By Tammy McCausland
Depending on who—or what tool—you ask, the definition of artificial intelligence (AI) and what it comprises can differ. Generally, machine learning, neural networks, deep learning and generative AI are technologies included under the general term “AI.” When OpenAI released its first version of ChatGPT (generative AI) to the public in November 2022, it garnered immediate interest, enthusiasm and concern. Competitor generative AI tools like Anthropic’s Claude.ai, Google’s NotebookLM and Perplexity.ai have since emerged. Their presence is pervasive across industries, including health care, and their impact indelible.
Though the hype may be quite recent—and ongoing—use of AI technologies in radiation oncology isn’t new. AI was already being used in auto-contouring and in locating tumors in medical imaging, for example, before generative AI.
In an academic article published in the journal Diagnostic and Interventional Imaging, Paul Giraud and Jean-Emmanuel Bibault discuss current applications and future trends of AI in radiation therapy.1 They explore how AI tools are being used to automate the radiotherapy treatment workflow and how deep learning models are used in CT imaging and MRI-based dosimetry planning. Giraud and Bibault also suggest AI may make adaptive radiotherapy less resource intensive and more accessible. They also caution that each new tool implemented requires care and close monitoring and that AI-based clinical decision support is still at the proof-of-concept stage.
Northwestern University published an article in May 2024 titled “Questions about your radiation therapy? Ask Chat GPT.” Marla Paul shares the findings of a study about ChatGPT’s responses to 115 common radiation oncology patients, which found that the responses were “on par or exceeded answers from professional societies in terms of accuracy, completeness and conciseness in a majority of cases.”2 The study authors noted a few issues—for example, ChatGPT’s answers were at a college reading level, it failed to mention tiny tattoos used to position patients and it omitted details about a specialized technology for brain tumors.2 Researchers are developing and testing an in-house app using gen AI to answer patients’ questions. In recent academic article published in Radiotherapy and Oncology, Ramadan et al. discuss the effectiveness of ChatGPT across 12 clinical scenarios.3 The researchers found that “AI performed best in the domains of epidemiology (93%) and cancer biology (93%) and reasonably in staging and workup (89%), physics (86%) and surveillance (82%). Weaker domains included treatment planning (78%) and clinical management (81%).”3
Doctors at Memorial Sloan Kettering Cancer Center (MSKCC) are using AI to improve radiation therapy treatments. In his article published in September 2024, Ian Demsky describes how Dr. Harini Veeraraghavan’s lab is developing AI models that can better isolate tumors and simultaneously spare surrounding healthy tissue.4 One example is a virtual digital twin that “would model how a patient’s abdominal organs move” so clinicians can improve treatment planning and the precision of delivery.4 Researchers are also developing AI methods that can track more precisely how much radiation a tumor or organ has received.4
Researchers in the United Kingdom are using AI to mimic cancer patient trials. In an article published by the BBC, Rob Bristow, the lead research professor was quoted as saying, “the simulated trials would help make research ‘faster, safe and less expensive’ than large-scale clinical studies.”5 “Groups of patients are created virtually based on real-data by an AI computer programme, with treatments or devices tested on them in simulated trials,” writes Ewan Gawne. “Researchers said this allowed for ideas to be tested at a faster rate than traditional trials which involve real patients.” Researchers received funding to investigate “patient-specific genetics and tumours, as well as comparing a new form of proton beam therapy with standard radiotherapy for lung cancer patients.”5
These are just a few examples of how AI technologies are being used and how they may reshape treatments for cancer patients. The technologies are evolving rapidly, so their impact and use will inevitably evolve over time.
References:
Share any helpful tips and suggestions.
We would love to hear your experiences.
Share your thoughts here, or login to SROA Connect and join the conversation. If you are not a member of SROA yet, learn more about joining the radiation oncology association serving the niche profession of Radiation Oncology Administrator.
Related Content:
SROA Blogs
Comments