The American Journal of Emergency Medicine, Journal Year: 2024, Volume and Issue: 88, P. 140 - 144
Published: Nov. 27, 2024
Language: Английский
Big Data and Cognitive Computing, Journal Year: 2025, Volume and Issue: 9(1), P. 16 - 16
Published: Jan. 20, 2025
As suicide rates increase globally, there is a growing need for effective, data-driven methods in mental health monitoring. This study leverages advanced artificial intelligence (AI), particularly natural language processing (NLP) and machine learning (ML), to identify suicidal ideation from Twitter data. A predictive model was developed process social media posts real time, using NLP sentiment analysis detect textual emotional cues associated with distress. The aims potential risks accurately, while minimizing false positives, offering practical tool targeted interventions. achieved notable performance, an accuracy of 85%, precision 88%, recall 83% detecting posts. Advanced preprocessing techniques, including tokenization, stemming, feature extraction term frequency–inverse document frequency (TF-IDF) count vectorization, ensured high-quality data transformation. random forest classifier selected its ability handle high-dimensional effectively capture linguistic patterns linked ideation. model’s reliability supported by precision–recall AUC score 0.93, demonstrating real-time monitoring intervention. By identifying behavioral triggers, such as isolation bullying, this framework provides scalable efficient solution support, contributing significantly prevention strategies worldwide.
Language: Английский
Citations
0
Journal of the Intensive Care Society, Journal Year: 2025, Volume and Issue: unknown
Published: March 4, 2025
Auto-intoxication represents a considerable number of Intensive care unit (ICU) admissions. Literature focusing on characteristics and outcomes ICU-admitted auto-intoxicated patients is limited. We aimed to identify the quantity intensive admissions attributable auto-intoxication in Belgian tertiary center, examine these patients, assess impact covid-19 pandemic variables. In this retrospective cohort study, all admitted ICU JESSA hospital, Hasselt, Belgium with diagnosis between January 1st 2017 December 31st 2022, were included study. collected data patient characteristics, comorbidities, type intoxication including length hospital stay mortality. total, 342 dataset, covering 2.44% from 2022. Although occurred age-categories, age group 18 29 years old, showed highest prevalence (24%). More woman (57.6%) than man (42.4%) Of 21.6% had history previous suicide attempt 36.5% auto-intoxication. The most common substances ingested sedatives (44.4%), illicit drugs (20.2%), analgesics (17.8%), antidepressants (16.7%) whereas 54.1% cases, combination was ingested. Type separated into four groups: attempt, accidental, iatrogenic recreational use accounting for 71.3% auto-intoxications. remained stable over 5-year study period. An Covid-19 could not be established. Overall mortality low an ICU- 1.75% 2.6% respectively. Total at time dataset closure, other hand, 7.2%. resource utilization relatively high risk recidivism substantial. in-hospital after low, although have substantial death come discharge.
Language: Английский
Citations
0
Annales Médico-psychologiques revue psychiatrique, Journal Year: 2025, Volume and Issue: unknown
Published: March 1, 2025
Citations
0
Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)
Published: May 29, 2024
Abstract Suicide is a major public health challenge worldwide with an increasing trend. Identifying risk factors for suicide attempts and deaths may help find useful ways to prevent suicide. We aimed determine the trend related of in Rafsanjan. This retrospective study included all cases from 2018 2022 Rafsanjan, city southeast Iran. The information was extracted registration system Rafsanjan University Medical Sciences. Univariable multivariable logistics regression models were used investigate affecting death. A total 2039 suicide, including 1932 (94.75%) 107 (5.25%) recorded during period. frequency per 100,000 people increased compared both genders. In adjusted model, odds males 6.48 (95% CI 3.39–12.42) times higher than females. Also, unemployed subjects housewives 2.64 1.50–4.67) 7.45 3.08–18.07) employed respectively. Finally, education less diploma had 10.85 1.48–79.54) university education. present research showed that pattern has been since 2018, we see upward coming years, which requires further investigation preventive measures. Male gender, low level, unemployment, being housewife associated highest
Language: Английский
Citations
1
The American Journal of Emergency Medicine, Journal Year: 2024, Volume and Issue: 88, P. 140 - 144
Published: Nov. 27, 2024
Language: Английский
Citations
1