Artificial Intelligence Techniques That May Be Applied to Primary Care Data to Facilitate Earlier Diagnosis of Cancer: Systematic Review (Preprint)

Owain T Jones* (Corresponding Author), Natalia Calanzani, Smiji Saji, Stephen W Duffy, Jon Emery, Willie Hamilton, Hardeep Singh, Niek J de Wit, Fiona M Walter

*Corresponding author for this work

Research output: Other contribution



More than 17 million people worldwide, including 360,000 people in the UK, were diagnosed with cancer in 2018. Cancer prognosis and disease burden is highly dependent on disease stage at diagnosis. Most people diagnosed with cancer first present in primary care settings, where improved assessment of the (often vague) presenting symptoms of cancer could lead to earlier detection, and improved outcomes for patients. There is accumulating evidence that artificial intelligence (AI) can assist clinicians in making better clinical decisions in some areas of healthcare.


We aimed to systematically review AI technologies based on electronic health record (EHR) data that may facilitate the earlier diagnosis of cancer in primary care settings. We evaluated the quality of the evidence, the phase of development the AI technologies have reached, the gaps that exist in the evidence, and the potential for use in primary care.


We searched Medline, Embase, SCOPUS, and Web of Science databases from 1st January 2000 to 11th June 2019 (PROSPERO ID CRD42020176674), and included all studies providing evidence for accuracy or effectiveness of applying AI technologies to early detection of cancer using electronic health records. We included all study designs, in all settings and all languages. We extended these searches through a scoping review of commercial AI technologies. The main outcomes assessed were measures of diagnostic accuracy for cancer.


We identified 10,456 studies: 16 met the inclusion criteria, representing the data of 3,862,910 patients. 13 studies described the initial development and testing of AI algorithms and three studies described the validation of an AI technology in independent datasets. One study was based on prospectively collected data; only three studies were based on primary care data. We found no data on implementation barriers or cost-effectiveness. Risk-of-bias assessment highlighted a wide range in study quality. The additional scoping review of commercial AI tools identified 21 technologies, only one meeting our inclusion criteria. Meta-analysis was not undertaken due to heterogeneity of AI modalities, dataset characteristics and outcome measures.


Applying AI technologies to electronic health records for early detection of cancer in primary care is at an early stage of maturity. Further evidence is needed on performance using primary care data, implementation barriers and cost-effectiveness before widespread adoption into routine primary care clinical practice can be recommended.

This study was supported by funding from the NIHR Cancer Policy Research Programme and Cancer Research UK.
Original languageEnglish
PublisherJMIR Publications Inc.
Number of pages56
Publication statusPublished - 3 Mar 2021

Bibliographical note

This research was commissioned and funded by the National Institute for Health Research (NIHR) Policy Research Programme, conducted through the Policy Research Unit in Cancer Awareness, Screening and Early Diagnosis, PR-PRU-1217-21601. The views expressed are those of the author(s) and not necessarily those of the NIHR, the Department of Health and Social Care. This work was also supported by the CanTest Collaborative (funded by Cancer Research UK C8640/A23385) of which FMW and WH are Directors, and JE, HS and NdW are Associate Directors. The views expressed in this publication are those of the authors and not necessarily those of the National Health Service, the NIHR or the Department of Health. The funding sources had no role in the study design, data collection, data analysis, data interpretation, writing of the report or in the decision to submit for publication. The authors would like to thank Isla Kuhn, Reader Services Librarian, University of Cambridge Medical Library, for her help in developing the search strategy.


Dive into the research topics of 'Artificial Intelligence Techniques That May Be Applied to Primary Care Data to Facilitate Earlier Diagnosis of Cancer: Systematic Review (Preprint)'. Together they form a unique fingerprint.

Cite this