We are delighted to continue the Academic Express section, which aims to facilitate academic exchanges and further the development of oncology nursing. In every issue, we present ground-breaking research contributions from esteemed members in the field.
In this issue, we are privileged to feature a significant study led by Dr Muna Alkhaifi from Odette Cancer Centre in Toronto, and a multinational team of researchers from the Multinational Association of Supportive Care in Cancer (MASCC) Survivorship Study Group. Their research, titled "Application of Artificial Intelligence in Symptom Monitoring in Adult Cancer Survivorship: A Systematic Review," provides a comprehensive evaluation of the role of artificial intelligence (AI) in monitoring and managing symptoms across the cancer survivorship trajectory.
As the global population ages and the number of cancer survivors continues to rise, the integration of AI technologies, such as machine learning and natural language processing, holds immense promise for transforming patient care. This study systematically reviews the effectiveness, safety, and usability of AI-driven approaches in symptom monitoring, offering valuable insights into their potential to enhance patient outcomes. The authors also provide actionable recommendations for the future integration of AI into clinical practice, emphasizing the need for standardized data collection, system interoperability, and patient-centered customization.
This research underscores the transformative potential of AI in oncology care and highlights critical considerations for its implementation, making it a vital contribution to the ongoing discourse on improving cancer survivorship care. With a growing and aging population, the number of cancer survivors is expected to rise. Artificial intelligence (AI) is becoming increasingly common in health care, with promises to transform patient care and administrative processes. AI technologies, such as machine learning (ML) and natural language processing (NLP) have already been applied to tasks in healthcare, including analyzing clinical documentation, treatment planning, toxicity management and follow-up care. Given the widespread application of AI in healthcare, it is imperative to assess the comprehensiveness, effectiveness and safety of AI in symptom monitoring and management for cancer survivors. There is also a lack of systemic evaluations addressing AI-driven symptom monitoring in cancer survivorship. In light of this gap, my team and I undertook a systematic review to assess the current state of science regarding the integration of AI in symptom monitoring within the adult cancer survivorship trajectory, from diagnosis through the end of life.
Forty-one studies were included in this review. AI prediction models typically performed well, with high precision and accuracy in tasks such as symptom detection. However, a small proportion of studies found limited accuracy, sensitivity and specificity of their AI models. Data was available to determine the effectiveness of AI approaches in monitoring and improving patient symptoms. Several studies found that their AI approach was highly accurate at identifying symptoms. Overall, AI showed utility in symptom prediction, monitoring and outcomes. User acceptance and satisfaction were mixed, with some studies reporting a higher user acceptance rate for their AI approach, while others reported mixed results regarding user acceptance, suggesting the need for tailored AI solutions.
After discussion and consensus among authors, we provided three broad recommendations on the future integration of AI into clinical practice for cancer symptom monitoring:
1. To effectively implement AI into clinical practice, symptom monitoring should be standardized in terms of frequency and method of data collection to achieve consistent data across all healthcare settings. This will allow for strategic use of analytics to detect trends and test interventions to improve patient outcomes.
2.System integration of AI requires significant investment in infrastructure and technology. It is instrumental to ensure electronic health records and analytical tools are interoperable and designed to meet individual needs so relevant and timely information can be collected that better supports decision-making.
3. Enhancing the usability of AI applications is essential for increasing patient engagement. Customizing these applications to accommodate individual patient preferences can foster active participation in symptom monitoring. This can include allowing patients to personalize features such as data input methods and notification settings, which can improve adherence with the system, improving patients’ experience and well-being.
Overall, the findings of this review indicate a promising potential for AI to enhance symptom monitoring in diverse cancer settings by incorporating various data sources, including patient text, patient-reported outcomes, and physiologic measurements. The review underscores the significant role of AI in both symptom monitoring and overall patient care throughout cancer survivorship.
Dr. Muna Alkhaifi (Corresponding author)
Reference:
Tabataba Vakili, S., Haywood, D., Kirk, D., Abdou, A. M., Gopalakrishnan, R., Sadeghi, S., Guedes, H., Tan, C. J., Thamm, C., Bernard, R., Wong, H. C. Y., Kuhn, E. P., Kwan, J. Y. Y., Lee, S. F., Hart, N. H., Paterson, C., Chopra, D. A., Drury, A., Zhang, E., Raeisi Dehkordi, S., … Multinational Association of Supportive Care in Cancer (MASCC) Survivorship Study Group (2024). Application of Artificial Intelligence in Symptom Monitoring in Adult Cancer Survivorship: A Systematic Review. JCO clinical cancer informatics, 8, e2400119. https://doi.org/10.1200/CCI.24.00119. Application of Artificial Intelligence in Symptom Monitoring in Adult Cancer Survivorship: A Systematic Review
