AI Is at the Door: Should Oncology Nurses Let It In?

Artificial intelligence (AI) is transforming oncology nursing, bringing cutting-edge tools such as machine learning (ML), deep learning (DL), and natural language processing (NLP) into daily practice.

Although there is enthusiasm about AI’s potential to predict patient prognosis, assess treatment adverse effects, streamline workflows, and personalize patient care, experts urge careful implementation and attention to ethical and safety considerations.

AI at the Bedside

Across the US, multiple National Cancer Institute–Designated Cancer Centers, including City of Hope in Duarte, California; Dana-Farber Cancer Institute in Boston, Massachusetts; Fred Hutch Cancer Center in Seattle, Washington; and The University of Texas MD Anderson Cancer Center in Houston, are among those investigating and integrating AI into their practices.^1,2

At MD Anderson, AI tools are used by physicians, and their potential for nursing practice is being explored, Kim Slusser, PhD, MSN, RN, NEA-BC, CHPN, chief nursing officer ad interim said in an interview with Oncology Nursing News.

AI has the potential to enhance nursing in palliative care, she explained. Nurses often use validated screening tools to identify patients who may benefit from earlier intervention. Advancing technologies may help consolidate key patient information and highlight who has palliative care needs, improving timing and integration, according to Slusser.

She also believes AI tools have the power to fuse relevant information from electronic health records (EHRs), allowing nurses to view a comprehensive patient profile without manually navigating multiple systems. This capability would support more meaningful discussions with patients about their goals, values, and priorities.

Although AI in palliative care is exciting, Slusser emphasized that its use is still in its early stages.

Additional opportunities include leveraging ambient listening technology, with patient consent, to capture conversations that automatically feed into documentation and clinical notes. Nurses would then be able to review and validate the information, reducing administrative burden and optimizing workflow.

“Something that we found…with our oncologists and palliative care providers is that the burden of administrative tasks and documentation can sometimes get in the way of having the time needed to have these rich conversations with patients and families,” Slusser said.

Between 2020 and 2022, researchers in several studies used AI for predictive modeling to forecast short-term mortality, risk of complications, and need for palliative services. In addition, NLP techniques have been used to extract key information from clinical notes to identify patients’ supportive care needs.³ Although these tools are promising, they require further validation.

Biomarkers and Prognostication

In biomarker testing, interpretation, and prognostication, AI has been applied to improve diagnostic accuracy, risk stratification, and individualized treatment planning.

For instance, advances in ML and DL have significantly improved cancer prognoses by accurately predicting tumor recurrence, metastasis, and survival outcomes. In patients with advanced or metastatic pancreatic cancer, ML models trained on multicenter data effectively predicted mortality risk and helped personalize chemotherapy selection.⁴ The retrospective analysis of 191 patients demonstrated a receiver operating characteristic–area under the curve of 0.81 for FOLFIRINOX (fluorouracil, leucovorin, irinotecan, and oxaliplatin) and 0.82 for gemcitabine/nab-paclitaxel.

In a study of more than 1200 patients with advanced melanoma from the Netherlands, higher levels of AI-detected tumor-infiltrating lymphocytes were independently associated with improved immune checkpoint inhibition response and survival, the authors noted.⁵

An article published in Cancers explored how AI may reduce cancer disparities by integrating social determinants of health into risk prediction and screening.⁶ Moreover, the researchers discussed AI’s role in radiology and digital pathology, as well as its potential to help close gaps in diversity among clinical trial enrollment via NLP algorithms that scan EHRs for key clinical and demographic data.

At City of Hope, oncologists are using Microsoft Azure to learn about patients before appointments and build treatment plans.² In 2024, the institution onboarded more than 150,000 patients with extensive medical histories. Through Azure, physicians were delivered a more digestible review of patients’ charts.

Nurses at Temple University Hospital in Philadelphia, Pennsylvania, are using their smartphones to record and transcribe patient interactions as part of a pilot program.⁷

“This allows us to look people in the eye again.… We stare at the computer too much,” said Melissa Culligan, PhD, RN, MS, in an interview with Oncology Nursing News. Culligan is the director of thoracic surgery research at Temple University Hospital and codirector of the Temple Health Mesothelioma and Pleural Disease Program at Fox Chase Cancer Center in Philadelphia, Pennsylvania.

Dragon Ambient eXperience (DAX) Copilot, a type of ambient AI scribe, records patient-provider discussions and generates detailed clinical notes. Afterwards, clinicians can review and amend the note as needed. Temple began using DAX to simplify processes, enabling greater focus on patient care, Culligan said.

In an NEJM Catalyst report, medical and technology experts at The Permanente Medical Group evaluated the use of ambient AI scribes across more than 2.5 million interactions during its first year of deployment. Usage increased consistently over time, and physician characteristics such as age or years in practice did not predict adoption. Most physicians reported positive experiences, and patients reported neutral to positive impacts on visit quality. Estimated documentation time savings exceeded 15,700 hours over the 1-year period for users, equivalent to more than 1800 workdays, compared with nonusers.⁸

AI is not only cutting down on note-taking but also assisting in analyzing CT scans and MRIs alongside a radiologist’s review. Coreline Soft, an AI-powered diagnostic imaging software, has been implemented at Temple University Hospital to more efficiently assess the extent of a patient’s disease, predict risks, and explore treatment plans.⁹

Although AI generates much optimism, Culligan is also cautious. “You have to be careful of where data [are] going and…trust that [the tools] have been vetted,” she said.

One of the most prominent and rapidly advancing areas of AI is large language models (LLM), such as ChatGPT and Gemini. In recent years, the oncology community has increased its use of LLM-based tools to identify clinical characteristics of cancer, develop targeted treatment approaches, and improve patient understanding.¹⁰

Researchers from UCLA evaluated the effectiveness and readability of AI-generated and physician-generated educational materials for patients with cancer. The materials were created by ChatGPT-4 and separately by a group of 3 physicians on topics such as cytopenia, diarrhea/constipation, fatigue, nausea, neuropathy, and loss of appetite. Both were then scored by blinded review, and the results showed that ChatGPT-4’s efficacy in creating materials was comparable to that of physician experts.¹¹

Real-World Adoption of AI

Although some oncology nurses are skeptical, others are eager to step onto the AI front lines.

Margia Fonseca, MSN, RN, OCN, of Hackensack Meridian Health System’s data science team, began touring all Hackensack Meridian Health sites earlier this year as a leader in using EKAM, a new AI tool powered by Google’s ML.

“Data scientists [and health care professionals are from] 2 different worlds,” Fonseca said in an interview with Oncology Nursing News. “I live at the bedside, so I see the gaps in real time that they don’t. EKAM…is built from what health care professionals need in the middle of a shift. I see myself as a liaison between the two.”

Initially started as a pilot project, EKAM is set to go network-wide in 2026. The tool can read and analyze 2 years of unstructured provider notes at lightning speed, providing providers with a comprehensive patient history in 45 seconds.

“I’ve launched a soft tour of EKAM to 2 units already, and I’ve had nurses tell me they don’t even read the progress note,” she said. “If you don’t read the note, you’re missing vital information. Maybe this patient had a mastectomy.… Maybe they had a past cancer.… Maybe they had a reaction to platelets.”

With EKAM, nurses and APPs can instantly learn about a patient, saving time that can be redirected to patient care. Fonseca says that, while her training focuses on reassuring nurses and helping them feel comfortable with AI, she will also consider their feedback.

“It’s not meant to replace your job.… It’s like an assistant that helps you with redundant tasks,” Fonseca said. “Is it 100% accurate? Humans make mistakes. Machines make mistakes. But I think it’s going to get better and better with time.”

Risk vs Reward

As AI tools move from pilot programs to everyday clinical use, questions about evidence, ethics, and equity are becoming increasingly urgent.

AI in oncology nursing remains in its early stages, but it’s already reshaping health care. This moment was evident at the 2025 American Society of Hematology Annual Meeting and Exposition, which featured 6 sessions dedicated entirely to AI research, emphasizing the need for nurses to be actively involved in AI development.¹²

Doing so requires continued education and training. A potential guide is the NURSES framework (navigate AI basics, utilize AI strategically, recognize AI pitfalls, skills support, ethics in action, and shape the future) published in Nursing Outlook. The framework offers an approach to help nurses adapt to digital health tools while maintaining clinical judgment.¹³

Despite its potential, barriers to the widespread adoption of AI remain. These include a shortage of randomized controlled trials, algorithmic bias, lack of standardized nursing data sets, and limited integration into everyday workflows.¹⁴

AI integration also raises important ethical and professional considerations, such as the risk of inaccurate outputs or hallucinations, as well as overreliance on technology, which could weaken nurses’ independent assessment.¹⁵

For patients, privacy is a main concern. Because AI systems require access to sensitive clinical data, keeping data safe and maintaining patient trust is essential.

Nursing leaders have emphasized the importance of transparency with patients about when and how AI is used in care, noting that there must be clear communication, particularly regarding consent and data use. To maintain the nurse-patient relationship, nurses must ensure these technologies don’t compromise the nature of human interactions, according to the American Nurses Association (ANA).¹⁶ In addition, the ANA states that it is the nurse’s role to educate patients and their families about AI to help dispel myths and alleviate fears.

The future of AI in oncology nursing will be shaped not only by innovation but also by nurses who help guide its ethical and patient-centered integration into everyday care.

“In time, we’re going to see that [AI will] be a lifesaver once we get more into the process and make [the technology] better,” Fonseca said.

References

1. Unlocking the unlimited potential of AI for cancer research. Cancer AI Alliance. Accessed January 11, 2026. https://www.canceralliance.ai
2. City of Hope uses Microsoft Azure and AI to rapidly onboard thousands of patients per year. Microsoft. April 11, 2025. Accessed January 11, 2026. https://www.microsoft.com/en/customers/story/23397-city-of-hope-azure-open-ai-service
3. Reddy V, Nafees A, Raman S. Recent advances in artificial intelligence applications for supportive and palliative care in cancer patients. Curr Opin Support Palliat Care. 2023;17(2):125-134. doi:10.1097/SPC.0000000000000645
4. Koo J, Choi G, Cheon J, et al. Predicting chemotherapy response in patients with advanced or metastatic pancreatic cancer using machine learning. JCO Clin Cancer Inform. 2025;9:e2500124. doi:10.1200/CCI-25-00124
5. Schuiveling M, van Duin IAJ, ter Maat L, et el. Artificial intelligence–detected tumor-infiltrating lymphocytes and outcomes in anti–PD-1–based treated melanoma. JAMA Oncol. 2025;11(12):1470-1478. doi:10.1001/jamaoncol.2025.4072
6. Srivastav AK, Singh A, Singh S, Rivers B, Lillard JW Jr, Singh R. Revolutionizing oncology through AI: addressing cancer disparities by improving screening, treatment, and survival outcomes via integration of social determinants of health. Cancers. 2025;17(17):2866. doi:10.3390/cancers17172866
7. Alvino G. Temple Health debuts AI clinical notetaking software to improve physician satisfaction and patient care. Temple Health. June 12, 2025. Accessed January 11, 2026. https://www.templehealth.org/about/news/ai-clinical-notetaking-software
8. Tierney AA, Gayre G, Hoberman B, et al. Ambient artificial intelligence scribes: learnings after 1 year and over 2.5 million uses. NEJM Catal Innov Care Deliv. 2025;6(5). doi:10.1056/CAT.25.0040
9. Leonard N. As Pennsylvania health systems rapidly adopt AI, lawmakers take steps to regulate the technology. WHYY. October 29, 2025. Accessed January 11, 2026. https://whyy.org/articles/artificial-intelligence-health-care-pennsylvania/
10. Benson R, Kenny C, Ganjouei AA, et al. Large language models in population oncology: a contemporary review on the use of LLMs to support data collection, aggregation, and analysis in cancer care and research. JCO Clin Cancer Inform. 2025;9:e2500112. doi:10.1200/CCI-25-00112
11. Hui G, Jiang J, Dommaraju S, et al. Artificial intelligence vs. physicians: quality of oncology patient education materials. JCO Oncol Pract. 2024;20(10):408. doi:10.1200/OP.2024.20.10_suppl.408
12. 68th ASH^® Annual Meeting, Meeting Program. American Society of Hematology. Accessed February 10, 2026. https://submit.hematology.org/program?is_sponsored=0&is_highlighted=0&is_on_demand=0&streaming_allowed=0&filterbox=false&legendbox=false&timezone=false&pre_filter=All&viewType=list&segment=programme&segmentname=Presentations&upcoming=0&order=starts_at
13. Hoelscher SH, Pugh A. N.U.R.S.E.S. embracing artificial intelligence: a guide to artificial intelligence literacy for the nursing profession. 2025;73(4):102466. doi:10.1016/j.outlook.2025.102466
14. Zhou T, Luo Y, Li J, et al. Application of artificial intelligence in oncology nursing: a scoping review. Cancer Nurs. 2024;47(6):436-450. doi:10.1097/NCC.0000000000001254
15. Yoon SM, Lyu J, Djunadi TA. Navigating artificial intelligence (AI) accuracy: a meta-analysis of hallucination incidence in large language model (LLM) responses to oncology questions. J Clin Oncol. 2025;43(suppl 16):e13686. doi:10.1200/JCO.2025.43.16_suppl.e13686
16. The ethical use of artificial intelligence in nursing practice. American Nurses Association. 2022. Accessed February 10, 2026. https://www.nursingworld.org/globalassets/practiceandpolicy/nursing-excellence/ana-position-statements/the-ethical-use-of-artificial-intelligence-in-nursing-practice_bod-approved-12_20_22.pdf