‘C the Signs’ AI program enhances early cancer detection by analyzing GP records, leading to significant improvements in identifying various types of cancer

The implementation of artificial intelligence in general practice has markedly improved cancer detection rates in England. The “C the Signs” AI tool, which scans patient records to identify hidden patterns indicative of cancer, has raised the detection rate from 58.7% to 66.0% at practices where it is used. This tool gathers comprehensive patient information, including medical history, test results, prescriptions, and other personal characteristics such as age, postcode, and family history, to assess cancer risk.

“C the Signs” prompts general practitioners to inquire about new symptoms from patients and recommends specific tests or clinical pathways if it detect high-risk patterns. Approximately 1,400 practices, or about 15% of those in England, currently use the AI tool. It was initially tested in 35 practices in the East of England, covering a population of 420,000 patients, beginning in May 2021.

Results published in the Journal of Clinical Oncology indicate a notable increase in cancer detection rates at these practices, whereas practices not using the tool showed no significant change. Dr. Bea Bakshi, a GP who co-created the system with her colleague Miles Payling, emphasized that the AI not only facilitates earlier diagnosis but also ensures faster diagnosis.

The AI tracks patients through the system, reminding doctors to review test results and referrals. According to Bakshi, the system has detected over 50 different types of cancer. Validation studies revealed that out of 118,677 patients, the algorithm correctly identified 7,056 out of 7,295 diagnosed cancer cases.

Dr. Bakshi was inspired to develop this tool after a patient with a late diagnosis of pancreatic cancer passed away shortly after diagnosis. This event highlighted the need for earlier detection methods. Despite existing screening programs for bowel, breast, and cervical cancers, many types of cancer are not screened for and may present with vague or easily misattributed symptoms.

GPs use guidelines from the National Institute for Health and Care Excellence to decide when to refer patients for cancer testing, but these guidelines can be complex and not always comprehensive. Dr. Peter Holloway, a GP and co-author of the study, recounted a case where the AI tool recommended a faecal test for a patient with common symptoms, leading to an early diagnosis and successful treatment of colorectal cancer.

The NHS England Long Term Plan for Cancer aims to diagnose 75% of all cancers at stages one or two by 2028. Alongside AI tools like “C the Signs,” the NHS is researching the Galleri blood test, which aims to detect DNA from over 50 types of tumours.

Professor Peter Johnson, National Clinical Director for Cancer at NHS England, noted that despite the pressures on services, the use of advanced technology is crucial in reducing wait times and improving early cancer detection rates. Initiatives like ‘teledermatology’ for skin cancers, community lung trucks, and home testing for bowel cancer are also part of these efforts.

Analysis

Political: The integration of AI in healthcare, particularly in cancer detection, reflects broader political initiatives aimed at modernizing the NHS and improving patient outcomes. The successful implementation of “C the Signs” supports government policies promoting the use of advanced technology in healthcare. This development may influence future healthcare funding and policy decisions, emphasizing the need for continued investment in AI and other innovative solutions. Politically, this success could be leveraged by current leadership to showcase progress in healthcare reform and bolster public confidence in government health policies.

Social: The use of AI in cancer detection addresses significant societal concerns about the effectiveness and efficiency of healthcare systems. Early cancer detection not only saves lives but also reduces the emotional and financial burden on patients and their families. The success of “C the Signs” can lead to increased trust in AI-driven healthcare solutions, potentially changing societal perceptions of AI from scepticism to acceptance. Furthermore, it highlights the importance of preventive care and regular check-ups, encouraging individuals to engage more proactively with their health.

Racial: Healthcare disparities often affect marginalized racial groups, who may have less access to early diagnostic services and advanced healthcare technologies. The widespread adoption of AI tools like “C the Signs” can help bridge these gaps by providing more consistent and accurate cancer detection across diverse populations. Ensuring equitable access to such technologies is crucial in addressing racial disparities in healthcare outcomes. Efforts must be made to implement these tools in underrepresented and underserved communities to ensure that the benefits of AI in healthcare are universally experienced.

Gender: The detection and treatment of cancer can vary significantly between genders, with certain cancers being more prevalent in one gender over the other. AI tools can help identify gender-specific risk factors and symptoms, leading to more personalized and effective care. For instance, AI can enhance the detection of breast and cervical cancers in women and prostate cancer in men. By improving early detection rates for gender-specific cancers, AI tools contribute to better health outcomes and potentially lower mortality rates among both men and women.

Economic: The economic impact of improved cancer detection through AI is substantial. Early diagnosis reduces the costs associated with late-stage cancer treatments, hospitalizations, and prolonged patient care. By catching cancers earlier, the healthcare system can allocate resources more efficiently, potentially leading to cost savings. Additionally, healthier populations contribute to economic productivity, as individuals are able to maintain their work and personal responsibilities. The investment in AI technology for healthcare, therefore, presents a long-term economic benefit by enhancing overall public health and reducing treatment costs.