Geriatric nursing (New York, N.Y.) · 2026
Leveraging patient and their surrogate caregiver communication with clinicians to predict palliative care decisions: A speech processing study.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Song Jiyoun, Beigi Homayoon, Davoudi Anahita, Moon Royce, McDonald Margaret V, Sridharan Sridevi, Stanley June, Bowles Kathryn H, Shang Jingjing, Stone Patricia W, Topaz Maxim
Аннотация
BACKGROUND: Managed Long-term Care (MLTC) integrates various services for individualized, holistic care, where palliative care (PC) is beneficial for complex patients, but initial communication processes in PC remain underexplored.
PURPOSE: (1) To establish a pipeline for analyzing speech characteristics and (2) to predict the PC preference (i.e., acceptance or refusal).
METHODS: This study analyzed 79 recorded phone calls (15.4 h) between patients and clinicians (i.e., nurses or social workers), extracting acoustic features and transcriptions. These features were combined with demographic and clinical data to predict the PC preference.
FINDINGS: Among the recordings, 42 participants (53%) accepted PC, while 37 (47%) declined. Higher energy and pitch levels were observed among those who accepted (0.065 vs. 0.59 and 289 vs. 260 Hz). Gradient Boosting was the most effective classifier (F-score of 65.5%). Discussion This study demonstrates the potential of using speech processing algorithms to analyze patient-clinician conversations, identifying PC preference-related speech characteristics.
PURPOSE: (1) To establish a pipeline for analyzing speech characteristics and (2) to predict the PC preference (i.e., acceptance or refusal).
METHODS: This study analyzed 79 recorded phone calls (15.4 h) between patients and clinicians (i.e., nurses or social workers), extracting acoustic features and transcriptions. These features were combined with demographic and clinical data to predict the PC preference.
FINDINGS: Among the recordings, 42 participants (53%) accepted PC, while 37 (47%) declined. Higher energy and pitch levels were observed among those who accepted (0.065 vs. 0.59 and 289 vs. 260 Hz). Gradient Boosting was the most effective classifier (F-score of 65.5%). Discussion This study demonstrates the potential of using speech processing algorithms to analyze patient-clinician conversations, identifying PC preference-related speech characteristics.