Search Results (10)

This paper explores advancements in real-time talking-head generation, focusing on overcoming challenges in Audio Feature Extraction (AFE), which often introduces latency and limits responsiveness in real-time applications. To address these issues, we propose and implement a fully integrated system that replaces conventional AFE models with OpenAI’s Whisper, leveraging its encoder to optimize processing and improve overall system efficiency. Our evaluation of two open-source real-time models across three different datasets shows that Whisper not only accelerates processing but also improves specific aspects of rendering quality, resulting in more realistic and responsive talking-head interactions. Although interviewer training systems are considered a potential application, the primary contribution of this work is the improvement of the technical foundations necessary for creating responsive AI avatars. These advancements enable more immersive interactions and expand the scope of AI-driven applications, including educational tools and simulated training environments. Full article

Background: Transphyseal humeral separations (TPHS) are rare injuries often associated with non-accidental trauma, necessitating accurate diagnosis. This study aims to assess the accuracy of diagnosis of TPHS. Methods: A retrospective review was conducted at five academic pediatric institutions to identify all surgically treated TPHS in patients up to 4 years of age over a 25-year period. Demographics, misdiagnosis rates, and reported misdiagnoses were noted. Comparative analyses were performed to analyze the effects of patient age and injury mechanism on misdiagnosis rates. Results: Seventy-nine patients (average age: 17.4 months) were identified, with injury mechanisms including accidental trauma (n = 49), non-accidental trauma (n = 21), Cesarean-section (n = 6), and vaginal delivery (n = 3). Neither age nor injury mechanism were significantly associated with diagnostic accuracy in the emergency department (ED)/consulting physician group. ED/consulting physicians achieved an accurate diagnosis 46.7% of the time, while radiologists achieved an accurate diagnosis 26.7% of the time. Diagnostic accuracy did not correlate with Child Protective Services (CPS) involvement or with a delay in surgery of more than 24 h. However, a significant correlation (p = 0.03) was observed between injury mechanism and misdiagnosis rates. Conclusion: This multicenter analysis is the largest study assessing TPHS misdiagnosis rates, highlighting the need for raising awareness and considering advanced imaging or orthopedic consultation for accurate diagnosis. This also reminds orthopedic surgeons to always have vigilant assessment in treating pediatric elbow injuries. Level of Evidence: Level III–Retrospective Cohort Study. Full article

When responding to allegations of child sexual, physical, and psychological abuse, Child Protection Service (CPS) workers and police personnel need to elicit detailed and accurate accounts of the abuse to assist in decision-making and prosecution. Current research emphasizes the importance of the interviewer’s ability to follow empirically based guidelines. In doing so, it is essential to implement economical and scientific training courses for interviewers. Due to recent advances in artificial intelligence, we propose to generate a realistic and interactive child avatar, aiming to mimic a child. Our ongoing research involves the integration and interaction of different components with each other, including how to handle the language, auditory, emotional, and visual components of the avatar. This paper presents three subjective studies that investigate and compare various state-of-the-art methods for implementing multiple aspects of the child avatar. The first user study evaluates the whole system and shows that the system is well received by the expert and highlights the importance of its realism. The second user study investigates the emotional component and how it can be integrated with video and audio, and the third user study investigates realism in the auditory and visual components of the avatar created by different methods. The insights and feedback from these studies have contributed to the refined and improved architecture of the child avatar system which we present here. Full article

In the current study, an observational procedure, recorded in video, was used to evaluate the quality of parent-child interactions in a sample of vulnerable Portuguese families (n = 47) with school-aged children followed by Child Protective Services (CPS). The study sought to explore if the families presented different profiles of parent-child interaction quality, and to characterize such profiles in terms of discrete behaviors observed, parenting outcome variables, and families’ sociodemographic and CPS referral characteristics. The parent-child dyads took part in a 15 minutes structured task and parents completed self-report measures (affection, parenting behaviors, and stress). Discrete behaviors of parents and children during interactions were coded with a micro-analytic coding procedure. The global dimensions of the parents’ interactions were coded with a global rating system. A latent profile analysis, estimated with global dimensions, identified two subgroups, one subgroup in which parents displayed higher quality interactions (n = 12), and another subgroup in which parents displayed lower quality interactions (n = 35). Further analyses comparing the subgroups determined that the higher quality subgroup presented more positive behaviors, and the lower quality subgroup presented more negative behaviors during the interactions. No further differences or associations were found regarding the parenting outcome variables, and the families’ sociodemographic and CPS referral characteristics. The findings are in line with prior studies, suggesting that vulnerable families may frequently present depleted parent-child interactions. However, given the small sample size, future studies should replicate the described procedures and analyses in larger sample sizes. Full article

The ability to narrate routine familiar events develops gradually during middle childhood, in increasingly higher levels of coherence and temporal cohesion. Improvements in episodic memory are also observed, reflecting children’s increasing ability to recall specific circumstances of past events and personal experiences. Even though several studies have evaluated children’s narrative abilities and episodic memory, little information is available regarding the children exposed to risks that justify their referral to Child Protective Services (CPS). The current study analysed children’s narrative abilities and episodic memory performance, according to the circumstances related to the referral to CPS. Event schema representation, narrative coherence, narrative temporal cohesion, and episodic memory concerning routine and specific personal events in family context were analysed in a sample of 56 school-aged children followed by the CPS in Portugal. Children referred to CPS due to disruptive behaviour presented higher episodic memory performance, compared to those exposed to domestic violence, neglect, and abuse. No significant differences were found between groups regarding narrative abilities related to familiar routine events. Results highlight the relevance of evaluating the adverse circumstances that lead to CPS referral, considering the levels of risk and danger involved, given its differential effects on children’s episodic memory development. Full article

Sleep-related infant deaths, including Sudden Infant Death Syndrome (SIDS), are the number one cause of death in infants between 28 days and one year of life. Nearly half of families experiencing a sleep-related infant death in Kansas were involved with the Department of Children and Families Child Protective Services (CPS), making CPS staff a priority for safe sleep training. This study assessed the impact of the two-day Kansas Infant Death and SIDS (KIDS) Network Safe Sleep Instructor (SSI) train-the-trainer program on CPS staffs’ knowledge of the American Academy of Pediatrics safe sleep recommendations. Training was attended by 43 participants, 27 (63%) of whom were employed by CPS. CPS staff had significantly lower baseline knowledge on the 10-item pretest (t = 3.33, p = 0.002), but both CPS and other attendees showed significant improvement by posttest (t = 8.53, p < 0.001 and t = 4.44, p < 0.001, respectively). Following SSI certification, CPS SSIs provided more safe sleep training to professionals than other SSIs (1051 vs. 165, respectively), and both groups of SSIs were able to significantly increase the knowledge of their trainees. Overall, the KIDS Network SSI training was successful. The innovative partnership with CPS allowed for provision of training to a group not historically targeted for safe sleep education. Full article

Education and healthcare professionals are crucial in detecting and reporting child abuse and neglect. However, signs of child abuse are often undetected, and professionals tend to underreport their suspicions of abuse and neglect. This qualitative study aimed to examine experiences, attitudes, perspectives, and decision-making skills of healthcare and education professionals with regard to identifying and reporting child abuse and to gain insight into how detection and reporting can be improved. Semi-structured interviews were conducted with 49 Dutch professionals working in child health care, mental health care, primary schools, and secondary schools. The I-Change model was used as a theoretical framework to organize the results. Many professionals believe they miss child abuse signs in their daily work, partially due to a lack of focus on child abuse. Further, professionals indicated having insufficient knowledge of child abuse, and lack communication skills to detect or discuss signs indicative of child abuse in conversations with parents or children. As for risk assessment, professionals barely use structured instruments even though these are regarded as very helpful in the decision-making process. Finally, professionals experience deficits in the cooperation with child welfare organizations, and in particular with Child Protective Services (CPS). Various directions for improvement were discussed to overcome barriers in child abuse detection and reporting, including developing tools for detecting and assessing the risk of child abuse and improving communication and information transfer between organizations. Full article

Background. Existing literature has documented a strong relationship between parental mental illness and child maltreatment, but little is known about the prevalence of mental illness among childbearing women. In the present study, linked administrative records were used to identify the prevalence of maternal mental health (MH) disorders documented at birth and determine the associated likelihood of maltreatment reports during infancy. Materials and Methods. Vital records for California’s 2006 birth cohort were linked to hospital discharge and Child Protective Services (CPS) records. The International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) billing codes from the mother’s delivery hospitalization were used to determine diagnosed maternal MH disorders for 551,232 infants born in 2006, and reports of alleged maltreatment were documented from CPS records. Vital birth records were used to control for sociodemographic factors. Finally, the associated risk of reported maltreatment during the first year of life was examined using generalized linear models. Results. Among infants in this statewide birth cohort, 2.8% were born to a mother with a documented MH disorder, of which 41.3% had documented maternal substance abuse issues versus less than 0.5% of infants born to mothers without a diagnosed MH disorder. Further, 34.6% of infants born to mothers with a MH disorder were reported to CPS within one year, and a majority of those reports were made within the first month of life (77.2%). In contrast, among children born to mothers without a MH disorder, 4.4% were reported to CPS during infancy. After controlling for sociodemographic factors, the rate of CPS reports during infancy for infants born to mothers with a MH disorder but no substance use disorder was 2.6 times that of infants born to mothers without a MH disorder (95% CI = 2.47, 2.73). Among infants born to mothers with MH and substance use disorders, the rate of CPS reports during infancy was 5.69 times that of infants born to mothers without a MH disorder (95% CI = 5.51, 5.87). Conclusions. Administrative records provide a method for identifying infants born to mothers with MH disorders, enabling researchers to track rates over time and generate population-level data to inform policy development and improve service delivery. Full article

This exploratory study examines combinations of income-tested welfare benefits and earnings, as they relate to the likelihood of child maltreatment investigations among low-income families with young children participating in a nutritional assistance program in one U.S. state (Wisconsin). Using a sample of 1065 parents who received the Special Supplemental Nutrition Assistance Program for Women, Infants, and Children (WIC) benefits in late 2010 and early 2011, we find that relying on either work in the absence of other means-tested welfare benefits, or a combination of work and welfare benefits, reduces the likelihood of CPS involvement compared to parents who rely on welfare benefits in the absence of work. Additionally, we find that housing instability increases the risk of CPS involvement in this population. The findings from this investigation may be useful to programs serving low-income families with young children, as they attempt to identify safety net resources for their clientele. Full article

Background: Children are believed to be more likely to be reported for maltreatment while they are working with mental health or social service professionals. This “surveillance bias” has been claimed to inflate reporting by fifty percent or more, and has been used to explain why interventions such as home visiting fail to reduce official maltreatment reporting rates. Methods: We use national child abuse reporting data (n = 825,763), supplemented by more detailed regional data from a multi-agency administrative data study (n = 7185). We determine the percentage of all re-reports made uniquely by mental health and social service providers within and across generations, the report sources which could be subject to surveillance bias. Results: At three years after the initial Child protective services (CPS) report, the total percentage of national reports uniquely made by mental health or social service providers is less than 10%, making it impossible that surveillance bias could massively inflate CPS reporting in this sample. Analysis of national data find evidence of a very small (+4.54%) initial surveillance bias “bump” among served cases which decays to +1.84% within three years. Our analysis of regional data showed similar or weaker effects. Conclusions: Surveillance bias effects appear to exist, but are very small. Full article