Multimodal Emotion Recognition on RAVDESS Dataset Using Transfer Learning.

Emotion Recognition is attracting the attention of the research community due to the multiple areas where it can be applied, such as in healthcare or in road safety systems. In this paper, we propose a multimodal emotion recognition system that relies on speech and facial information. For the speech-based modality, we evaluated several transfer-learning techniques, more specifically, embedding extraction and Fine-Tuning. The best accuracy results were achieved when we fine-tuned the CNN-14 of the PANNs framework, confirming that the training was more robust when it did not start from scratch and the tasks were similar. Regarding the facial emotion recognizers, we propose a framework that consists of a pre-trained Spatial Transformer Network on saliency maps and facial images followed by a bi-LSTM with an attention mechanism. The error analysis reported that the frame-based systems could present some problems when they were used directly to solve a video-based task despite the domain adaptation, which opens a new line of research to discover new ways to correct this mismatch and take advantage of the embedded knowledge of these pre-trained models. Finally, from the combination of these two modalities with a late fusion strategy, we achieved 80.08% accuracy on the RAVDESS dataset on a subject-wise 5-CV evaluation, classifying eight emotions. The results revealed that these modalities carry relevant information to detect users' emotional state and their combination enables improvement of system performance.

Tannin-based supplementation influences gut microbiota composition and activity in IBS-D patients with a potential impact on symptoms: a pilot study.

As the involvement of the intestinal microbiota in the etiopathology of irritable bowel syndrome, subtype diarrhoea (IBS-D) is now increasingly recognised, a preliminary, quasi-experimental, before-after and prospective study was conducted on 28 patients to test the effect of a tannin-based supplement on the composition and activity of the microbiota, after 8 weeks of treatment. No statistically significant differences were found in α- or ß-diversity. However, sparse Partial Least Squares Discriminant Analysis (sPLS-DA) and Boruta algorithm did reveal significant changes in the relative abundance of specific groups of bacteria, highlighting the involvement of recognized of IBS-D biomarkes, namely Blautia (adj p = 3.5 × 10-11), Eubacterium hallii group (adj p = 5.1 × 10-12) and Dorea (adj p = 1.8 × 10-18), which resulted significantly depleted by the treatment. The modulation of the composition of the gut microbiota had an impact also in the production of short chain fatty acids (SCFAs), which were modulated: acetate and butyrate (n.s. and p = 0.000143) increased while propionate and formate resulted to be significantly reduced (p = 0.00476 and p = 0.00011, respectively), following the supplementation. Finally, the sPLS analysis showed that the strongest association between faecal microbiome composition and clinical symptoms of IBS-D was given by Catenibacterium, which showed a positive correlation with evacuation-related symptoms. Such preliminary findings suggest that tannin supplementation could play an outstanding role in microbiota modulation in IBS-D patients, potentially improving their symptomatology, by selectively acting on the growth and the activity of specific groups of taxa.