Two new advanced predictive algorithms have been developed to help GPs accurately predict if their patients have undiagnosed cancer, particularly liver and oral cancers. These tools, created by researchers from Queen Mary University of London and the University of Oxford, use anonymized electronic health records from over 7.4 million adults in England. By incorporating data like age, family history, medical diagnoses, symptoms, and general health, along with results from seven routine blood tests, these algorithms aim to enhance early cancer detection and improve clinical decision-making. The existing prediction algorithms used by the NHS, such as QCancer scores, are effective but less sensitive compared to the new models. The newly developed algorithms can identify additional medical conditions, family history, and symptoms linked to a higher risk of 15 different types of cancer, including those affecting the liver, kidneys, and pancreas. For instance, the new models pinpointed four more medical conditions associated with an increased risk of cancer, two additional family history associations with lung and blood cancers, and seven new symptoms, such as itching, bruising, back pain, hoarseness, flatulence, abdominal mass, and dark urine. The integration of blood test results as biomarkers is a significant advancement. These tests, which measure a person’s full blood count and liver function, are already part of routine healthcare, making the new algorithms both cost-effective and practical. This approach is unique in its ability to estimate the likelihood of having an undiagnosed liver cancer, filling a critical gap in current diagnostic practices. According to Professor Julia Hippisley-Cox, the lead author from Queen Mary University of London, these algorithms are designed to be integrated into clinical systems and utilized during routine GP consultations. They provide higher accuracy in recognizing cancers, especially at early stages, which are more treatable. This enhancement can help the NHS achieve its goal of improving early cancer diagnosis rates by 2028. Dr. Carol Coupland, co-author and senior researcher at Queen Mary University of London, noted that the algorithms can identify individuals at risk of 15 types of cancer based on their symptoms, blood test results, lifestyle factors, and medical records. This capability extends to people aged 18 and older, including those with rare cancers, offering the potential for earlier diagnoses and better outcomes. The development of these new algorithms represents a significant leap forward in cancer diagnostics, particularly in primary care settings. The ability to detect undiagnosed cancers using readily available patient data and routine blood tests could lead to more timely and effective treatment. Industry insiders praise these advancements for their potential to transform early cancer detection, highlighting the importance of continued research and the implementation of these tools in clinical practice. Queen Mary University of London and the University of Oxford are renowned for their contributions to medical and technological innovation, and this collaboration demonstrates their commitment to leveraging data and analytics to improve patient care.