Search

This paper presents a user interface for the exploration of music libraries based on parametric t-SNE. Each song in the music library is represented as a stack of 34-dimensional vectors containing features related to genres, emotions and other musical characteristics. Parametric t-SNE is used to construct a model that extracts a pair of coordinates from these features for each song, preserving similarity rela- tions between songs in the high dimensional-feature space and their projection in a two-dimensional space. The two-dimensional output of the model will be used for projecting and rendering a song catalogue onto a plane. We have investigated different models, which have been obtained by using different structures of hidden layers, pre-training technique, features selection, and data pre-processing. These results are an extension of a previous project published by Moodagent Company, which show that the clustering adaptation of genres and emotions, that is obtained by using parametric t-SNE, is by far more accurate than the previous methods based on PCA. Finally, our study proposes a visual representation of the resulting model. The model has been used to build a music-space of 20000 songs, visually rendered for browser interaction. This provides the user with a certain degree of freedom to explore the music-space by changing the highlighted features and it offers an immersive experience for music exploration and playlist generation.

Objective To estimate the proportion of patients with very severe osteoporosis (those covered by the reimbursement criteria of the Italian National Health Service) experiencing new vertebral and non-vertebral fragility fractures in the first 24 months of a new anti-osteoporosis treatment. Methods Prospective observational study in men and post-menopausal women (aged > 21 years) initiating anti-osteoporosis treatment for very severe osteoporosis. Eligibility was based on teriparatide (TPD) reimbursement criteria in Italy: Incident of vertebral or hip fracture during anti-resorptive treatment (minimum 1 year), or at least three prevalent severe vertebral fractures, or two prevalent severe vertebral fractures and a historical proximal hip fracture. Incidence of new clinical vertebral and non-vertebral fractures was documented by original x-rays and/or radiological reports, and a post-hoc analysis compared data from the TPD monotherapy population versus the total treated group. Results Overall, 767 patients (mean age 72.8 years, 90.7% women) were enrolled in the study, of whom 628, 538, 419 and 424 attended visits at 6, 12, 18 and 24 months, respectively. The most commonly prescribed therapy was TPD (single-agent; 64.5%), then bisphosphonates and other anti-resorptives (33.3%). A combination of different oral treatments was given to 22.5% of the patients. Overall treatment adherence at 24 months was 65.7%. In a post-hoc analysis, the overall incidence of new clinical vertebral and non-vertebral fractures in the total treated population was, respectively, 4.7% and 2.3% in the first 6 months; 1.8% and 1.6% in the 6-12 month period; 2.9% and 1.4% in the 12-18 month period; and 2.2% and 1.0% in the 18-24 month period. Conclusion In patients with very severe osteoporosis, the risk of new vertebral and non-vertebral fractures declined after the first 6 months and remained low throughout the study.