Infection Inspection: Using the power of citizen science to help with image-based prediction of antibiotic resistance in Escherichia coli

Antibiotic resistance is an urgent global health challenge, necessitating rapid diagnostic tools to combat its escalating threat. This study introduces innovative approaches for expedited bacterial antimicrobial resistance profiling, addressing the critical need for swift clinical responses. Between February and April 2023, we conducted the Infection Inspection project, a citizen science initiative in which the public could participate in advancing an antimicrobial susceptibility testing method based on single-cell images of cellular phenotypes in response to ciprofloxacin exposure. A total of 5,273 users participated, classifying 1,045,199 images. Notably, aggregated user accuracy in image classification reached 66.8%, lower than our deep learning model's performance at 75.3%, but accuracy increased for both users and the model when ciprofloxacin treatment was greater than a strain's own minimum inhibitory concentration. We used the users' classifications to elucidate which visual features influence classification decisions, most importantly the degree of DNA compaction and heterogeneity. We paired our classification data with an image feature analysis which showed that most of the incorrect classifications were due to cellular features that varied from the expected response. This understanding informs ongoing efforts to enhance the robustness of our deep learning-based bacterial classifier and diagnostic methodology. Our successful engagement with the public through citizen science is another demonstration of the potential for collaborative efforts in scientific research, specifically increasing public awareness and advocacy on the pressing issue of antibiotic resistance, and empowering individuals to actively contribute to the development of novel diagnostics. ### Competing Interest Statement The original image data were obtained using a wide-field microscope from Oxford Nanoimaging, a company in which A.N.K. is a co-founder and shareholder. The other authors declare no competing interests. ### Funding Statement This work was supported by the Oxford Martin School (by the establishment of the Oxford Martin School Programme on Antimicrobial Resistance Testing; to A.N.K., N.S., C.N., D.C. and M.A.), by Wellcome Trust grant 110164/Z/15/Z (to A.N.K.), by the Clarendon Fund Scholarships (to A.F.), and by UK Biotechnology and Biological Sciences Research Council grants BB/N018656/1 and BB/S008896/1 (to A.N.K.). The research was additionally supported by the National Institute for Health Research (NIHR) Health Protection Research Unit in Healthcare Associated Infections and Antimicrobial Resistance (NIHR200915) at the University of Oxford in partnership with United Kingdom Health Security Agency (UKHSA) and by the NIHR Oxford Biomedical Research Centre. N.S. is an NIHR Oxford BRC Senior Research Fellow. ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes The raw data images used to build this project are available from the Oxford University Research Archive. The individual segmented single cell images and classification metadata are available at Zenodo.