Artificial intelligence across the cancer care continuum.

INTRODUCTION

Artificial intelligence (AI) is poised to be a transformative technology that promises to change the way we care for patients with cancer. Our evolving understanding of cancer biology, coupled with the exponential increase in the availability of genomic information, electronic health records (EHRs), and medical imaging, has created a unique and timely opportunity to leverage the power of AI to revolutionize cancer care. The potential of AI to improve cancer care spans the entire spectrum of the disease, from prevention and early detection to treatment, survivorship, and end‐of‐life care. 1 , 2

AI‐powered tools are being developed and implemented to enhance our ability to assess individual cancer risk, improve the accuracy and efficiency of screening programs, refine the precision of diagnostic procedures, personalize treatment strategies based on individual patient characteristics, and provide comprehensive support to patients and their families throughout their cancer journey. For example, AI algorithms can analyze genomic data to identify individuals at high risk for developing specific cancers, enabling targeted prevention strategies. AI‐powered image analysis tools can assist radiologists and pathologists in detecting subtle abnormalities in medical images, improving the accuracy and speed of cancer diagnosis. AI can also be used to develop personalized treatment plans based on a patient's unique genetic profile and clinical history, maximizing the chances of treatment success and minimizing the risk of side effects.

This review article aims to provide an overview of AI's current and potential applications in oncology practice, encompassing the entire cancer care continuum. In addition to exploring the diverse applications of AI in oncology, this review also addresses the key challenges and ethical considerations associated with implementing AI in cancer care. These include the critical need for rigorous AI algorithm validation to ensure accuracy, reliability, and clinical utility. We also discuss the importance of addressing ethical concerns, such as data privacy and security, potential algorithm bias, and the need for transparency and explainability in AI‐driven decision making. Furthermore, we emphasize the crucial role of seamless integration of AI tools into existing clinical workflows to facilitate their effective adoption by health care professionals. Finally, we highlight the importance of interdisciplinary collaboration between clinicians, researchers, data scientists, technology developers, and policymakers to overcome these challenges and fully realize the transformative potential of AI in oncology. Our goal is to provide practicing physicians, researchers, and leaders in oncology and related fields with a comprehensive and nuanced understanding of the current state of the art of AI in oncology and the future directions and immense potential of this rapidly evolving field. By fostering informed discussion and promoting collaborative efforts, we hope to accelerate the responsible and effective implementation of AI in cancer care, ultimately improving outcomes and quality of life for all patients with cancer and their families.

PREVENTION AND EARLY INTERCEPTION

Cancer prevention represents a critical frontier in the fight against this disease. Identifying high‐risk individuals and improving early detection can significantly affect patient outcomes by personalizing screening. AI is a promising tool in this domain and has the potential to revolutionize how we approach cancer prevention. AI's ability to analyze vast data sets, identify complex patterns, and learn from experience is paving the way for more personalized and effective prevention strategies (Table 1 ). 3 , 15 , 20

One of the most promising applications of AI in cancer prevention lies in its ability to assess and predict individual cancer risk. Algorithms can integrate diverse data sources, including multi‐omic information, EHRs, lifestyle factors (e.g., smoking, diet, physical activity), environmental exposures, and family history, to generate personalized risk profiles. 3 , 15 , 20 For example, AI‐powered tools can analyze mammography images in conjunction with genetic markers and clinical data to predict a woman's risk of developing breast cancer. 4 , 5 Similarly, AI systems can assess an individual's risk for lung cancer based on their smoking history, exposure to carcinogens, and other relevant factors. 9 , 10 A recent study demonstrated the power of AI in predicting pancreatic cancer risk by analyzing disease trajectories from EHRs. Their deep learning models, particularly using Transformer architectures, achieved remarkable predictive performance (area under the receiver operating characteristic curve = 0.88) for identifying high‐risk patients up to 36 months before diagnosis by analyzing temporal sequences of disease codes from 6 million Danish patients and 3 million US veterans. This work represents a significant advance in using AI for early detection of aggressive cancers that typically present late. 14 These predictive models can empower health care providers to identify high‐risk individuals who may benefit from more intensive screening, chemoprevention, or lifestyle modifications. However, it is crucial to acknowledge the ethical considerations surrounding AI‐driven risk prediction. Potential biases in the data used to train these algorithms must be carefully addressed to ensure equitable access to preventive interventions. 21 Furthermore, responsible data handling and privacy protection are paramount. 22 Pathologists, with their expertise in interpreting tissue samples and understanding disease mechanisms, play a crucial role in developing and validating AI‐driven risk‐prediction models that incorporate pathologic data. Pathologists and radiologists, with their expertise in interpreting tissue samples and medical images, respectively, play a crucial role in developing and validating AI‐driven risk‐prediction models. Radiologists ensure imaging data quality and consistency, reducing biases from heterogeneous data sets, whereas pathologists refine AI algorithms through accurate annotations and histopathologic validation. Their collaboration with AI developers is essential for refining these tools, ensuring clinical applicability, and maintaining high diagnostic accuracy and patient care standards.

SCREENING AND EARLY DIAGNOSIS

An accurate and timely cancer diagnosis is essential for effective treatment planning and improved patient outcomes. Radiology plays a pivotal role by providing noninvasive imaging of tumors, and AI is augmenting the expertise of radiologists by streamlining image interpretation and improving diagnostic accuracy (Table 1 ). AI algorithms analyze various imaging modalities, including computed tomography (CT) scans, magnetic resonance images (MRIs), and positron‐emission tomography scans, to detect subtle abnormalities that may indicate cancer, improving sensitivity and reducing false‐negative rates.

Expanding on AI's transformative role in radiology, one particularly effective application is that for mammography, in which AI enhances breast cancer detection through advanced image analysis. AI‐powered algorithms assist in analyzing mammograms, identifying subtle patterns and features indicative of early stage breast cancer. A landmark trial involving 105,934 participants demonstrated that AI‐assisted screening achieved a 29% higher cancer detection rate (6.4 vs. 5.0 per 1000 screened), primarily identifying early stage cancers and reducing radiologist workload by 44% without increasing the false‐positive rate compared with standard double reading. 8

AI is also advancing lung cancer screening by improving the detection of small lung nodules on CT scans. Identification of these nodules is essential for guiding biopsy decisions and enabling early diagnosis, which significantly improves the chances of successful curative treatment. In addition, AI is enhancing the accuracy and efficiency of tissue acquisition. AI‐powered image‐guidance systems assist interventional radiologists in precisely targeting lesions during biopsies, reducing sampling errors and improving diagnostic yield. For instance, AI algorithms can analyze real‐time imaging data from CT scans or MRIs to enhance needle placement accuracy for lung and prostate biopsies, thereby increasing the detection rate of clinically significant cancers and reducing the need for repeat procedures. 16 , 17

Once tissue samples are obtained, pathologists play a central role in diagnosing cancer through microscopic examination. AI is significantly enhancing pathology by automating histopathology image analysis, assisting pathologists in identifying cancer subtypes, grading tumors, and assessing disease progression. 18 , 19 , 23 , 24 AI‐powered tools can analyze digital pathology slides, detecting subtle cellular features and biomarkers that may otherwise be overlooked, thus improving diagnostic precision (Table 1 ). For example, AI is being applied in cytopathology to analyze Papanicolaou smears for the early detection of cervical cancer, help identify prostate cancer, and provide Gleason scores. Several US Food and Drug Administration‐approved, AI‐powered digital pathology solutions are now available, such as Paige Prostate, 25 which identifies areas in prostate biopsy images likely to harbor cancer, and Ibex Galen Prostate, 26 which assists in prostate cancer detection and grading.

Beyond imaging‐based and tissue‐based diagnostics, AI is playing an increasingly important role in molecular diagnostics by complementing these traditional methods with genomic and molecular insights, enabling a more comprehensive approach to cancer detection and treatment planning. AI from histopathology slides can assess tumor microenvironment and identify genetic mutations driving cancer growth and predict responses to targeted therapies or immunotherapy. 27 , 28 Deep‐learning approaches have been used to analyze tumor heterogeneity, predicting tumor evolution and resistance mechanisms and identifying prognostic biomarkers. Examples include analyses of histopathology slides to classify molecular subtypes of lung and colorectal cancer, evaluation of tumor microenvironments in renal cell carcinoma by quantifying immune cell infiltration, aiding in immunotherapy response predictions, detecting genetic alterations, such as EGFR and KRAS mutations in lung cancer and BRCA mutations in ovarian cancer, all of which hold significant predictive value and influence treatment decisions.

Although AI is improving cancer screening and early diagnosis, its success depends on collaboration between AI systems and medical professionals. 12 , 13 AI enhances efficiency by reducing workload and expediting image analysis, yet expert oversight remains essential to ensure accuracy, interpret complex cases, and provide clinical context. Radiologists and pathologists play a crucial role in integrating AI findings into practice, ensuring that these technologies enhance—not replace—expert clinical judgment. AI tools enable radiologists to work more efficiently, focusing on complex cases and improving overall patient care while reducing turnaround time for diagnostic interpretations.

TREATMENT

Effective cancer treatment requires a multifaceted approach, often involving combinations of surgical intervention, radiation therapy, and systemic therapies. AI is being integrated into each of these treatment modalities, offering the potential for more precise, personalized, and effective cancer care (Table 2 ). 3 , 15 , 20

Surgical treatment

Surgical intervention remains a cornerstone of cancer treatment for many solid tumors. AI is enhancing preoperative surgical planning, intraoperative navigation, and postoperative risk assessment, demonstrating an improvement in patient outcomes through several key applications.

AI‐assisted surgical planning allows surgeons to create detailed, patient‐specific surgical plans based on preoperative imaging data. 37 AI algorithms can analyze CT scans, MRI, and other imaging modalities to generate three‐dimensional models of the tumor and surrounding anatomy, enabling surgeons to visualize the surgical field and plan the optimal surgical approach. 32 , 38 This is particularly valuable in complex surgical cases, such as those involving tumors near critical structures. For example, in urologic oncology, AI can assist in planning complex robotic prostatectomies, helping surgeons to preserve nerve function and improve postoperative outcomes. 29

AI‐powered, intraoperative navigation systems provide real‐time guidance during surgery, helping surgeons to precisely locate tumors and avoid damaging healthy tissue. 37 These systems can overlay preoperative imaging data onto the surgical field, providing surgeons with a clear view of the tumor's location and margins. In robotic surgery, AI is being used to enhance surgical dexterity and precision. AI algorithms can analyze real‐time video feeds from the surgical robot to provide surgeons with enhanced visual feedback and assist with complex surgical maneuvers. Intraoperative data, such as tissue perfusion and blood flow, can also be analyzed to provide surgeons with real‐time information about the surgical field and help them make informed decisions during the procedure. 30 Furthermore, AI is being used to develop systems that can automate certain aspects of surgery, such as suturing or tissue dissection. Although fully autonomous robotic surgery is still in its early stages of development, AI‐assisted robotic systems are already improving surgical efficiency and precision. 39

AI also plays a role in postoperative risk assessment. AI algorithms can analyze patient data, including preoperative clinical information, intraoperative data, and pathologic findings, to predict the risk of postoperative complications. This information can help surgeons to identify high‐risk patients and implement preventive measures to reduce the likelihood of complications. In urologic oncology, AI can be used to predict the risk of complications after radical cystectomy, helping surgeons to tailor postoperative care and improve patient outcomes. 31 Pathologists, with their expertise in tissue analysis and understanding of tumor biology, are essential in providing the data that inform AI‐driven surgical planning and risk assessment. Their input ensures that the AI algorithms are trained on accurate and representative data, leading to more reliable and clinically relevant results.

Radiation treatment

Radiation therapy plays a crucial role in treating many types of cancer. AI is enhancing the precision and effectiveness of radiation therapy through several key applications. AI‐powered target volume delineation automates the process of identifying the tumor and surrounding healthy tissues on CT scans and other imaging modalities. This process, which is traditionally done manually by radiation oncologists, can be time‐consuming and prone to variability. In the future, AI algorithms may be able to quickly and accurately delineate target volumes, reducing the workload on radiation oncologists and improving the consistency of treatment planning. 34 Radiation treatment plans are also being optimized by using AI. AI algorithms can analyze patient‐specific data, including tumor location, size, and surrounding anatomy, to generate treatment plans that maximize the dose to the tumor while minimizing the dose to healthy tissues. This can reduce the risk of side effects and improve the therapeutic ratio of radiation therapy. Adaptive radiotherapy, which allows treatment plans to be adjusted during the course of radiation therapy based on changes in the tumor or surrounding tissues, is also being studied. 33 AI algorithms can analyze daily imaging data to track tumor shrinkage and changes in patient anatomy, allowing radiation oncologists to modify the treatment plan in real time to ensure that the tumor receives the optimal dose of radiation while sparing healthy tissues. Furthermore, AI is being used to personalize radiation dosing. AI algorithms can integrate patient‐specific data, including tumor characteristics, genetic information, and clinical factors, to determine the optimal radiation dose for each individual patient. This personalized approach to radiation dosing has the potential to improve treatment outcomes and reduce the risk of side effects.

Systemic treatment

Systemic treatments, such as chemotherapy, targeted therapy, and immunotherapy, play a vital role in treating many cancers. AI is transforming systemic treatment by enabling more personalized treatment selection, accelerating drug discovery, and improving toxicity monitoring. 35 AI algorithms can analyze vast amounts of genomic data, clinical information, and pathologic findings to predict a patient's response to different systemic therapies. 40 This allows medical oncologists to select the most effective treatment for each individual patient, maximizing the chances of success and minimizing the risk of side effects. AI is also accelerating the process of drug discovery and development. AI algorithms can analyze vast data sets of molecular information to identify potential drug targets and predict the efficacy of new drugs. 36 , 41 This can significantly reduce the time and cost associated with bringing new cancer drugs to market. AI is also playing an important role in monitoring treatment toxicity. AI algorithms can analyze patient data, including EHRs, laboratory results, and imaging data, to identify early signs of treatment‐related toxicity. 35 , 42 This allows medical oncologists to intervene early and adjust the treatment plan to minimize the risk of serious side effects. AI‐powered systems can also provide personalized recommendations for managing treatment‐related side effects, improving patient comfort and quality of life. 43

SURVIVORSHIP

As therapies improve, cancer survivorship is an increasingly important phase of cancer care because the number of individuals living beyond a cancer diagnosis continues to grow. AI has the potential to significantly improve the quality of life for cancer survivors by enhancing monitoring and surveillance, personalizing care plans, and addressing the long‐term physical and psychosocial effects of cancer and its treatment (Table 3 ). 3 , 15 , 20

AI can play a crucial role in monitoring cancer survivors for recurrence or complications. AI algorithms can analyze data from EHRs, including laboratory results, imaging reports, and clinical notes, to identify patterns and trends that may be indicative of cancer recurrence or treatment‐related complications. 44 , 52 For example, AI systems can track changes in tumor markers or analyze imaging data to detect early signs of recurrence, allowing for timely intervention.

Wearable sensor data, such as activity trackers or smartwatches, can also be integrated with AI systems to monitor patients' physical activity levels, sleep patterns, and other physiologic parameters, providing valuable insights into their overall health and well‐being. 45 These data can help health care providers to identify individuals who may be at risk for developing long‐term side effects, such as fatigue or cardiovascular problems, and to implement preventive measures.

Survivorship care plans can also be personalized using AI. AI algorithms can analyze individual patient data, including their cancer type, treatment history, genetic information, and lifestyle factors, to generate personalized recommendations for managing long‐term side effects, improving physical function, and addressing psychosocial needs. 53 For instance, AI can help tailor exercise programs to individual patients' abilities and preferences or provide personalized recommendations for managing pain or fatigue. AI‐powered systems can also connect cancer survivors with relevant resources and support groups, helping them to navigate the challenges of survivorship. 54

Furthermore, AI can play a crucial role in addressing the psychosocial needs of cancer survivors. AI‐powered chatbots or virtual assistants can provide emotional support, answer questions about survivorship issues, and connect patients with mental health professionals when needed. AI algorithms can also analyze patient data to identify individuals who may be at risk for developing anxiety, depression, or other mental health conditions, allowing for early intervention and support. 46

The expertise of medical oncologists, radiation oncologists, surgeons, pathologists, radiologists, and other specialists is crucial in developing and implementing AI‐driven survivorship care programs. 55 Their input ensures that the AI algorithms are trained on accurate and comprehensive data with the possibility that the recommendations generated by these systems are clinically relevant and tailored to the individual needs of each cancer survivor. 56 By leveraging the power of AI, we can improve the long‐term health and well‐being of cancer survivors, helping them to live full and productive lives after their cancer diagnosis.

END‐OF‐LIFE CARE

End‐of‐life care focuses on providing comfort, dignity, and support to patients with advanced cancer and their families. However, because of prognostic uncertainty and optimism bias, end‐of‐life planning does not occur in a timely manner, offering suboptimal benefit to patients. AI has the opportunity to become a valuable tool to enhance end‐of‐life care by improving symptom management, predicting prognosis, supporting decision making, and providing timely support to caregivers and families (Table 3 ). This is the most under‐developed area in applying AI to the cancer care continuum and thus offers a great deal of opportunity.

AI can play a crucial role in optimizing symptom management for patients receiving palliative care. AI algorithms can analyze patient data, including vital signs, medication records, and symptom reports, to identify patterns and trends that may indicate changes in a patient's condition. 50 , 51 This allows health care providers to proactively address emerging symptoms, such as pain, nausea, or shortness of breath, and to adjust treatment plans accordingly. AI‐powered systems can also provide personalized recommendations for managing symptoms, helping patients to achieve the best possible quality of life during this challenging time.

The prognosis for patients with advanced cancer is also being predicted using AI. AI algorithms can analyze patient data, including clinical information, laboratory results, and imaging data, to generate more accurate and personalized predictions of disease progression and survival. 47 , 48 This information can help patients and their families make informed decisions about their care, including whether to pursue further treatment or focus on comfort measures. AI‐powered prognostic tools can also assist health care providers in planning end‐of‐life care, ensuring that patients receive the appropriate level of support and resources. 49

In the future, AI could also support decision making for patients and families facing difficult choices about end‐of‐life care. AI‐powered decision‐support tools can provide patients and families with information about different treatment options, potential outcomes, and quality‐of‐life considerations. These tools could help patients and families to understand complex medical information and make informed decisions that are aligned with their values and preferences. AI can also facilitate communication between patients, families, and health care providers, ensuring that everyone is on the same page and that the patient's wishes are respected.

Furthermore, AI can provide valuable support to caregivers and families of patients receiving end‐of‐life care. AI‐powered chatbots or virtual assistants can offer emotional support, answer questions about end‐of‐life care, and connect caregivers with resources and support groups. AI algorithms could also analyze patient data to identify caregivers who may be at risk for burnout or compassion fatigue, allowing for early intervention and support. By providing caregivers with the tools and resources they need, AI can help them to provide the best possible care for their loved ones during this difficult time.

The expertise of medical oncologists, palliative care specialists, nurses, social workers, and other health care professionals is essential in developing and implementing AI‐driven end‐of‐life care programs. Their input ensures that the AI algorithms are trained on accurate and comprehensive data and that the recommendations generated by these systems are clinically relevant and tailored to the individual needs of each patient and their family. By leveraging the power of AI, we can enhance the quality of end‐of‐life care, providing patients and their families with comfort, dignity, and support during this challenging time.

CARE DELIVERY

Cancer care today resembles the management of other chronic conditions: every patient experiences a long trial of consultations, test results, treatments, and psychosocial encounters. The administrative load of documenting, synthesizing, and acting on that information now rivals the clinical work itself. Recent advances in natural language processing (NLP), speech‐to‐text, and large language models (LLMs) have turned summarization into one of the most mature, immediately deployable AI capabilities with the potential of directly improving patient experience and system capacity.

Current LLMs can summarize, assimilate, and rephrase information from complex clinical narratives, extract salient details, and draft encounter notes that require minimal clinician editing. 57 , 58 When paired with retrieval‐augmented generation, open‐source LLMs have used longitudinal patient trajectories from the MIMIC‐III Clinical Database (Medical Information Mart for Intensive Care, version 3; Beth Israel Deaconess Medical Center) for clinical text summarization while preserving temporal order and causal relationships. 59 These same models also excel at evidence distillation, positioning them as powerful tools for clinical‐decision support (CDS). 60 Pilot studies in oncology have illustrated LLMs as CDS tools in gynecologic, 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 prostate, 58 , 61 , 66 , 69 , 70 , 71 , 72 , 73 oropharyngeal, 74 , 75 , 76 , 77 , 78 , 79 lung 57 , 61 , 66 , 67 , 80 and various other cancers. 81 , 82 , 83 , 84

State‐of‐the‐art LLMs can now accept multimodal input, including live speech, and generate structured documentation in real time. This advance underpins digital scribes that capture the entire encounter and draft notes autonomously. Controlled and observational studies indicate that ambient documentation both shortens charting time and lowers cognitive burden. Microsoft's Nuance DAX Copilot, an Epic‐embedded tool, has reduced time‐in‐notes by 1 minute per visit, halved scores on the National Aeronautics and Space Administration Task Load Index, and signaled a reduction in clinician burnout. 85 Vendor surveys suggest that 70% of clinicians report better work–life balance and clinics regain 13–26 extra appointment slots per provider per month when documentation writes itself . 86 By using ABRIDGE, another ambient clinical documentation tool, 81% of clinicians reported easier documentation compared with the current workflows according to vendor surveys. A multispecialty trial at the University of Kansas Medical Center indicated higher professional satisfaction and lower cognitive load for nearly 100 clinicians already using legacy speech‐recognition tools. 87 Leading cancer institutes are now deploying ABRIDGE across hematology and medical oncology services 88 after pilots confirmed accurate handling of complex terminology and multilingual conversations. 89 Beyond easing clinician workload, ambient systems automatically produce lay‐language after‐visit summaries, reinforcing education and shared decision making for patients navigating intricate oncology care plans.

Beyond point‐of‐care support, AI is increasingly automating downstream administrative workflows, such as appointment scheduling, revenue‐cycle management, and population‐health outreach. Machine‐learning scheduling platforms, such as iQueue for Infusion Centers (LeanTaaS) level‐load daily infusion demand, shortening patient wait times and increasing chair use. 90 For billing, a Peterson Health Technology Institute task force report notes that an ambient scribe tech produces structured notes for preparing authorizations, submitting claims, and billing functions. 91 An IMO Health brief further demonstrates that ambient AI can auto‐map diagnoses to standardized codes, improving billing accuracy and overall return on investment, 92 and a separate case study of an AI‐driven charge‐capture platform documented a 15% increase in captured revenue and a 20% reduction in denials. 93 Finally, by writing discrete EHR fields, such as diagnosis, regimen, and next test date, AI systems enable population‐health dashboards to flag overdue survivorship or surveillance visits and trigger timely patient outreach. 94 , 95

EVIDENCE GENERATION AND SYNTHESIS

Contemporary clinical practice guidelines are informed by high‐quality, up‐to‐date evidence generated through clinical trials and synthesized through systematic reviews and meta‐analyses. Evidence generation requires protocol drafting, regulatory submission, eligibility screening, enrolment, and longitudinal outcome capture. Evidence synthesis includes literature search, study selection, data extraction, quality appraisal, and statistical pooling. When performed manually, these processes become time‐consuming, resource‐intensive, and prone to human error and variability, rendering the evidence outdated by the time a guideline is published. AI solutions, particularly LLMs, are now streamlining evidence‐generation and synthesis tasks. As these platforms mature, they promise to deliver a faster, more reliable flow of evidence for guideline developers and, ultimately, for patients.

LLMs are shrinking the interval between trial conception and first patient. A recent experiment has demonstrated that ChatGPT‐4 (OpenAI) produces protocol skeletons with objectives, eligibility criteria, and statistical plans, compressing weeks of authoring into hours. 96 Similarly, ChatGPT‐4 97 and Mistral 8x22B 98 (Mistral AI) generated multilingual, plain‐language consent forms that meet US readability standards and regulatory key‐information requirements, reducing iterative changes with ethics committees. Clinical trial matching is another avenue of active research. Real‐time NLP engines mine EHRs to match complex protocol criteria against patient attributes. 99 , 100 Beyond static matching, machine‐learning–driven, predictive‐enrichment phenomapping identifies biologically homogeneous subgroups most likely to benefit from an investigational therapy, enabling adaptive eligibility refinements during the trial itself. 99 To curb recruitment demands and ethical concerns about placebo exposure, sponsors increasingly build synthetic control cohorts from real‐world data or historical trials. A recent proof‐of‐concept study generated an entire cohort of synthetic patients with acute myeloid leukemia using generative AI and successfully reproduced the survival curves of a traditional control arm, highlighting double‐digit reductions in projected trial cost and duration without compromising fidelity. 101

Given the rapid advances in available evidence, automated workflows to facilitate different steps of living‐evidence synthesis is an area of growing interest. LLMs can transform the practice of evidence synthesis by formulating a comprehensive literature search, screening relevant citations, extracting prespecified data elements, performing critical appraisal, and completing statistical analyses. 102 , 103 , 104 , 105 , 106 , 107 These models can support a framework of living practice guidelines that are updated on a near real‐time basis as new evidence becomes available.

PATIENT‐REPORTED OUTCOMES

As cancer treatments prolong survival, focus has shifted from mere longevity to the cumulative burden of disease and treatment‐related comorbidities. Consequently, patient‐reported outcomes (PROs), such as health‐related quality of life, functional status, and self‐reported toxicities, have become core end points in both clinical trials and real‐world evidence studies. Yet traditional methods to capture PROs through clinic visits and through periodic surveys yield sporadic, heterogeneous data, limiting their usefulness for day‐to‐day management and policy decisions.

AI platforms now provide continuous, scalable capture and interpretation of PRO data. 108 NLP pipelines can extract symptom information from free‐text clinic notes, patient diaries, and secure‐message threads 109 while downstream machine‐learning models flag clinically meaningful deterioration and prompt timely intervention. 110 , 111 In multiple myeloma, for instance, a rule‐based NLP algorithm identified descriptors of fatigue, neuropathy, and bone pain in unstructured notes with near‐perfect agreement to manual abstraction (F1 = 0.95–0.99). 112 Beyond cross‐sectional capture, AI supports longitudinal symptom monitoring. 110 Examples include a hybrid model that combined weekly PROs (i.e., pruritus, pain, breast swelling) with serum cytokine levels and predicted grade ≥2 radiation dermatitis in patients with breast cancer before it became clinically evident, showcasing an area under the receiver operating characteristic curve of 0.78 (95% CI, 0.701–0.854). 113 In oropharyngeal cancers, a systematic review demonstrated that combining longitudinal PROs (dysphagia, xerostomia, and pain) with radiomics features forecasted late toxicities of chemoradiotherapy, with the majority of algorithms reaching area under the receiver operating characteristic curve ≥0.80 for grade ≥2 dysphagia. 114 At a population level, AI can harmonize disparate PRO instruments and impute missing data and surface trends that inform formulary decisions and supportive‐care resource allocation.

FUTURE DIRECTIONS

The next decade will be defined by a shift from stand‐alone AI tools to an interoperable, continuously learning ecosystem. Multimodal, long‐context foundation models already accept text, images, and, increasingly, genomic data in a single prompt, allowing cross‐modal reasoning that links, for example, radiographic response with emerging resistance mutations. 115 , 116 In addition, digital‐twin platforms are beginning to simulate individual tumor trajectories and treatment courses, offering a sandbox for dose optimization or surgery–radiation sequencing before any real‐world intervention. 117 Unlocking the data volume required for these advances will depend on privacy‐preserving federation and differential‐privacy overlays that let algorithms train across institutions without moving raw data, thereby increasing generalizability while satisfying data‐privacy guardrails. 118

Clinically, AI will drive a new generation of adaptive, equity‐conscious research and care. Automated protocol drafting, real‐time EHR eligibility alerts, predictive enrichment phenomapping, and synthetic or real‐world external controls are converging to shorten trial timelines and lower costs, making truly adaptive oncology studies feasible even for rare disease cohorts. 119 , 120 To keep guidelines current, LLM‐assisted, living systematic‐review pipelines will feed a feedback loop in which new evidence, once published or posted, is screened, extracted, and meta‐analyzed with minimal human lag. At the point of care, richer PRO streams, ambient documentation, and multimodal decision support will enable human‐AI teams in which the system surfaces uncertainty, presents ranked rationales, and invites expert override rather than replacing clinician judgement.

Sustaining these gains will require robust governance, life‐cycle monitoring, and a commitment to equity and environmental stewardship. 121 Regulatory frameworks, such as the US Food and Drug Administration's evolving CDS guidance and the European Union's AI Act are moving toward mandatory bias audits, subgroup performance reporting, and postdeployment drift detection, mirroring pharmacovigilance for drugs. Health‐system dashboards that track model performance, trigger automatic recalibration, and log clinician overrides will become standard infrastructure. Finally, the field must address the computational footprint of ever larger models through research into sparse architecture and on‐device inference. Investments in federated learning networks, bias‐remediation toolkits, digital‐twin validation consortia, and adaptive reimbursement schemes will ensure that the promise of AI translates into durable, equitable improvements for every patient living with or beyond cancer.

CONCLUSION

AI is rapidly transforming oncology, offering unprecedented opportunities to improve cancer care across the entire disease spectrum—from prevention and early detection to treatment, survivorship, and end‐of‐life care (Figure 1 ). AI‐powered tools enhance cancer risk assessment, optimize screening programs, refine diagnostic precision, and personalize treatment strategies, ultimately improving patient outcomes. Moreover, AI facilitates care delivery, streamlines evidence generation and syntheses, and optimizes the recording of PROs. In addition, AI‐driven advancements in fundamental biology, such as protein structure prediction and cancer genomics, are enhancing our understanding of the disease. At the same time, applications in robotic surgery, radiation therapy planning, drug discovery, and patient support are well poised to improve patient care and treatment outcomes.

However, realizing the full potential of AI in cancer care requires addressing health care workforce education 122 and critical challenges, such as rigorous validation of AI algorithms, ensuring data privacy, mitigating algorithmic bias, and seamlessly integrating AI into clinical workflows. 123 , 124

Educating health care professionals on AI principles, applications, and limitations is crucial for successful implementation. By fostering confidence in AI tools and ensuring their role as decision‐support systems, we enable meaningful engagement between AI and clinicians. Human–AI collaboration is at the heart of AI's success in oncology—AI excels in data analysis and predictive modeling, but human expertise is essential for clinical decision making, patient communication, and ethical considerations. The synergy between AI insights and physician judgment enables more precise, personalized, and patient‐centered care, making this collaboration the key to transforming cancer treatment.

Multidisciplinary collaboration among clinicians, researchers, data scientists, and technology developers is vital to refining AI models, ensuring interpretability, and expanding AI applications in oncology. As AI continues to evolve and access to high‐quality data improves, its role in cancer care will become even more effective.

The future of oncology is increasingly intertwined with AI. By embracing this technology responsibly, educating health care professionals, and fostering human–AI collaboration, we can usher in a new era of personalized, precise, and effective cancer care—ultimately improving outcomes and quality of life for all patients.

AUTHOR CONTRIBUTIONS

Irbaz Bin Riaz : Conceptualization; formal analysis; methodology; visualization; writing–original draft; writing–review and editing. Muhammad Ali Khan : Visualization; writing–original draft; writing–review and editing. Travis Osterman : Conceptualization; methodology; project administration; supervision; writing–original draft; writing–review and editing.

CONFLICT OF INTEREST STATEMENT

Travis J. Osterman reports research support and/or grants/contracts from the Conquer Cancer Foundation, the Evelyn Selby Stead Fund for Innovation, GE Healthcare, IBM, Microsoft, the National Cancer Institute, and Outcomes4Me; consulting/personal and/or speaking fees from Amazon, the American Society of Clinical Oncology, European Society of Medical Oncology, AstraZeneca, Biodesix, Cota Healthcare, Dedham Group, MD Outlook, the National Comprehensive Cancer Network, and Outcomes Insights; and ownership of the business FacultyCoaching.com outside the submitted work. Irbaz Bin Riaz and Muhammad Ali Khan disclosed no conflicts of interest.

ACKNOWLEDGMENTS

This work was supported by the National Cancer Institute (Grants 1R21CA274545‐01 and P30 CA068485).

DATA AVAILABILITY STATEMENT

Data are available in the article supporting information.

REFERENCES

1 Swanson K , Wu E , Zhang A , Alizadeh AA , Zou J . From patterns to patients: advances in clinical machine learning for cancer diagnosis, prognosis, and treatment . Cell . 2023 ; 186 ( 8 ): 1772 ‐ 1791 . doi: 10.1016/j.cell.2023.01.035 36905928

2 Riaz IB , Khan MA , Haddad TC . Potential application of artificial intelligence in cancer therapy . Curr Opin Oncol . 2024 ; 36 ( 5 ): 437 ‐ 448 . doi: 10.1097/cco.0000000000001068 39007164

3 Kakileti ST , Madhu HJ , Manjunath G , Wee L , Dekker A , Sampangi S . Personalized risk prediction for breast cancer pre‐screening using artificial intelligence and thermal radiomics . Artif Intell Med . 2020 ; 105 : 101854 . doi: 10.1016/j.artmed.2020.101854 32505418

4 Arefan D , Mohamed AA , Berg WA , Zuley ML , Sumkin JH , Wu S . Deep learning modeling using normal mammograms for predicting breast cancer risk . Med Phys . 2020 ; 47 ( 1 ): 110 ‐ 118 . doi: 10.1002/mp.13886 31667873 PMC6980268

5 Badré A , Zhang L , Muchero W , Reynolds JC , Pan C . Deep neural network improves the estimation of polygenic risk scores for breast cancer . J Hum Genet . 2021 ; 66 ( 4 ): 359 ‐ 369 . doi: 10.1038/s10038-020-00832-7 33009504

6 Yala A , Mikhael PG , Strand F , et al. Toward robust mammography‐based models for breast cancer risk . Sci Transl Med . 2021 ; 13 ( 578 ): eaba4373 . doi: 10.1126/scitranslmed.aba4373 33504648

7 Ha R , Chang P , Karcich J , et al. Convolutional neural network based breast cancer risk stratification using a mammographic dataset . Acad Radiol . 2019 ; 26 ( 4 ): 544 ‐ 549 . doi: 10.1016/j.acra.2018.06.020 30072292 PMC8114104

8 Hernström V , Josefsson V , Sartor H , et al. Screening performance and characteristics of breast cancer detected in the Mammography Screening with Artificial Intelligence trial (MASAI): a randomised, controlled, parallel‐group, non‐inferiority, single‐blinded, screening accuracy study . Lancet Digit Health . 2025 ; 7 ( 3 ): e175 ‐ e183 . doi: 10.1016/S2589-7500(24)00267-X 39904652

9 Levi M , Lazebnik T , Kushnir S , Yosef N , Shlomi D . Machine learning computational model to predict lung cancer using electronic medical records . Cancer Epidemiol . 2024 ; 92 : 102631 . doi: 10.1016/j.canep.2024.102631 39053365

10 Yeh MCH , Wang YH , Yang HC , Bai KJ , Wang HH , Li YCJ . Artificial intelligence–based prediction of lung cancer risk using nonimaging electronic medical records: deep learning approach . J Med Internet Res . 2021 ; 23 ( 8 ): e26256 . doi: 10.2196/26256 34342588 PMC8371476

11 Mikhael PG , Wohlwend J , Yala A , et al. Sybil: a validated deep learning model to predict future lung cancer risk from a single low‐dose chest computed tomography . J Clin Oncol . 2023 ; 41 ( 12 ): 2191 ‐ 2200 . doi: 10.1200/JCO.22.01345 36634294 PMC10419602

12 Wang J , Yu Y , Tan Y , et al. Artificial intelligence enables precision diagnosis of cervical cytology grades and cervical cancer . Nat Commun . 2024 ; 15 ( 1 ): 4369 . doi: 10.1038/s41467-024-48705-3 38778014 PMC11111770

13 Xu H , Tang RSY , Lam TYT , et al. Artificial intelligence–assisted colonoscopy for colorectal cancer screening: a multicenter randomized controlled trial . Clin Gastroenterol Hepatol . 2023 ; 21 ( 2 ): 337 ‐ 346.e3 . doi: 10.1016/j.cgh.2022.07.006 35863686

14 Placido D , Yuan B , Hjaltelin JX , et al. A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories . Nat Med . 2023 ; 29 ( 5 ): 1113 ‐ 1122 . doi: 10.1038/s41591-023-02332-5 37156936 PMC10202814

15 Nazha A , Komrokji R , Meggendorfer M , et al. Personalized prediction model to risk stratify patients with myelodysplastic syndromes . J Clin Oncol . 2021 ; 39 ( 33 ): 3737 ‐ 3746 . doi: 10.1200/JCO.20.02810 34406850 PMC8601291

16 Harder C , Pryalukhin A , Quaas A , et al. Enhancing prostate cancer diagnosis: artificial intelligence–driven virtual biopsy for optimal magnetic resonance imaging‐targeted biopsy approach and Gleason grading strategy . Mod Pathol . 2024 ; 37 ( 10 ): 100564 . doi: 10.1016/j.modpat.2024.100564 39029903

17 Too CW , Fong KY , Hang G , et al. Artificial intelligence–guided segmentation and path planning software for transthoracic lung biopsy . J Vasc Interv Radiol . 2024 ; 35 ( 5 ): 780 ‐ 789.e1 . doi: 10.1016/j.jvir.2024.02.006 38355040

18 Darbandsari A , Farahani H , Asadi M , et al. AI‐based histopathology image analysis reveals a distinct subset of endometrial cancers . Nat Commun . 2024 ; 15 ( 1 ): 4973 . doi: 10.1038/s41467-024-49017-2 38926357 PMC11208496

19 Abbas SF , Vuong TTL , Kim K , Song B , Kwak JT . Multi‐cell type and multi‐level graph aggregation network for cancer grading in pathology images . Med Image Anal . 2023 ; 90 : 102936 . doi: 10.1016/j.media.2023.102936 37660482

20 Li L , Sun M , Wang J , Wan S . Chapter Nine—Multi‐omics based artificial intelligence for cancer research . In: Madan E , Fisher PB , Gogna R , eds. Advances in Cancer Research . Academic Press ; 2024 : 303 ‐ 356 . 10.1016/bs.acr.2024.06.005 PMC12701827 39271266

21 Mittermaier M , Raza MM , Kvedar JC . Bias in AI‐based models for medical applications: challenges and mitigation strategies . NPJ Digit Med . 2023 ; 6 ( 1 ): 113 . doi: 10.1038/s41746-023-00858-z 37311802 PMC10264403

22 El Mestari SZ , Lenzini G , Demirci H . Preserving data privacy in machine learning systems . Comput Secur . 2024 ; 137 : 103605 . doi: 10.1016/j.cose.2023.103605

23 Li Z , Jiang Y , Liu L , Xia Y , Li R . Single‐cell spatial analysis of histopathology images for survival prediction via graph attention network. In: Wu S, Shabestari B, Xing L, eds. Applications of Medical Artificial Intelligence. AMAI 2023. Lecture Notes in Computer Science, Vol. 14313. Springer, Cham ; 2024 : 114 ‐ 124 . doi: 10.1007/978-3-031-47076-9_12

24 Ochi M , Komura D , Onoyama T , et al. Registered multi‐device/staining histology image dataset for domain‐agnostic machine learning models . Sci Data . 2024 ; 11 ( 1 ): 330 . doi: 10.1038/s41597-024-03122-5 38570515 PMC10991301

25 Campanella G , Hanna MG , Geneslaw L , et al. Clinical‐grade computational pathology using weakly supervised deep learning on whole slide images . Nat Med . 2019 ; 25 ( 8 ): 1301 ‐ 1309 . doi: 10.1038/s41591-019-0508-1 31308507 PMC7418463

26 Pantanowitz L , Quiroga‐Garza GM , Bien L , et al. An artificial intelligence algorithm for prostate cancer diagnosis in whole slide images of core needle biopsies: a blinded clinical validation and deployment study . Lancet Digit Health . 2020 ; 2 ( 8 ): e407 ‐ e416 . doi: 10.1016/S2589-7500(20)30159-X 33328045

27 Pati P , Karkampouna S , Bonollo F , et al. Accelerating histopathology workflows with generative AI‐based virtually multiplexed tumour profiling . Nat Mach Intell . 2024 ; 6 ( 9 ): 1077 ‐ 1093 . doi: 10.1038/s42256-024-00889-5 39309216 PMC11415301

28 Shmatko A , Ghaffari Laleh N , Gerstung M , Kather JN . Artificial intelligence in histopathology: enhancing cancer research and clinical oncology . Nat Cancer . 2022 ; 3 ( 9 ): 1026 ‐ 1038 . doi: 10.1038/s43018-022-00436-4 36138135

29 Auffenberg GB , Ghani KR , Ramani S , et al. askMUSIC: leveraging a clinical registry to develop a new machine learning model to inform patients of prostate cancer treatments chosen by similar men . Eur Urol . 2019 ; 75 ( 6 ): 901 ‐ 907 . doi: 10.1016/j.eururo.2018.09.050 30318331 PMC6459726

30 Park SH , Park HM , Baek KR , Ahn HM , Lee IY , Son GM . Artificial intelligence based real‐time microcirculation analysis system for laparoscopic colorectal surgery . World J Gastroenterol . 2020 ; 26 ( 44 ): 6945 ‐ 6962 . doi: 10.3748/wjb.v26.i44.6946 33311942 PMC7701947

31 Sharma V , Fadel A , Tollefson MK , et al. Artificial intelligence‐based assessment of preoperative body composition is associated with early complications after radical cystectomy . J Urol . 2025 ; 213 ( 2 ): 228 ‐ 237 . doi: 10.1097/JU.0000000000004292 39392812

32 Lyuksemburg V , Abou‐Hanna J , Marshall JS , et al. Virtual reality for preoperative planning in complex surgical oncology: a single‐center experience . J Surg Res . 2023 ; 291 : 546 ‐ 556 . doi: 10.1016/j.jss.2023.07.001 37540972

33 Bohoudi O , Bruynzeel AME , Senan S , et al. Fast and robust online adaptive planning in stereotactic MR‐guided adaptive radiation therapy (SMART) for pancreatic cancer . Radiother Oncol . 2017 ; 125 ( 3 ): 439 ‐ 444 . doi: 10.1016/j.radonc.2017.07.028 28811038

34 Rajendran P , Chen Y , Qiu L , et al. Autodelineation of treatment target volume for radiation therapy using large language model‐aided multimodal learning . Int J Radiat Oncol Biol Phys . 2025 ; 121 ( 1 ): 230 ‐ 240 . doi: 10.1016/j.ijrobp.2024.07.2149 39117164 PMC13041240

35 Lippenszky L , Mittendorf KF , Kiss Z , et al. Prediction of effectiveness and toxicities of immune checkpoint inhibitors using real‐world patient data . JCO Clin Cancer Inform . 2024 ; 8 ( 8 ): e2300207 . doi: 10.1200/CCI.23.00207 38427922 PMC10919473

36 Jeon J , Nim S , Teyra J , et al. A systematic approach to identify novel cancer drug targets using machine learning, inhibitor design and high‐throughput screening . Genome Med . 2014 ; 6 ( 7 ): 57 . doi: 10.1186/s13073-014-0057-7 25165489 PMC4143549

37 Prasad K , Miller A , Sharif K , et al. Augmented‐reality surgery to guide head and neck cancer re‐resection: a feasibility and accuracy study . Ann Surg Oncol . 2023 ; 30 ( 8 ): 4994 ‐ 5000 . doi: 10.1245/s10434-023-13532-1 37133570 PMC11563582

38 Aweeda M , Adegboye F , Yang SF , Topf MC . Enhancing surgical vision: augmented reality in otolaryngology—head and neck surgery . J Med Ext Real . 2024 ; 1 ( 1 ): 124 ‐ 136 . doi: 10.1089/jmxr.2024.0010 39091667 PMC11290041

39 Moglia A , Morelli L , D’Ischia R , et al. Ensemble deep learning for the prediction of proficiency at a virtual simulator for robot‐assisted surgery . Surg Endosc . 2022 ; 36 ( 9 ): 6473 ‐ 6479 . doi: 10.1007/s00464-021-08999-6 35020053 PMC9402513

40 Torkamani A , Wineinger NE , Topol EJ . The personal and clinical utility of polygenic risk scores . Nat Rev Genet . 2018 ; 19 ( 9 ): 581 ‐ 590 . doi: 10.1038/s41576-018-0018-x 29789686

41 Riniker S , Wang Y , Jenkins JL , Landrum GA . Using information from historical high‐throughput screens to predict active compounds . J Chem Inform Model . 2014 ; 54 ( 7 ): 1880 ‐ 1891 . doi: 10.1021/ci500190p 24933016

42 Yagi R , Goto S , Himeno Y , et al. Artificial intelligence‐enabled prediction of chemotherapy‐induced cardiotoxicity from baseline electrocardiograms . Nat Commun . 2024 ; 15 ( 1 ): 2536 . doi: 10.1038/s41467-024-45733-x 38514629 PMC10957877

43 Satheeskumar R . AI‐driven diagnostics and personalized treatment planning in oral oncology: innovations and future directions . Oral Oncol Rep . 2025 ; 13 : 100704 . doi: 10.1016/j.oor.2024.100704

44 Lee KS , Jang JY , Yu YD , et al. Usefulness of artificial intelligence for predicting recurrence following surgery for pancreatic cancer: retrospective cohort study . Int J Surg . 2021 ; 93 : 106050 . doi: 10.1016/j.ijsu.2021.106050 34388677

45 Ito K , Suzuki Y , Sakaguchi T , et al. Physical activity using a wearable device as an alternative to performance status in patients with advanced lung cancer . JAMA Oncol . 2024 ; 10 ( 5 ): 648 ‐ 651 . doi: 10.1001/jamaoncol.2024.0023 38546663 PMC10979353

46 Nunez JJ , Leung B , Ho C , Ng RT , Bates AT . Predicting which patients with cancer will see a psychiatrist or counsellor from their initial oncology consultation document using natural language processing . Commun Med (Lond) . 2024 ; 4 ( 1 ): 69 . doi: 10.1038/s43856-024-00495-x 38589545 PMC11001970

47 Bychkov D , Linder N , Turkki R , et al. Deep learning based tissue analysis predicts outcome in colorectal cancer . Sci Rep . 2018 ; 8 ( 1 ): 3395 . doi: 10.1038/s41598-018-21758-3 29467373 PMC5821847

48 Manz CR , Chen J , Liu M , et al. Validation of a machine learning algorithm to predict 180‐day mortality for outpatients with cancer . JAMA Oncol . 2020 ; 6 ( 11 ): 1723 ‐ 1730 . doi: 10.1001/jamaoncol.2020.4331 32970131 PMC7516810

49 Manz CR , Parikh RB , Small DS , et al. Effect of integrating machine learning mortality estimates with behavioral nudges to clinicians on serious illness conversations among patients with cancer: a stepped‐wedge cluster randomized clinical trial . JAMA Oncol . 2020 ; 6 ( 12 ): e204759 . doi: 10.1001/jamaoncol.2020.4759 33057696 PMC7563672

50 Gajra A , Zettler ME , Miller KA , et al. Augmented intelligence to predict 30‐day mortality in patients with cancer . Future Oncol . 2021 ; 17 ( 29 ): 3797 ‐ 3807 . doi: 10.2217/fon-2021-0302 34189965

51 Gajra A , Zettler ME , Miller KA , et al. Impact of augmented intelligence on utilization of palliative care services in a real‐world oncology setting . JCO Oncol Pract . 2021 ; 18 ( 1 ): e80 ‐ e88 . doi: 10.1200/OP.21.00179 34506215 PMC8758123

52 Tran TTV , Surya Wibowo A , Tayara H , Chong KT . Artificial intelligence in drug toxicity prediction: recent advances, challenges, and future perspectives . J Chem Inform Model . 2023 ; 63 ( 9 ): 2628 ‐ 2643 . doi: 10.1021/acs.jcim.3c00200 37125780

53 Bitterman DS , Downing A , Maués J , Lustberg M . Promise and perils of large language models for cancer survivorship and supportive care . J Clin Oncol . 2024 ; 42 ( 14 ): 1607 ‐ 1611 . doi: 10.1200/JCO.23.02439 38452323 PMC11095890

54 Leung YW , Ng S , Duan L , et al. Therapist Feedback and implications on adoption of an artificial intelligence–based co‐facilitator for online cancer support groups: mixed methods single‐arm usability study . JMIR Cancer . 2023 ; 9 : e40113 . doi: 10.2196/40113 37294610 PMC10334721

55 Jindal S , Jindal M , Bhati P . Artificial intelligence in cancer survivorship care plans: what lies beyond diagnostics? IEEE Xplore . 2023 ; 17 ; 1 ‐ 7 . Accessed February 18, 2025. https://ieeexplore.ieee.org/document/10065173/citations?tabFilter=papers#citations

56 Amann J , Blasimme A , Vayena E , Frey D , Madai VI ; Precise4Q Consortium . Explainability for artificial intelligence in healthcare: a multidisciplinary perspective . BMC Med Inform Decis Mak . 2020 ; 20 ( 1 ): 310 . doi: 10.1186/s12911-020-01332-6 33256715 PMC7706019

57 Haver HL , Lin CT , Sirajuddin A , Yi PH , Jeudy J . Use of ChatGPT, GPT‐4, and Bard to improve readability of ChatGPT's answers to common questions about lung cancer and lung cancer screening . Am J Roentgenol . 2023 ; 221 ( 5 ): 701 ‐ 704 . doi: 10.2214/AJR.23.29622 37341179

58 Chung EM , Zhang SC , Nguyen AT , Atkins KM , Sandler HM , Kamrava M . Feasibility and acceptability of ChatGPT generated radiology report summaries for cancer patients . Digit Health . 2023 ; 9 : 20552076231221620 . doi: 10.1177/20552076231221620 38130802 PMC10734360

59 Kruse M , Hu S , Derby N , et al. Zero‐shot large language models for long clinical text summarization with temporal reasoning [preprint]. arXiv. Published online January 30, 2025 . doi: 10.48550/arXiv.2501.18724

60 Rubinstein S , Mohsin A , Banerjee R , et al. Summarizing clinical evidence utilizing large language models for cancer treatments: a blinded comparative analysis . Front Digit Health . 2025 ; 7 : 1569554 . doi: 10.3389/fdgth.2025.1569554 40364850 PMC12069342

61 Chen S , Kann BH , Foote MB , et al. Use of artificial intelligence chatbots for cancer treatment information . JAMA Oncol . 2023 ; 9 ( 10 ): 1459 ‐ 1462 . doi: 10.1001/jamaoncol.2023.2954 37615976 PMC10450584

62 Sorin V , Klang E , Sklair‐Levy M , et al. Large language model (ChatGPT) as a support tool for breast tumor board . NPJ Breast Cancer . 2023 ; 9 ( 1 ): 44 . doi: 10.1038/s41523-023-00557-8 37253791 PMC10229606

63 Lukac S , Dayan D , Fink V , et al. Evaluating ChatGPT as an adjunct for the multidisciplinary tumor board decision‐making in primary breast cancer cases . Arch Gynecol Obstet . 2023 ; 308 ( 6 ): 1831 ‐ 1844 . doi: 10.1007/s00404-023-07130-5 37458761 PMC10579162

64 Griewing S , Gremke N , Wagner U , Lingenfelder M , Kuhn S , Boekhoff J . Challenging ChatGPT 3.5 in senology—an assessment of concordance with breast cancer tumor board decision making . J Pers Med . 2023 ; 13 ( 10 ): 1502 . doi: 10.3390/jpm13101502 37888113 PMC10608120

65 Benary M , Wang XD , Schmidt M , et al. Leveraging large language models for decision support in personalized oncology . JAMA Netw Open . 2023 ; 6 ( 11 ): e2343689 . doi: 10.1001/jamanetworkopen.2023.43689 37976064 PMC10656647

66 Pan A , Musheyev D , Bockelman D , Loeb S , Kabarriti AE . Assessment of artificial intelligence chatbot responses to top searched queries about cancer . JAMA Oncol . 2023 ; 9 ( 10 ): 1437 ‐ 1440 . doi: 10.1001/jamaoncol.2023.2947 37615960 PMC10450581

67 Nguyen D , Swanson D , Newbury A , Kim YH . Evaluation of ChatGPT and Google Bard using prompt engineering in cancer screening algorithms . Acad Radiol . 2024 ; 31 ( 5 ): 1799 ‐ 1804 . doi: 10.1016/j.acra.2023.11.002 38103973

68 Hermann CE , Patel JM , Boyd L , Growdon WB , Aviki E , Stasenko M . Let's chat about cervical cancer: assessing the accuracy of ChatGPT responses to cervical cancer questions . Gynecol Oncol . 2023 ; 179 : 164 ‐ 168 . doi: 10.1016/j.ygyno.2023.11.008 37988948

69 Coskun B , Ocakoglu G , Yetemen M , Kaygisiz O . Can ChatGPT, an artificial intelligence language model, provide accurate and high‐quality patient information on prostate cancer? Urology . 2023 ; 180 : 35 ‐ 58 . doi: 10.1016/j.urology.2023.05.040 37406864

70 Lombardo R , Gallo G , Stira J , et al. Quality of information and appropriateness of Open AI outputs for prostate cancer . Prostate Cancer Prostatic Dis . 2025 ; 28 ( 1 ): 229 ‐ 231 . doi: 10.1038/s41391-024-00789-0 38228809

71 Ozgor F , Caglar U , Halis A , et al. Urological cancers and ChatGPT: assessing the quality of information and possible risks for patients . Clin Genitourin Cancer . 2024 ; 22 ( 2 ): 454 ‐ 457.e4 . doi: 10.1016/j.clgc.2023.12.017 38246831

72 Gebrael G , Sahu KK , Chigarira B , et al. Enhancing triage efficiency and accuracy in emergency rooms for patients with metastatic prostate cancer: a retrospective analysis of artificial intelligence‐assisted triage using ChatGPT 4.0 . Cancers (Basel . 2023 ; 15 ( 14 ): 3717 . doi: 10.3390/cancers15143717 37509379 PMC10378202

73 Musheyev D , Pan A , Loeb S , Kabarriti AE . How well do artificial intelligence chatbots respond to the top search queries about urological malignancies? Eur Urol . 2024 ; 85 ( 1 ): 13 ‐ 16 . doi: 10.1016/j.eururo.2023.07.004 37567827

74 Davis RJ , Ayo‐Ajibola O , Lin ME , et al. Evaluation of oropharyngeal cancer information from revolutionary artificial intelligence chatbot . Laryngoscope . 2024 ; 134 ( 5 ): 2252 ‐ 2257 . doi: 10.1002/lary.31191 37983846

75 Marchi F , Bellini E , Iandelli A , Sampieri C , Peretti G . Exploring the landscape of AI‐assisted decision‐making in head and neck cancer treatment: a comparative analysis of NCCN guidelines and ChatGPT responses . Eur Arch Otorhinolaryngol . 2024 ; 281 ( 4 ): 2123 ‐ 2136 . doi: 10.1007/s00405-024-08525-z 38421392

76 Lechien JR , Chiesa‐Estomba CM , Baudouin R , Hans S . Accuracy of ChatGPT in head and neck oncological board decisions: preliminary findings . Eur Arch Otorhinolaryngol . 2024 ; 281 ( 4 ): 2105 ‐ 2114 . doi: 10.1007/s00405-023-08326-w 37991498

77 Kuşcu O , Pamuk AE , Sütay Süslü N , Hosal S . Is ChatGPT accurate and reliable in answering questions regarding head and neck cancer? Front Oncol . 2023 ; 13 : 1256459 . doi: 10.3389/fonc.2023.1256459 38107064 PMC10722294

78 Wei K , Fritz C , Rajasekaran K . Answering head and neck cancer questions: an assessment of ChatGPT responses . Am J Otolaryngol . 2024 ; 45 ( 1 ): 104085 . doi: 10.1016/j.amjoto.2023.104085 37844413

79 Lee JC , Hamill CS , Shnayder Y , Buczek E , Kakarala K , Bur AM . Exploring the role of artificial intelligence chatbots in preoperative counseling for head and neck cancer surgery . Laryngoscope . 2024 ; 134 ( 6 ): 2757 ‐ 2761 . doi: 10.1002/lary.31243 38126511 PMC12374230

80 Rahsepar AA , Tavakoli N , Kim GHJ , Hassani C , Abtin F , Bedayat A . How AI responds to common lung cancer questions: ChatGPT versus Google Bard . Radiology . 2023 ; 307 ( 5 ): e230922 . doi: 10.1148/radiol.230922 37310252

81 Johnson SB , King AJ , Warner EL , Aneja S , Kann BH , Bylund CL . Using ChatGPT to evaluate cancer myths and misconceptions: artificial intelligence and cancer information . JNCI Cancer Spectr . 2023 ; 7 ( 2 ): pkad015 . doi: 10.1093/jncics/pkad015 36929393 PMC10020140

82 Holmes J , Liu Z , Zhang L , et al. Evaluating large language models on a highly‐specialized topic, radiation oncology physics . Front Oncol . 2023 ; 13 : 1219326 . doi: 10.3389/fonc.2023.1219326 37529688 PMC10388568

83 Huang Y , Gomaa A , Semrau S , et al. Benchmarking ChatGPT‐4 on a radiation oncology in‐training exam and Red Journal Gray Zone cases: potentials and challenges for ai‐assisted medical education and decision making in radiation oncology . Front Oncol . 2023 ; 13 : 1265024 . doi: 10.3389/fonc.2023.1265024 37790756 PMC10543650

84 Dennstädt F , Hastings J , Putora PM , et al. Exploring capabilities of large language models such as ChatGPT in radiation oncology . Adv Radiat Oncol . 2024 ; 9 ( 3 ): 101400 . doi: 10.1016/j.adro.2023.101400 38304112 PMC10831180

85 Stults CD , Deng S , Martinez MC , et al. Evaluation of an ambient artificial intelligence documentation platform for clinicians . JAMA Netw Open . 2025 ; 8 ( 5 ): e258614 . doi: 10.1001/jamanetworkopen.2025.8614 40314951 PMC12048851

86 Microsoft . Automatically document care with DAX™ Copilot . Microsoft; 2025 . Accessed February 18, 2025. https://www.nuance.com/asset/en_us/collateral/healthcare/data‐sheet/ds‐ambient‐clinical‐intelligence‐en‐us.pdf

87 Albrecht M , Shanks D , Shah T , et al. Enhancing clinical documentation with ambient artificial intelligence: a quality improvement survey assessing clinician perspectives on work burden, burnout, and job satisfaction . JAMA Open . 2025 ; 8 ( 1 ): ooaf013 . doi: 10.1093/jamiaopen/ooaf013 PMC11843214 39991073

88 ABRIDGE . Company website . Accessed February 18, 2025. https://www.abridge.com/customers#More

89 Memorial Sloan Kettering Cancer Center . MSK Innovation Hub . Memorial Sloan Kettering Cancer Center ; 2025 . Accessed February 18, 2025. https://www.mskcc.org/commercialization/programs‐accelerators/msk‐innovation‐hub

90 LeanTaas . iQueue for Infusion Centers . LeanTaas ; 2022 . Accessed February 2, 2025. https://leantaas.com/products/infusion‐centers/

91 Peterson Health Technology Institute . Adoption of Artificial Intelligence in Healthcare Delivery Systems: Early Applications and Impacts. Report from the Peterson Health Technology Institute AI Taskforce . Peterson Health Technology Institute ; 2025 . Accessed February 18, 2025. https://phti.org/wp‐content/uploads/sites/3/2025/03/PHTI‐Adoption‐of‐AI‐in‐Healthcare‐Delivery‐Systems‐Early‐Applications‐Impacts.pdf

92 Bookner M . Clinical AI: enhancing documentation accuracy with ambient technology . IMO Health ; November 21, 2024 . Accessed February 18, 2025. https://www.imohealth.com/resources/clinical‐ai‐enhancing‐documentation‐accuracy‐with‐ambient‐technology/

93 Jorie AI . How AI‐Driven Charge Capture Boosts Healthcare Revenue Integrity . Jorie AI ; 2025 . Accessed February 18, 2025. https://www.jorie.ai/post/how‐ai‐driven‐charge‐capture‐boosts‐healthcare‐revenue‐integrity

94 Emamekhoo H , Carroll CB , Stietz C , et al. Supporting structured data capture for patients with cancer: an initiative of the University of Wisconsin Carbone Cancer Center Survivorship Program to improve capture of malignant diagnosis and cancer staging data . JCO Clin Cancer Inform . 2022 ; 6 ( 6 ): e2200020 . doi: 10.1200/CCI.22.00020 35802837 PMC9296185

95 Hesse BW , Suls JM . Informatics‐enabled behavioral medicine in oncology . Cancer J . 2011 ; 17 ( 4 ): 222 ‐ 230 . doi: 10.1097/ppo.0b013e318227c811 21799329 PMC3148812

96 Wei Y , Guo X . Revolutionizing clinical experimental protocol design through the ChatGPT technology . Am J Med Sci . 2023 ; 366 ( 6 ): 468 ‐ 470 . doi: 10.1016/j.amjms.2023.09.004 37699445

97 Beattie J , Neufeld S , Yang DX , et al. Using large language models to create patient centered consent forms . Int J Radiat Oncol Biol Phys . 2024 ; 120 ( 2 ): e612 . doi: 10.1016/j.ijrobp.2024.07.1346

98 Shi Q , Luzuriaga K , Allison JJ , et al. Transforming informed consent generation using large language models: mixed methods study . JMIR Med Inform . 2025 ; 13 : e68139 . doi: 10.2196/68139 39946400 PMC11841745

99 Oikonomou EK , Thangaraj PM , Bhatt DL , et al. An explainable machine learning‐based phenomapping strategy for adaptive predictive enrichment in randomized clinical trials . NPJ Digit Med . 2023 ; 6 ( 1 ): 217 . doi: 10.1038/s41746-023-00963-z 38001154 PMC10673945

100 Cerami E , Trukhanov P , Paul MA , et al. MatchMiner‐AI: an open‐source solution for cancer clinical trial matching [preprint]. ArXiv. Published online December 23, 2024 . doi: 10.48550/arXiv.2412.17288

101 Eckardt JN , Hahn W , Röllig C , et al. Mimicking clinical trials with synthetic acute myeloid leukemia patients using generative artificial intelligence . NPJ Digit Med . 2024 ; 7 ( 1 ): 76 . doi: 10.1038/s41746-024-01076-x 38509224 PMC10954666

102 Reason T , Benbow E , Langham J , Gimblett A , Klijn SL , Malcolm B . Artificial intelligence to automate network meta‐analyses: four case studies to evaluate the potential application of large language models . Pharmacoecon Open . 2024 ; 8 ( 2 ): 205 ‐ 220 . doi: 10.1007/s41669-024-00476-9 38340277 PMC10884375

103 Tsafnat G , Glasziou P , Choong MK , Dunn A , Galgani F , Coiera E . Systematic review automation technologies . Syst Rev . 2014 ; 3 ( 1 ): 74 . doi: 10.1186/2046-4053-3-74 25005128 PMC4100748

104 Cao C , Sang J , Arora R , et al. Development of prompt templates for large language model–driven screening in systematic reviews . Ann Intern Med . 2025 ; 178 ( 3 ): 389 ‐ 401 . doi: 10.7326/ANNALS-24-02189 39993313

105 Sanghera R , Thirunavukarasu AJ , El Khoury M , et al. High‐performance automated abstract screening with large language model ensembles . J Am Med Inform Assoc . 2025 ; 32 ( 5 ): 893 ‐ 904 . doi: 10.1093/jamia/ocaf050 40119675 PMC12012331

106 Lai H , Liu J , Bai C , et al. Language models for data extraction and risk of bias assessment in complementary medicine . NPJ Digit Med . 2025 ; 8 ( 1 ): 74 . doi: 10.1038/s41746-025-01457-w 39890970 PMC11785717

107 Khan MA , Ayub U , Naqvi SAA , et al. Collaborative large language models for automated data extraction in living systematic reviews . J Am Med Inform Assoc . 2025 ; 32 ( 4 ): 638 ‐ 647 . doi: 10.1093/jamia/ocae325 39836495 PMC12005628

108 Krepper D , Cesari M , Hubel NJ , Zelger P , Sztankay MJ . Machine learning models including patient‐reported outcome data in oncology: a systematic literature review and analysis of their reporting quality . J Patient Rep Outcomes . 2024 ; 8 ( 1 ): 126 . doi: 10.1186/s41687-024-00808-7 39499409 PMC11538124

109 Sim JA , Xiaolei H , Horan MR , Baker JN , Huang IC . Using natural language processing to analyze unstructured patient‐reported outcomes data derived from electronic health records for cancer populations: a systematic review . Expert Rev Pharmacoecon Outcomes Res . 2024 ; 24 ( 4 ): 467 ‐ 475 . doi: 10.1080/14737167.2024.2322664 38383308 PMC11001514

110 Tabataba Vakili S , Haywood D , Kirk D , et al. Application of artificial intelligence in symptom monitoring in adult cancer survivorship: a systematic review . JCO Clin Cancer Inform . 2024 ; 8 ( 8 ): e2400119 . doi: 10.1200/CCI.24.00119 39621952

111 Lin N , Abbas‐Aghababazadeh F , Su J , et al. Development of machine learning models for predicting radiation dermatitis in breast cancer patients using clinical risk factors, patient‐reported outcomes, and serum cytokine biomarkers . Clin Breast Cancer . 2025 ; 25 ( 5 ): e622 ‐ e634.e6 . doi: 10.1016/j.clbc.2025.03.002 40155248

112 Goryachev SD , Yildirim C , DuMontier C , et al. Natural language processing algorithm to extract multiple myeloma stage from oncology notes in the Veterans Affairs Healthcare System . JCO Clin Cancer Inform . 2024 ; 8 ( 8 ): e2300197 . doi: 10.1200/CCI.23.00197 39038255 PMC11371094

113 Mauro J , Mueller M , Perrone E , et al. SLYMEC II study: overall survival analysis of the impact of LVSI in apparent early stage endometrioid endometrial cancer . Eur J Surg Oncol . 2025 ; 51 ( 7 ): 109750 . doi: 10.1016/j.ejso.2025.109750 40163975

114 Araújo ALD , Moraes MC , Pérez‐de‐Oliveira ME , et al. Machine learning for the prediction of toxicities from head and neck cancer treatment: a systematic review with meta‐analysis . Oral Oncol . 2023 ; 140 : 106386 . doi: 10.1016/j.oraloncology.2023.106386 37023561

115 Xiao H , Zhou F , Liu X , et al. A comprehensive survey of large language models and multimodal large language models in medicine . Inf Fusion . 2024 ; 117 : 102888 . doi: 10.1016/j.inffus.2024.102888

116 Jung J , Kim H , Bae S , Park JY . Comparative analysis of multimodal large language models GPT‐4o and o1 vs clinicians in clinical case challenge questions [preprint]. medRxiv. Published online June 23, 2025 . doi: 10.1101/2025.06.22.25330068 PMC12851745 41578521

117 Mollica L , Leli C , Sottotetti F , Quaglini S , Locati LD , Marceglia S . Digital twins: a new paradigm in oncology in the era of big data . ESMO Real World Data Dig Oncol . 2024 ; 5 : 100056 . doi: 10.1016/j.esmorw.2024.100056 PMC12836754 41648658

118 Shukla S , Rajkumar S , Sinha A , Esha M , Elango K , Sampath V . Federated learning with differential privacy for breast cancer diagnosis enabling secure data sharing and model integrity . Sci Rep . 2025 ; 15 ( 1 ): 13061 . doi: 10.1038/s41598-025-95858-2 40240790 PMC12003885

119 Azenkot T , Rivera DR , Stewart MD , Patel SP . Artificial intelligence and machine learning innovations to improve design and representativeness in oncology clinical trials . Am Soc Clin Oncol Edu Book . 2025 ; 45 ( 3 ): e473590 . doi: 10.1200/EDBK-25-473590 40403202

120 Bryant AK , Zamora‐Resendiz R , Dai X , et al. Artificial intelligence to unlock real‐world evidence in clinical oncology: a primer on recent advances . Cancer Med . 2024 ; 13 ( 12 ): e7253 . doi: 10.1002/cam4.7253 38899720 PMC11187737

121 Singh Y Fau‐Patel H , Patel H , et al. Beyond the hype: navigating bias in AI‐driven cancer detection . Oncotarget . 2024 ; 15 : 764 ‐ 766 . doi: 10.18632/oncotarget.28665 39513852 PMC11546210

122 Schubert T , Oosterlinck T , Stevens RD , Maxwell PH , van der Schaar M . AI education for clinicians . eClinicalMedicine . 2025 ; 79 : 102968 . doi: 10.1016/j.eclinm.2024.102968 39720600 PMC11667627

123 Fountzilas E , Pearce T , Baysal MA , Chakraborty A , Tsimberidou AM . Convergence of evolving artificial intelligence and machine learning techniques in precision oncology . NPJ Digit Med . 2025 ; 8 ( 1 ): 75 . doi: 10.1038/s41746-025-01471-y 39890986 PMC11785769

124 Ghassemi M , Gusev A . Limiting bias in AI models for improved and equitable cancer care . Nat Rev Cancer . 2024 ; 24 ( 12 ): 823 ‐ 824 . doi: 10.1038/s41568-024-00739-x 39191902