BMJ supportive & palliative care · 2025
Artificial intelligence for early palliative referral in adult oncology: opportunities, challenges and future directions.
СтатьяОбзорНазвание 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.
Gupta Atul Kumar
Аннотация
BACKGROUND: In oncology, early palliative care enhances quality of life and may increase survival; yet, because of resource limitations and overestimation of prognosis, referrals frequently happen late. Due to a shortage of specialised workers, this issue is made worse in low- and middle-income countries (LMICs).
KEY EVIDENCE: Referral systems powered by artificial intelligence (AI) (such as electronic health record-based triggers and machine learning risk models) have the potential to identify individuals who might benefit from earlier palliative integration. According to preliminary research, AI interventions resulted in more timely referrals. In a large cancer cohort, a machine learning model, for example, could increase access to early palliative treatment by approximately 8-15% without requiring more consultations. Although algorithm-based default referrals in a randomised experiment greatly increased consultation rates (44% vs 8%), patient-reported outcomes were not improved.
GLOBAL RELEVANCE: By giving referrals priority, such technologies have the potential to revolutionise LMICs, where specialised palliative care is limited. Nevertheless, there is a lack of evidence from LMIC contexts, and context-specific adaptation is required.
CONCLUSION: Globally, AI-powered referral systems have a great deal of promise to enable earlier integration of palliative care in oncology. Among the main drawbacks are algorithmic bias, problems with data quality and the requirement for ethical supervision and clinical workflow integration. Although preliminary data appear promising, further prospective validation is necessary to guarantee that more referrals result in significant patient benefits. With a focus on patient-centred care, AI systems should support clinical judgement rather than replace it.
KEY EVIDENCE: Referral systems powered by artificial intelligence (AI) (such as electronic health record-based triggers and machine learning risk models) have the potential to identify individuals who might benefit from earlier palliative integration. According to preliminary research, AI interventions resulted in more timely referrals. In a large cancer cohort, a machine learning model, for example, could increase access to early palliative treatment by approximately 8-15% without requiring more consultations. Although algorithm-based default referrals in a randomised experiment greatly increased consultation rates (44% vs 8%), patient-reported outcomes were not improved.
GLOBAL RELEVANCE: By giving referrals priority, such technologies have the potential to revolutionise LMICs, where specialised palliative care is limited. Nevertheless, there is a lack of evidence from LMIC contexts, and context-specific adaptation is required.
CONCLUSION: Globally, AI-powered referral systems have a great deal of promise to enable earlier integration of palliative care in oncology. Among the main drawbacks are algorithmic bias, problems with data quality and the requirement for ethical supervision and clinical workflow integration. Although preliminary data appear promising, further prospective validation is necessary to guarantee that more referrals result in significant patient benefits. With a focus on patient-centred care, AI systems should support clinical judgement rather than replace it.