Recent Articles

For patients with cancer, the pathway to diagnosis will most often begin in general practice. But delays in diagnosis can occur in the absence of strong diagnostic features or in patients with non-specific symptoms. Initial presentations and routine blood tests are important in determining whether a patient requires further investigation. Quality improvement (QI) interventions including auditing tools and clinical decision support (CDS), have been developed for use in general practice to support this diagnostic process. We conducted a process evaluation of a pragmatic, cluster randomised trial which evaluated the effectiveness of a new technology, Future Health Today (FHT), implemented in general practice to assist with the appropriate follow-up of patients at risk of an undiagnosed cancer.

Commonly used digital health technologies, such as electronic health record systems and patient portals as well as custom-built digital decision aids, have the potential to enhance person-centered shared decision-making (SDM) in cancer care. SDM is a 2-way exchange of information between at least a clinician and the patient and a shared commitment to make informed decisions. However, there is little evidence in the literature on how technologies are used for SDM or how best they can be designed and integrated into workflows and practice. This may be due to the nature of SDM, which is fundamentally human interactions and conversations that produce desired human outcomes. Therefore, technology must be nonintrusive while supporting the human decision-making process.


The increasing demand for population-wide genomic screening (PGS) and the limited availability of genetic counseling resources have created a pressing need for innovative service delivery models. Chatbots powered by large language models (LLMs) have shown potential in genomic services, particularly in pre-test counseling, but their application in returning positive PGS results remains underexplored. Leveraging advanced LLMs like GPT-4 offers an opportunity to address this gap by delivering accurate, contextual, and user-centered communication to individuals receiving positive genetic test results. This project aimed to design, implement, and evaluate a chatbot integrated with GPT-4, tailored to support the return of positive genomic screening results in the context of South Carolina's In Our DNA SC program. This initiative offers free genetic screening to 100,000 individuals, with over 33,000 results returned and numerous positive findings for conditions such as Lynch syndrome, hereditary breast and ovarian cancer syndrome, and familial hypercholesterolemia. A three-step prompt engineering process using Retrieval-Augmented Generation (RAG) and few-shot techniques was employed to create the chatbot. Training materials included patient frequently asked questions, genetic counseling scripts, and patient-derived queries. The chatbot underwent iterative refinement based on 13 training questions, while performance was evaluated through expert ratings on responses to two hypothetical patient scenarios. The two scenarios were intended to represent common but distinct patient profiles in terms of gender, race, ethnicity, age, and background knowledge. Domain experts rated the chatbot using a 5-point Likert scale across eight predefined criteria: tone, clarity, program accuracy, domain accuracy, robustness, efficiency, boundaries, and usability. The chatbot achieved an average score of 3.88 across all evaluation metrics. The highest-rated criteria were tone (4.25) and usability (4.25), reflecting the chatbot’s ability to communicate effectively and provide a seamless user experience. Boundary management (4.0) and efficiency (3.88) also scored well, while clarity and robustness received ratings of 3.81. Domain accuracy was rated 3.63, indicating satisfactory performance in delivering genetic information, whereas program accuracy received the lowest score of 3.25, highlighting the need for improvements in delivering program-specific details. This project demonstrates the feasibility of using LLM-powered chatbots to support the return of positive genomic screening results. The chatbot effectively handled open-ended patient queries, maintained conversational boundaries, and delivered user-friendly responses. However, enhancements in program-specific accuracy are essential to maximize its utility. Future research will explore hybrid chatbot designs that combine the strengths of LLMs with rule-based components to improve scalability, accuracy, and accessibility in genomic service delivery. The findings underscore the potential of generative AI tools to address resource limitations and improve the accessibility of genomic healthcare services.

Head and neck cancer (HNC) survivors face challenging treatment consequences that can lead to severe disruptions in swallowing and result in weight loss, malnutrition and feeding tube dependence. HNC caregivers (family/friends who provide support) therefore often encounter distressing nutritional caregiving burdens and feel unprepared to provide adequate support at home.

Cancer imposes significant physical and emotional distress not only on patients but also on their caregivers. In recent years, there has been a growing focus on the mental and physical well-being of caregivers. Among various psychological interventions, cognitive behavioral therapy (CBT) is widely recognized as one of the most effective approaches. However, traditional CBT is often limited by time and geographical constraints, resulting in delayed or inefficient support for caregivers. Internet-based cognitive behavioral therapy (ICBT) presents a valuable alternative for alleviating the caregiving burden and the negative emotions experienced by caregivers.

Despite its potential to predict and detect early cancer risks, genetic testing remains underutilized by the public. This study, guided by the Health Belief Model, examined key factors influencing an individual’s willingness to undergo genetic testing for cancer, with a particular focus on gender, caregiver status, and participation in online social support groups.

Due to multifaceted outpatient regiments, children receiving hematopoietic stem cell transplant (HCT) are at high risk of medication non-adherence, leading to life-threatening complications. mHealth interventions have proven effective in improving adherence in various pediatric conditions, however adherence intervention literature on HCT is limited.



Generative AI Chatbots may be useful tools for supporting shared prostate cancer screening decisions, but the information produced by these tools sometimes lack quality or credibility. ‘Prostate Cancer Info’ is a custom GPT chatbot developed to provide plain-language PrCA information only from websites of key authorities on cancer and peer-reviewed literature.

Progression-free survival (PFS) is a crucial endpoint in cancer drug research. The clinician-confirmed cancer progression, namely real-world PFS (rwPFS) in unstructured text (i.e. clinical notes) has been shown to serve as a reasonable surrogate for real-world indicators in ascertaining progression endpoints. Response Evaluation Criteria in Solid Tumors(RECIST) is traditionally used in clinical trials using serial imaging evaluations, which is not practical when working with real-world data. Manual abstraction of clinical progression from unstructured notes continues to be the gold standard. However, this process is a resource-intensive and time-consuming process. Natural Language processing(NLP), a subdomain of machine learning, has shown promise in accelerating the extraction of tumor progression from real world data in recent years.