A family of experiments about how developers perceive delayed system response time.

Collecting and analyzing data about developers working on their development tasks can help improve development practices, finally increasing the productivity of teams. Indeed, monitoring and analysis tools have already been used to collect data from productivity tools. Monitoring inevitably consumes resources and, depending on their extensiveness, may significantly slow down software systems, interfering with developers' activity. There is thus a challenging trade-off between monitoring and validating applications in their operational environment and preventing the degradation of the user experience. The lack of studies about when developers perceive an overhead introduced in an application makes it extremely difficult to fine-tune techniques working in the field. In this paper, we address this challenge by presenting an empirical study that quantifies how developers perceive overhead. The study consists of three replications of an experiment that involved 99 computer science students in total, followed by a small-scale experimental assessment of the key findings with 12 professional developers. Results show that non-negligible overhead can be introduced for a short period into applications without developers perceiving it and that the sequence in which complex operations are executed influences the perception of the system response time. This information can be exploited to design better monitoring techniques.

Deep Representation Learning for Open Vocabulary Electroencephalography-to-Text Decoding.

Previous research has demonstrated the potential of using pre-trained language models for decoding open vocabulary Electroencephalography (EEG) signals captured through a non-invasive Brain-Computer Interface (BCI). However, the impact of embedding EEG signals in the context of language models and the effect of subjectivity, remain unexplored, leading to uncertainty about the best approach to enhance decoding performance. Additionally, current evaluation metrics used to assess decoding effectiveness are predominantly syntactic and do not provide insights into the comprehensibility of the decoded output for human understanding. We present an end-to-end architecture for non-invasive brain recordings that brings modern representational learning approaches to neuroscience. Our proposal introduces the following innovations: 1) an end-to-end deep learning architecture for open vocabulary EEG decoding, incorporating a subject-dependent representation learning module for raw EEG encoding, a BART language model, and a GPT-4 sentence refinement module; 2) a more comprehensive sentence-level evaluation metric based on the BERTScore; 3) an ablation study that analyses the contributions of each module within our proposal, providing valuable insights for future research. We evaluate our approach on two publicly available datasets, ZuCo v1.0 and v2.0, comprising EEG recordings of 30 subjects engaged in natural reading tasks. Our model achieves a BLEU-1 score of 42.75%, a ROUGE-1-F of 33.28%, and a BERTScore-F of 53.86%, achieving an increment over the previous state-of-the-art by 1.40%, 2.59%, and 3.20%, respectively.

Embedding the Patient-Citizen Perspective into an Operational Framework for the Development and the Introduction of New Technologies in Rehabilitation Care: The Smart&Touch-ID Model.

To date, at least 2.41 billion people with Non-Communicable Diseases (NCDs) are in need of rehabilitation. Rehabilitation care through innovative technologies is the ideal candidate to reach all people with NCDs in need. To obtain these innovative solutions available in the public health system calls for a rigorous multidimensional evaluation that, with an articulated approach, is carried out through the Health Technology Assessment (HTA) methodology. In this context, the aim of the present paper is to illustrate how the Smart&TouchID (STID) model addresses the need to incorporate patients' evaluations into a multidimensional technology assessment framework by presenting a feasibility study of model application with regard to the rehabilitation experiences of people living with NCDs. After sketching out the STID model's vision and operational process, preliminary evidence on the experiences and attitudes of patients and citizens on rehabilitation care will be described and discussed, showing how they operate, enabling the co-design of technological solutions with a multi-stakeholder approach. Implications for public health are discussed including the view on the STID model as a tool to be integrated into public health governance strategies aimed at tuning the agenda-setting of innovation in rehabilitation care through a participatory methodology.

[Internet- and Mobile-based interventions aimed at suicide risks: a review of the literature.] / Interventi Internet- e Mobile-based rivolti al rischio suicidario: una revisione della letteratura.

Many suicidal people do not receive professional assistance due to economic problems, the desire to autonomously solve the problem, or the stigma. Internet and Mobile-based Interventions (IMIs) can overcome many of these obstacles, offering flexible and accessible interventions. We conducted a narrative literature review about the characteristics of IMIs focused on suicide risks. We studied their efficacy and some examples of applications and websites used to prevent and treat suicide risk. METHOD: Literature research was conducted on Pub Med and PsycINFO databases. RESULTS: IMIs are a viable alternative to traditional treatments and several studies supported their effectiveness in reducing suicidal ideation and behaviors. Most IMIs are based on cognitive-behavioral therapy, which is very structured and easily transferable to digital format. Empowerment, reinforcement mechanisms and human support are the main features that promote the change of the individual and her or his adherence to the treatment. Some IMIs developed for suicide prevention are Stay Alive (UK), Virtual Hope Box (USA), This way up (Australia), and Jaspr Health (USA). CONCLUSIONS: IMIs can be considered a valid aid in suicide prevention. It is necessary to continue the research supporting their effectiveness and implement their interventions in Italy.

An Internet-Based Multi-Approach Intervention Targeting University Students Suffering from Psychological Problems: Design, Implementation, and Evaluation.

Despite the widespread prevalence of mental health problems, most psychological distress remains untreated. Internet-based psychological interventions can be an essential tool for increasing treatment availability and accessibility. The main objective of the MindBlooming project is to design and implement an innovative Internet-based multi-approach treatment for university students suffering from psychological or physical problems. The intervention will focus on symptoms of depression, anxiety, sleep problems, self-destructive thoughts, job- and study-related stress and burnout, and chronic pain. It will be based on different approaches, primarily psychoeducation, Cognitive-Behavioral Treatment (CBT), and third-wave CBT. At the end of the treatment, user satisfaction and usability will be assessed. In addition, two further aims will be evaluating the treatment efficacy through a randomized controlled trial and tuning a predictive model through Machine Learning techniques. The intervention consists of a 7-week treatment on two problematic areas according to each students' personal needs, identified through an initial assessment. Besides the treatment assigned following the initial screening, participants will also be assigned to a different module to improve their relational skills. The treatment, which can be accessed through a mobile app, consists of psychoeducational videos followed by related exercises. We expect MindBlooming to be a remarkable tool for promoting the mental health of university students.

The House of Carbs: Personalized Carbohydrate Dispenser for People with Diabetes.

Patients with diabetes are often worried about having low blood glucose because of the unpleasant feeling and possible dangerous situations this can lead to. This can make patients consume more carbohydrates than necessary. Ad-hoc carbohydrate estimation and dosing by the patients can be unreliable and may produce unwanted periods of high blood glucose. In this paper we present a system that automatically estimates and dispenses the amount of juice (or similar) according to the current patients' blood glucose values. The system is remotely accessible and customizable from a chatbot, exploits sensors and actuators to dispense the necessary amount of liquid carbohydrates. It relies on a cloud solution (Nightscout) to acquire the patient's blood glucose values, which are constantly updated thanks to a commercial wearable continuous glucose monitor (CGM).

From source code to test cases: A comprehensive benchmark for resource leak detection in Android apps.

Android apps share resources, such as sensors, cameras, and Global Positioning System, that are subject to specific usage policies whose correct implementation is left to programmers. Failing to satisfy these policies may cause resource leaks, that is, apps may acquire but never release resources. This might have different kinds of consequences, such as apps that are unable to use resources or resources that are unnecessarily active wasting battery. Researchers have proposed several techniques to detect and fix resource leaks. However, the unavailability of public benchmarks of faulty apps makes comparison between techniques difficult, if not impossible, and forces researchers to build their own data set to verify the effectiveness of their techniques (thus, making their work burdensome). The aim of our work is to define a public benchmark of Android apps affected by resource leaks. The resulting benchmark, called AppLeak, is publicly available on GitLab and includes faulty apps, versions with bug fixes (when available), test cases to automatically reproduce the leaks, and additional information that may help researchers in their tasks. Overall, the benchmark includes a body of 40 faults that can be exploited to evaluate and compare both static and dynamic analysis techniques for resource leak detection.