Search Results (31)

This study introduces and validates a novel instrument to measure aggressive gender backlash (AGB), a distinct and underexplored dimension of gender backlash (GB) within intimate partner relationships. Based on the General Aggression Model, a multidimensional scale was developed and tested using data from 513 Peruvian female microentrepreneurs. Results demonstrate solid evidence of reliability, discriminant validity, and predictive validity across five dimensions: hostility, the withdrawal of support, sabotage/coercion, gender stereotyping, and masculine victimization. The findings reveal that AGB is more prevalent than intimate partner violence against women (IPVAW) and often precedes it. AGB encompasses covert, non-violent behaviors that aim to resist female empowerment, such as emotional sabotage, manipulation, and disqualification, often normalized within relationships. This construct is significantly associated with lower levels of empowerment, increased subordination, emotional morbidity, and decreased work productivity. This study redefines GB as an interpersonal process measurable at the individual level and provides the first validated tool for its assessment. By conceptualizing AGB as a persistent, harmful, and functionally equivalent mechanism to IPVAW, though not necessarily physically violent, this research fills a key gap in gender violence literature. It offers practical implications for early detection and prevention strategies. Full article

Interpersonal violence involves intentional physical harm or force with psychological effects, influenced by interpersonal and societal factors. Health systems play a vital role in detecting and addressing such violence, requiring improved training and surveillance. Our hospital established a registry of suspected violence cases reported by healthcare professionals to enhance understanding, prevention strategies, and recognition of violence types and risk factors. Since 2009, all admitted patients suspected of experiencing violence were included, regardless of age or gender. Data from 2009 to 2022 covered demographics, violence details, medical interventions, and legal actions. Among 1284 patients, 83.4% were seen in the emergency department, with women comprising 80.8%, and with a mean age of 33.19 years. Reports of violence rose from 1.9% in 2009 to 16.9% in 2022. Risk factors included pregnancy [5.6%], age below 18 or over 80 [18.9%], disability [10.2%], and psychiatric conditions [11.3%]. Perpetrators were known in 56.8% of cases, mainly intimate partners [25.2%], with 29.4% of victims living with the aggressor. Doctors were primary reporters, and injury reports were issued in 65.5% of cases. Violence types included physical [44.5%], sexual [22.4%], psychological [13.3%], and economic [12.5%], with 36.3% of cases involving multiple types. Routine hospital screening and trained staff can improve victim support and enable injury prevention programs. Full article

This study conducts a systematic literature review following the PRISMA framework and the guidelines of Kitchenham and Charters to analyze the application of Internet of Things (IoT) technologies and deep learning models in monitoring violent actions and criminal activities in smart cities. A total of 45 studies published between 2010 and 2024 were selected, revealing that most research, primarily from India and China, focuses on cybersecurity in IoT networks (76%), while fewer studies address the surveillance of physical violence and crime-related events (17%). Advanced neural network models, such as Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, and hybrid approaches, have demonstrated high accuracy rates, averaging over 97.44%, in detecting suspicious behaviors. These models perform well in identifying anomalies in IoT security; however, they have primarily been tested in simulation environments (91% of analyzed studies), most of which incorporate real-world data. From a legal perspective, existing proposals mainly emphasize security and privacy. This study contributes to the development of smart cities by promoting IoT-based security methodologies that enhance surveillance and crime prevention in cities in developing countries. Full article

Emotion recognition through artificial intelligence and smart sensing of physical and physiological signals (affective computing) is achieving very interesting results in terms of accuracy, inference times, and user-independent models. In this sense, there are applications related to the safety and well-being of people (sexual assaults, gender-based violence, children and elderly abuse, mental health, etc.) that require even more improvements. Emotion detection should be done with fast, discrete, and non-luxurious systems working in real time and real life (wearable devices, wireless communications, battery-powered). Furthermore, emotional reactions to violence are not equal in all people. Then, large general models cannot be applied to a multi-user system for people protection, and health and social workers and law enforcement agents would welcome customized and lightweight AI models. These semi-personalized models will be applicable to clusters of subjects sharing similarities in their emotional reactions to external stimuli. This customization requires several steps: creating clusters of subjects with similar behaviors, creating AI models for every cluster, continually updating these models with new data, and enrolling new subjects in clusters when required. An initial approach for clustering labeled data compiled (physiological data, together with emotional labels) is presented in this work, as well as the method to ensure the enrollment of new users with unlabeled data once the AI models are generated. The idea is that this complete methodology can be exportable to any other expert systems where unlabeled data are added during in-field operation and different profiles exist in terms of data. Experimental results demonstrate an improvement of 5% in accuracy and 4% in F1 score with respect to our baseline general model, along with a 32% to 58% reduction in variability, respectively. Full article

Background: We investigated the predictors of poor SRH in a representative sample of Lithuanian mainstream school students in grades 7–8. We also checked for gender differences in the associations between SRH and depressive symptoms and other predictors. Methods: A total of 2104 7th–8th-grade students participated (response rate 73.95%) and were asked about depressive symptoms, psychosomatic health complaints, negative acts at school, feeling at school, family stress and violence, sense of coherence, self-esteem, and lifestyle. We used a hierarchical regression analysis including a variety of self-rated health predictors. Results: Boys scored significantly higher on physical activity and smoking, whereas girls scored significantly higher on SRH, depressive symptoms, psychosomatic health complaints, and family stress and violence, though the significance was lost in the hierarchical regression. Depressive symptoms were the strongest predictor of poor SRH (standardized β = 0.309, p < 0.001), though other investigated predictors were also significant but had lower effect sizes. Strong evidence was found supporting the buffering role of sense of coherence in the relationship between depressive symptoms and SRH (standardized β = −0.266, p < 0.001). Conclusions: We can conclude that the magnitude of the relationship between depressive symptoms and self-rated health is dependent on the levels of sense of coherence. We did not find gender differences in those associations. As poor SRH is easy to determine, especially with a one-item question, the cases of poorly rated health should be detected early and corrected by interventions in order to prevent poor health outcomes in the future. Full article

Public authorities and private companies have used video cameras as part of surveillance systems, and one of their objectives is the rapid detection of physically violent actions. This task is usually performed by human visual inspection, which is labor-intensive. For this reason, different deep learning models have been implemented to remove the human eye from this task, yielding positive results. One of the main problems in detecting physical violence in videos is the variety of scenarios that can exist, which leads to different models being trained on datasets, leading them to detect physical violence in only one or a few types of videos. In this work, we present an approach for physical violence detection on images obtained from video based on threshold active learning, that increases the classifier’s robustness in environments where it was not trained. The proposed approach consists of two stages: In the first stage, pre-trained neural network models are trained on initial datasets, and we use a threshold ($\mu$) to identify those images that the classifier considers ambiguous or hard to classify. Then, they are included in the training dataset, and the model is retrained to improve its classification performance. In the second stage, we test the model with video images from other environments, and we again employ ($\mu$) to detect ambiguous images that a human expert analyzes to determine the real class or delete the ambiguity on them. After that, the ambiguous images are added to the original training set and the classifier is retrained; this process is repeated while ambiguous images exist. The model is a hybrid neural network that uses transfer learning and a threshold $\mu$ to detect physical violence on images obtained from video files successfully. In this active learning process, the classifier can detect physical violence in different environments, where the main contribution is the method used to obtain a threshold $\mu$ (which is based on the neural network output) that allows human experts to contribute to the classification process to obtain more robust neural networks and high-quality datasets. The experimental results show the proposed approach’s effectiveness in detecting physical violence, where it is trained using an initial dataset, and new images are added to improve its robustness in diverse environments. Full article

Physical aggression is a serious and widespread problem in society, affecting people worldwide. It impacts nearly every aspect of life. While some studies explore the root causes of violent behavior, others focus on urban planning in high-crime areas. Real-time violence detection, powered by artificial intelligence, offers a direct and efficient solution, reducing the need for extensive human supervision and saving lives. This paper is a continuation of a systematic mapping study and its objective is to provide a comprehensive and up-to-date review of AI-based video violence detection, specifically in physical assaults. Regarding violence detection, the following have been grouped and categorized from the review of the selected papers: 21 challenges that remain to be solved, 28 datasets that have been created in recent years, 21 keyframe extraction methods, 16 types of algorithm inputs, as well as a wide variety of algorithm combinations and their corresponding accuracy results. Given the lack of recent reviews dealing with the detection of violence in video, this study is considered necessary and relevant. Full article

Traditional methods for predicting sow reproductive cycles are not only costly but also demand a larger workforce, exposing workers to respiratory toxins, repetitive stress injuries, and chronic pain. This occupational hazard can even lead to mental health issues due to repeated exposure to violence. Managing health and welfare issues becomes pivotal in group-housed animal settings, where individual care is challenging on large farms with limited staff. The necessity for computer vision systems to analyze sow behavior and detect deviations indicative of health problems is apparent. Beyond observing changes in behavior and physical traits, computer vision can accurately detect estrus based on vulva characteristics and analyze thermal imagery for temperature changes, which are crucial indicators of estrus. By automating estrus detection, farms can significantly enhance breeding efficiency, ensuring optimal timing for insemination. These systems work continuously, promptly alerting staff to anomalies for early intervention. In this research, we propose part of the solution by utilizing an image segmentation model to localize the vulva. We created our technique to identify vulvae on pig farms using infrared imagery. To accomplish this, we initially isolate the vulva region by enclosing it within a red rectangle and then generate vulva masks by applying a threshold to the red area. The system is trained using U-Net semantic segmentation, where the input for the system consists of grayscale images and their corresponding masks. We utilize U-Net semantic segmentation to find the vulva in the input image, making it lightweight, simple, and robust enough to be tested on many images. To evaluate the performance of our model, we employ the intersection over union (IOU) metric, which is a suitable indicator for determining the model’s robustness. For the segmentation model, a prediction is generally considered ‘good’ when the intersection over union score surpasses 0.5. Our model achieved this criterion with a score of 0.58, surpassing the scores of alternative methods such as the SVM with Gabor (0.515) and YOLOv3 (0.52). Full article

Older adult maltreatment (OAM) is a global problem that has attracted increasing attention due to the ageing population and its severe impact on victim health. Thus, this study aims to analyse the prevalence of certain health conditions in people ≥ 60 years old whom physicians from a local healthcare unit suspected to be victims of maltreatment. The specific objectives are to determine the prevalence rates of health-related risk factors, traumatic injuries and intoxications, mental disorders, and physical disorders. We conducted a real-world, retrospective, observational, and cross-sectional study based on secondary data analyses of electronic health records and healthcare registers of patients at the Local Healthcare Unit of Matosinhos (2001–2021). Information was obtained based on codes from the International Classification of Diseases, codes from the International Classification of Primary Care, and clinical notes (according to previously defined keywords). We identified 3092 suspected victims of OAM, representing 4.5% of the total population analysed. This prevalence is lower than the known rates. We also found that some health risk factors, traumatic injuries and intoxications, mental health disorders, and physical disorders presented higher rates in the suspected victims than among the total population. In this age group, we cannot assume that these health problems are only related to a possible current victimisation process; they could also be associated with adverse childhood experiences or intimate partner violence, among other forms of violence, all of which can lead to cumulative effects on the victim’s health. This evidence increases healthcare providers’ responsibility in detecting and reporting all cases of suspected maltreatment. Full article

The double vulnerability of women with disabilities places them at the center of this research paper. Intersectionality is key in research on gender-based violence. This study analyzes the perspective of the victims and non-victims themselves on this issue, through a comparative analysis between women with and without disabilities, at two levels of analysis: quantitative, through the adaptation of various scales (Assessment Screen-Disability/AAS-D, and the Woman Abuse Screening Tool/WAST), and qualitative, with semi-structured interviews (open scripts and different themes), and focus groups with experts from the associative network. The results obtained indicate that the most frequent type of violence is physical, followed by psychological and sexual, mainly perpetrated by partners. The higher their level of education, the more they defend themselves; receiving public aid can be a risk factor for domestic and sexual violence, and belonging to the associative movement and having paid work outside the home act as preventive measures. In conclusion, it is necessary to establish strategic protection measures and effective detection and intervention systems to make victims visible and care for them. Full article

Social media applications, such as Twitter and Facebook, allow users to communicate and share their thoughts, status updates, opinions, photographs, and videos around the globe. Unfortunately, some people utilize these platforms to disseminate hate speech and abusive language. The growth of hate speech may result in hate crimes, cyber violence, and substantial harm to cyberspace, physical security, and social safety. As a result, hate speech detection is a critical issue for both cyberspace and physical society, necessitating the development of a robust application capable of detecting and combating it in real-time. Hate speech detection is a context-dependent problem that requires context-aware mechanisms for resolution. In this study, we employed a transformer-based model for Roman Urdu hate speech classification due to its ability to capture the text context. In addition, we developed the first Roman Urdu pre-trained BERT model, which we named BERT-RU. For this purpose, we exploited the capabilities of BERT by training it from scratch on the largest Roman Urdu dataset consisting of 173,714 text messages. Traditional and deep learning models were used as baseline models, including LSTM, BiLSTM, BiLSTM + Attention Layer, and CNN. We also investigated the concept of transfer learning by using pre-trained BERT embeddings in conjunction with deep learning models. The performance of each model was evaluated in terms of accuracy, precision, recall, and F-measure. The generalization of each model was evaluated on a cross-domain dataset. The experimental results revealed that the transformer-based model, when directly applied to the classification task of the Roman Urdu hate speech, outperformed traditional machine learning, deep learning models, and pre-trained transformer-based models in terms of accuracy, precision, recall, and F-measure, with scores of 96.70%, 97.25%, 96.74%, and 97.89%, respectively. In addition, the transformer-based model exhibited superior generalization on a cross-domain dataset. Full article

Domestic violence (DV) is a public health crisis that threatens both the mental and physical health of people. With the unprecedented surge in data available on the internet and electronic health record systems, leveraging machine learning (ML) to detect obscure changes and predict the likelihood of DV from digital text data is a promising area health science research. However, there is a paucity of research discussing and reviewing ML applications in DV research. Methods: We extracted 3588 articles from four databases. Twenty-two articles met the inclusion criteria. Results: Twelve articles used the supervised ML method, seven articles used the unsupervised ML method, and three articles applied both. Most studies were published in Australia (n = 6) and the United States (n = 4). Data sources included social media, professional notes, national databases, surveys, and newspapers. Random forest (n = 9), support vector machine (n = 8), and naïve Bayes (n = 7) were the top three algorithms, while the most used automatic algorithm for unsupervised ML in DV research was latent Dirichlet allocation (LDA) for topic modeling (n = 2). Eight types of outcomes were identified, while three purposes of ML and challenges were delineated and are discussed. Conclusions: Leveraging the ML method to tackle DV holds unprecedented potential, especially in classification, prediction, and exploration tasks, and particularly when using social media data. However, adoption challenges, data source issues, and lengthy data preparation times are the main bottlenecks in this context. To overcome those challenges, early ML algorithms have been developed and evaluated on DV clinical data. Full article

Background: Violence against women is a relevant health and social problem with negative consequences on women’s health. The interaction between genome and environmental factors, such as violence, represents one of the major challenges in molecular medicine. The Epigenetics for WomEn (EpiWE) project is a multidisciplinary pilot study that intends to investigate the epigenetic signatures associated with intimate partner and sexual violence-induced stress-related disorders. Materials and Methods: In 2020, 62 women exposed to violence (13 women suffering from sexual violence and 49 from Intimate Partner Violence, IPV) and 50 women with no history of violence were recruited at the Service for Sexual and Domestic Violence. All women aged 18–65 were monitored for their physical and psychological conditions. Blood samples were collected, and DNAs were extracted and underwent the epigenetic analysis of 10 stress-related genes. Results: PTSD prevalence in victims was assessed at 8.1%. Quantitative methylation evaluation of the ten selected trauma/stress-related genes revealed the differential iper-methylation of brain-derived neurotrophic factor, dopamine receptor D2 and insulin-like growth factor 2 genes. These genes are among those related to brain plasticity, learning, and memory pathways. Conclusions: The association of early detection of posttraumatic distress and epigenetic marker identification could represent a new avenue for addressing women survivors toward resilience. This innovative approach in gender-based violence studies could identify new molecular pathways associated with the long-term effects of violence and implement innovative protocols of precision medicine. Full article

Studies conducted worldwide indicate that adverse childhood experiences (ACEs) are among the most intense and frequent sources of stress, considerably influencing mental and physical health while also resulting in risk behaviors in adulthood. Methodology: We used data from the Pilot National Survey of Adversity and Sexual Abuse in Childhood (2020), conducted by CUIDA UC, which comprises the Adverse Childhood Experiences International Questionnaire [ACE-IQ] (Adapted). The cross-sectional methodology used made it possible to directly calculate the prevalence of adverse childhood experiences in the population sampled, at a single point in time. We performed a bivariate and univariate descriptive analysis, a correlation analysis, and a multivariate analysis, all of which will be detailed in the section entitled “General Data Analysis Procedure”. Results: We found equally high rates of adverse childhood experiences in men and women, with community violence exhibiting the highest prevalence. We found significant low- to moderate-sized associations between the multiple types of ACEs considered and mental health problems, substance use problems, criminal behaviors, and intrafamily violence (IFV), which differed between men and women. Significant correlations were detected between the ACE score and mental health, substance use, criminal behaviors, and IFV in both men and women. Importantly, ACEs were found to be predictors of all of these variables, with differences observed between men and women. Conclusions: Nearly all participants reported having had at least one ACE and more than half reported had four or more ACEs. Those who had had four or more ACEs were more likely to report problems throughout their life. Having an ACE of any type was found to be a better predictor of mental health problems and IFV in men than in women and might be a stronger risk factor for substance use and criminal behaviors in women than in men. Full article

Each year, 275 million children worldwide are exposed to domestic violence (DV) and suffer negative mental and physical health consequences; however, only a small proportion receive assistance. Pediatricians and child psychiatrists can play a central role in identifying threatened children. We reviewed experiences of DV screening in pediatric and child and adolescent mental health services (CAMHS) to understand its feasibility and provide clues for its implementation. We performed bibliographic research using the Sapienza Library System, PubMed, and the following databases: MEDLINE, American Psychological Association PsycArticles, American Psychological Association PsycInfo, ScienceDirect, and Scopus. We considered a 20-year interval when selecting the articles and we included studies published in English between January 2000 and March 2021. A total of 23 out of 2335 studies satisfied the inclusion criteria. We found that the prevalence of disclosed DV ranged from 4.2% to 48%, with most prevalence estimates between 10% and 20%. Disclosure increases with a detection plan, which is mostly welcomed by mothers (70–80% acceptance rates). Written tools were used in 55% of studies, oral interviews in 40%, and computer instruments in 20%. Mixed forms were used in three studies (15%). The most used and effective tool appeared to be the Conflict Tactics Scale (CTS) (30% of studies). For young children, parental reports are advisable and written instruments are the first preference; interviews can be conducted with older children. Our research pointed out that the current literature does not provide practical clinical clues on facilitating the disclosure in pediatric clinics and CAMHS. Further studies are needed on the inpatient population and in the field of children psychiatry. Full article