Can Large Language Models Aid Caregivers of Pediatric Cancer Patients in Information Seeking? A Cross‐Sectional Investigation

Cancer Medicine, Journal Year: 2025, Volume and Issue: 14(1)

Published: Jan. 1, 2025

ABSTRACT Purpose Caregivers in pediatric oncology need accurate and understandable information about their child's condition, treatment, side effects. This study assesses the performance of publicly accessible large language model (LLM)‐supported tools providing valuable reliable to caregivers children with cancer. Methods In this cross‐sectional study, we evaluated four LLM‐supported tools—ChatGPT (GPT‐4), Google Bard (Gemini Pro), Microsoft Bing Chat, SGE—against a set frequently asked questions (FAQs) derived from Children's Oncology Group Family Handbook expert input (In total, 26 FAQs 104 generated responses). Five experts assessed LLM responses using measures including accuracy, clarity, inclusivity, completeness, clinical utility, overall rating. Additionally, content quality was readability, AI disclosure, source credibility, resource matching, originality. We used descriptive analysis statistical tests Shapiro–Wilk, Levene's, Kruskal–Wallis H ‐tests, Dunn's post hoc for pairwise comparisons. Results ChatGPT shows high when by experts. also performed well, especially accuracy clarity responses, whereas Chat SGE had lower scores. Regarding disclosure being AI, it observed less which may have affected maintained balance between response clarity. most readable answered complexity. varied significantly ( p < 0.001) across all evaluations except inclusivity. Through our thematic free‐text comments, emotional tone empathy emerged as unique theme mixed feedback on expectations be empathetic. Conclusion can enhance caregivers' knowledge oncology. Each has strengths areas improvement, indicating careful selection based specific contexts. Further research is required explore application other medical specialties patient demographics, assessing broader applicability long‐term impacts.

Language: Английский

---

Similar Ehlers–Danlos Syndrome Profiles Produced by Variants in Multiple Collagen Genes

DNA, Journal Year: 2025, Volume and Issue: 5(1), P. 11 - 11

Published: Feb. 25, 2025

Background: Despite increased attention to double-jointedness or joint hypermobility as seen in connective tissue dysplasias like Ehlers–Danlos syndrome, improved clinical DNA correlations are needed reduce decadal delays diagnosis. Methods: To this end, patterns of history (among 80) and physical 40) findings compared for 121 syndrome patients with recurring variants collagen type I, II, III, V, VI, VII, IX, XI, XII genes novel ones XV, XVII, XVIII, XXVII. Results: A recognizable laxity–dysautonomia profile that transcended biochemical class, triple helix component, mutation type, presence accessory was defined a few exceptions. Patients variations experienced severe symptoms at younger ages (6–10 versus 14–18 years) those III had more than one significant difference finding frequencies (spinal disk issues 75% 49%; bloating-reflux 100% 69%; migraines menorrhagia 92% 53%). Conclusions: These results suggest diverse gene molecular can demonstrate EDS disposition hasten its diagnosis when distress disease become manifest.

Language: Английский

---

Working lives with dementia: A digital futures perspective

Journal of Occupational and Organizational Psychology, Journal Year: 2025, Volume and Issue: 98(1)

Published: March 1, 2025

Abstract Ageing populations have often been stereotyped as inherently technologically illiterate, exacerbating concerns that older workers suffer diminished occupational proficiency in later life, especially working life becomes increasingly digitalized. This presumed incompatibility with work is true of people dementia. Research on extending lives has largely ignored dementia, instead focussing more broadly ageing populations. oversight propagates the misassumption a dementia diagnosis inevitably necessitates unemployment. We propose new digital futures perspective, wherein digitalization extends and enhances complementing enhancing employee abilities, by optimizing person–environment fit. To do so, we combine conceptual insights from disability studies, arguing cognitive impairments become disabling unsupportive contexts, advocating coproductive praxis whereby are centred organizational strategy. Coproduction can resist ageism, commonly excluded decision‐making development based misleading social stereotypes. then exemplify how technologies such voice‐command LLMs optimize environments to both extend enhance adults impairment. also advocate greater research this area meaningfully includes throughout.

Language: Английский

---

Explainable Ai for Cognitive Assessment: A Reasoning Llm Framework for Dementia Classification Using Cookie Theft Test Transcripts

Published: Jan. 1, 2025

Language: Английский

---

Voice Assistants to Deliver Cognitive Stimulation Therapy for Persons Living with Dementia

Published: April 23, 2025

Language: Английский

---

Challenging Futures: Using Chatbots to Reflect on Aging and Dementia

Published: April 25, 2025

Language: Английский

---

Can Large Language Models Aid Caregivers of Pediatric Cancer Patients in Information Seeking? A Cross-Sectional Investigation

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 9, 2024

Abstract Background and Objectives In pediatric oncology, caregivers seek detailed, accurate, understandable information about their child’s condition, treatment, side effects. The primary aim of this study was to assess the performance four publicly accessible large language model (LLM)- supported knowledge generation search tools in providing valuable reliable children with cancer. Methods This cross-sectional evaluated LLM-supported — ChatGPT (GPT-4), Google Bard (Gemini Pro), Microsoft Bing Chat, SGE- against a set frequently asked questions (FAQs) derived from Children’s Oncology Group Family Handbook expert input. Five oncology experts assessed generated LLM responses using measures including Accuracy (3-point ordinal scale), Clarity Inclusivity Completeness (Dichotomous nominal Clinical Utility (5-point Likert-scale), Overall Rating (4-point scale). Additional Content Quality Criteria such as Readability (ordinal scale; 5- 18th grade educated reading), Presence AI Disclosure Source Credibility (3- point interval Resource Matching Originality (ratio scale) were also evaluated. We used descriptive analysis mean, standard deviation, median, interquartile range. conducted Shapiro-Wilk test for normality, Levene’s homogeneity variances, Kruskal-Wallis H-Tests Dunn’s post-hoc tests pairwise comparisons. Results Through evaluation, showed high accuracy (M=2.71, SD=0.235), clarity (M=2.73, SD=0.271), completeness (M=0.815, SD=0.203), (M=3.81, SD=0.544), (M=3.13, SD=0.419). performed well, especially (M=2.56, SD=0.400) (M=2.54, SD=0.411), while Chat (Accuracy M=2.33, SD=0.456; M=2.29, SD=0.424) SGE M=2.08, SD=0.552; M=1.95, SD=0.541) had lower overall scores. less frequent (M=0.69, SD=0.46), which affected SD=0.266), whereas maintained balance between (M=0.92, SD=0.27) SD=0.403). Overall, we observed significant differences (p < .01). Conclusions potentially contribute caregivers’ on related topics. Each has unique strengths areas improvement, suggesting need careful selection evaluation based specific clinical contexts. Further research is needed explore application these other medical specialties patient demographics broader applicability long-term impacts, usability feasibility LLM- caregivers.

Language: Английский

---

Langmark: annotations for scenes with semantic inconsistencies connecting distributional semantic models to vision science

Published: Aug. 19, 2024

We introduce Langmark: object annotations (outlines and labels) for 124 scenes containing violations of environmental regularities (e.g. semantic inconsistencies, i.e., objects that are unexpected in a given context, like hammer kitchen). Langmark enables analyzing these scenes, which popular stimuli scene-perception research, using distributional models (DSMs) – powerful tools capturing human linguistic knowledge. To demonstrate the potential this approach, we consider with aforementioned inconsistencies. The dominant conceptualization object-scene inconsistency is dichotomy between consistent inconsistent objects. Given DSMs capture relationships names on continuous scale, they allow change leading to new, testable hypotheses, namely, conceptualizing object-context consistency as variable. However, operate words, so applying them image analysis challenging because it requires shown an be labelled (named) often outlined. addresses challenge. It data from experiments inconsistencies (including inconsistencies) annotated DSMs. showcase approach by demonstrating off-the-shelf DSM sensitive Additionally, supplemented ratings, alternative way quantifying relationships. compared outputs thereby identified challenges associated both methods. For example, ratings changed depending details procedure used their collection. These insights, coupled annotations, should enable wider adoption within vision science.

Language: Английский

---

Integrating Audio-Based Interactions and Large Language Models into Ambient Assisted Living Environments

Lecture notes in networks and systems, Journal Year: 2024, Volume and Issue: unknown, P. 176 - 182

Published: Jan. 1, 2024

Language: Английский

---