Multi-channel lung sound classification with convolutional recurrent neural networks.

In this paper, we present an approach for multi-channel lung sound classification, exploiting spectral, temporal and spatial information. In particular, we propose a frame-wise classification framework to process full breathing cycles of multi-channel lung sound recordings with a convolutional recurrent neural network. With our recently developed 16-channel lung sound recording device, we collect lung sound recordings from lung-healthy subjects and patients with idiopathic pulmonary fibrosis (IPF), within a clinical trial. From the lung sound recordings, we extract spectrogram features and compare different deep neural network architectures for binary classification, i.e. healthy vs. pathological. Our proposed classification framework with the convolutional recurrent neural network outperforms the other networks by achieving an F-score of F1≈92%. Together with our multi-channel lung sound recording device, we present a holistic approach to multi-channel lung sound analysis.

Interposition of Retrosternal Pedicled Jejunum After Hypopharyngolaryngo-Esophagogastrectomy.

Reconstruction of the upper gastrointestinal tract presents a surgical challenge after esophagogastrectomy, especially when it includes hypopharyngolaryngectomy. Reconstruction is generally undertaken with interposed colon as a substitute conduit, but it carries several risks. Alternative reconstruction of the foregut with pedicled retrosternal jejunum anastomosed at the level of the base of the tongue is described.

Crackle and Breathing Phase Detection in Lung Sounds with Deep Bidirectional Gated Recurrent Neural Networks.

In this paper, we present a method for event detection in single-channel lung sound recordings. This includes the detection of crackles and breathing phase events (inspiration/expiration). Therefore, we propose an event detection approach with spectral features and bidirectional gated recurrent neural networks (BiGRNNs). In our experiments, we use multichannel lung sound recordings from lung-healthy subjects and patients diagnosed with idiopathic pulmonary fibrosis, collected within a clinical trial. We achieve an event-based F-score of F1 ≈ 86% for breathing phase events and F1 ≈ 72% for crackles. The proposed method shows robustness regarding the contamination of the lung sound recordings with noise, bowel and heart sounds.

Analysis of horse-related injuries in children.

PURPOSE: The purpose of the present study was to investigate factors affecting the nature, characteristics, severity and outcome of horseback and horse care injuries in paediatric patients and to create guidelines for injury prevention. METHODS: Detailed clinical records of 265 children sustained horse-riding related injuries have been analysed. Questionnaires were mailed to provide follow-up information for patients who have been treated in either Department of Paediatric Surgery in Pécs, Hungary, or Department of Paediatric Surgery in Graz, Austria between 1999 and 2003. Those 112 children (42%) who answered the questionnaire were included in the study and further analyses were performed. RESULTS: Female to male ratio of the 112 patients was 101/11. Trauma occurred during horseback riding accounted for 76.8% of all cases; these injuries represented more severe cases comparing to those which happened while handling a horse (23.2%). The mechanism as well as the localisation of injury displayed a close association with age. CONCLUSIONS: Prevention strategies targeting horse-related injuries at children should appreciate the age-dependent nature of injury as well as the fact that injury severity is not necessarily associated with the experience of the rider.