Background Suicide risk assessment usually involves an interaction between doctors and patients. However, a significant number of people with mental disorders receive no treatment for their condition due to the limited access to mental health care facilities; the reduced availability of clinicians; the lack of awareness; and stigma, neglect, and discrimination surrounding mental disorders. In contrast, internet access and social media usage have increased significantly, providing experts and patients with a means of communication that may contribute to the development of methods to detect mental health issues among social media users. Objective This paper aimed to describe an approach for the suicide risk assessment of Spanish-speaking users on social media. We aimed to explore behavioral, relational, and multimodal data extracted from multiple social platforms and develop machine learning models to detect users at risk. Methods We characterized users based on their writings, posting patterns, relations with other users, and images posted. We also evaluated statistical and deep learning approaches to handle multimodal data for the detection of users with signs of suicidal ideation (suicidal ideation risk group). Our methods were evaluated over a dataset of 252 users annotated by clinicians. To evaluate the performance of our models, we distinguished 2 control groups: users who make use of suicide-related vocabulary (focused control group) and generic random users (generic control group). Results We identified significant statistical differences between the textual and behavioral attributes of each of the control groups compared with the suicidal ideation risk group. At a 95% CI, when comparing the suicidal ideation risk group and the focused control group, the number of friends (P=.04) and median tweet length (P=.04) were significantly different. The median number of friends for a focused control user (median 578.5) was higher than that for a user at risk (median 372.0). Similarly, the median tweet length was higher for focused control users, with 16 words against 13 words of suicidal ideation risk users. Our findings also show that the combination of textual, visual, relational, and behavioral data outperforms the accuracy of using each modality separately. We defined text-based baseline models based on bag of words and word embeddings, which were outperformed by our models, obtaining an increase in accuracy of up to 8% when distinguishing users at risk from both types of control users. Conclusions The types of attributes analyzed are significant for detecting users at risk, and their combination outperforms the results provided by generic, exclusively text-based baseline models. After evaluating the contribution of image-based predictive models, we believe that our results can be improved by enhancing the models based on textual and relational features. These methods can be extended and applied to different use cases related to other mental disorders.

The statistics presented by the World Health Organization inform that 90% of the suicides can be attributed to mental illnesses in high-income countries. Besides, previous studies concluded that people with mental illnesses tend to reveal their mental condition on social media, as a way of relief. Thus, the main objective of this work is the analysis of the messages that a user posts online, sequentially through a time period, and detect as soon as possible if this user is at risk of depression. This paper is a preliminary attempt to minimize measures that penalize the delay in detecting positive cases. Our experiments underline the importance of an exhaustive sentiment analysis and a combination of learning algorithms to detect early symptoms of depression.

Indigenous beers (chicha) are part of the indigenous culture in Ecuador. The fermentation process of these beers probably relies on microorganisms from fermented substrates, environment and human microbiota. We analyzed the microbiota of artisanal beers (including a type of beer produced after chewing boiled cassava) using bacterial culture and 16S ribosomal RNA (rRNA) gene-based tag-encoded FLX amplicon pyrosequencing (bTEFAP). Surprisingly, we found that Streptococcus salivarius and Streptococcus mutans (part of the human oral microbiota) were among the most abundant bacteria in chewed cassava and in non-chewed cassava beers. We also demonstrated that S. salivarius and S. mutans (isolated from these beers) could proliferate in cassava mush. Lactobacillus sp. was predominantly present in most types of Ecuadorian chicha.

The impact of cancer in the society has created the necessity of new and faster theoretical models that may allow earlier cancer detection. The present review gives the prediction of cancer by using the star graphs of the protein sequences and proteome mass spectra by building a Quantitative Protein -Disease Relationships (QPDRs), similar to Quantitative Structure Activity Relationship (QSAR) models. The nodes of these star graphs are represented by the amino acids of each protein or by the amplitudes of the mass spectra signals and the edged are the geometric and/or functional relationships between the nodes. The star graphs can be numerically described by the invariant values named topological indices (TIs). The transformation of the star graphs (graphical representation) of proteins into TIs (numbers) facilitates the manipulation of protein information and the search for structure-function relationships in Proteomics. The advantages of this method include simplicity, fast calculations and free resources such as S2SNet and MARCH-INSIDE tools. Thus, this ideal theoretical scheme can be easily extended to other types of diseases or even other fields, such as Genomics or Systems Biology.

Anorexia Nervosa (AN) is a serious mental disorder that has been proved to be traceable on social media through the analysis of users' written posts. Here we present an approach to generate word embeddings enhanced for a classification task dedicated to the detection of Reddit users with AN. Our method extends Word2vec's objective function in order to put closer domain-specific and semantically related words. The approach is evaluated through the calculation of an average similarity measure, and via the usage of the embeddings generated as features for the AN screening task. The results show that our method outperforms the usage of fine-tuned pre-learned word embeddings, related methods dedicated to generate domain adapted embeddings, as well as representations learned on the training set using Word2vec. This method can potentially be applied and evaluated on similar tasks that can be formalized as document categorization problems. Regarding our use case, we believe that this approach can contribute to the development of proper automated detection tools to alert and assist clinicians.