandem GMM-HMM decision trees for the training set languages. Adapted hybrid systems performed slightly worse on average than the adapted Tandem systems. A language independent acoustic model test on the target language showed that retraining or adapting of the acoustic models to the target language is currently minimally needed to achieve reasonable performance. © 2013 IEEE.