research_article August 17, 2026

Rethinking Biological Categories: Nancy Cartwright on AI in Medicine

Chikere Ugwulebo
Volume1
Issue3
Pages122-128
DOI10.5281/zenodo.21981144
Abstract

Artificial Intelligence (AI) is now central to pharmaceutical research and clinical practice across the globe. It promises faster drug discovery and more precise care. Despite heavy investment, the clinical impact of AI in medicine remains limited in both high- and low-income settings. This paper argues that the limitation is philosophical, not only technical. Medicine continues to use disease categories built for 20th century regulation and billing. When AI systems are trained on these categories, they learn and scale the error. Drawing on Nancy Cartwright’s concept of “Evidence for Use” in her The Dappled World and on African and Asian critiques of universalism in medicine, I argue that a treatment’s capacity to work is local. It depends on the specific causal arrangement in which it was established. The paper uses conceptual analysis, drawing on Cartwright's philosophy of science and African and Asian philosophy of medicine, with hypertension as an illustrative case, focusing on salt-sensitive hypertension in West Africa and East Asia. Since AI learns from broad labels, it reproduces mismatch at scale. To make AI work in medicine, we must rethink biological categories. We must move from universal labels to mechanism-based categories grounded in evidence for use. The paper shows that AI bias is rooted partly in biological classification, not only in data quality. It concludes that philosophy is infrastructure. Without recalibrating categories, AI in medicine will be fast but not good, especially for patients in Africa and Asia.

Keywords
Artificial IntelligenceBiological CategoriesEvidence for UseHypertensionNancy CartwrightPhilosophy of Medicine.
Download Article PDF
References
  1. Aronowitz, R. A. (1998). Making sense of illness: Science, society, and disease. Cambridge University Press.
  2. Boyd, R. (1999). Homeostasis, species, and higher taxa. In Species: New interdisciplinary essays (pp. 141–185). MIT Press. https://mitpress.mit.edu/9780262231989/species/
  3. Bujo, B. (2001). Foundations of an African ethic: Beyond the universal claims of Western morality. Crossroad.
  4. Cartwright, N. (1989). Nature's capacities and their measurement. Clarendon Press.
  5. Cartwright, N. (1999). The dappled world: A study of the boundaries of science. Cambridge University Press.
  6. Cartwright, N. (2007). Hunting causes and using them: Approaches in philosophy and economics. Cambridge University Press.
  7. Cartwright, N., & Hardie, J. (2012). Evidence-based policy: A practical guide to doing it better. Oxford University Press.
  8. Chaudhari, A. S., Fang, Z., Kogan, F., Wood, J., Stevens, K. J., Gibbons, E. K., ... & Gold, G. E. (2019). Discovery and clinical application of deep learning algorithms for artificial intelligence accelerated MRI. Journal of Magnetic Resonance Imaging, 49(6), 1570–1585. https://doi.org/10.1002/jmri.26581
  9. Chen, I., Pierson, E., Rose, S., Joshi, S., Ferrara, E., & Ghassemi, M. (2021). Ethical machine learning in healthcare. Annual Review of Biomedical Data Science, 4, 123–144. https://doi.org/10.1146/annurev-biodatasci-092820-114757
  10. Collins, F. S., & Varmus, H. (2015). A new initiative on precision medicine. New England Journal of Medicine, 372(9), 793–795. https://doi.org/10.1056/NEJMp1500523
  11. Cooper, R., Rotimi, C., Ataman, S., McGee, D., Osotimehin, B., Kadiri, S., ... & Muna, W. (1997). The prevalence of hypertension in seven populations of West African origin. American Journal of Public Health, 87(2), 1482–1485. https://doi.org/10.2105/AJPH.87.9.1482
  12. Deaton, A., & Cartwright, N. (2018). Understanding and misunderstanding randomized controlled trials. Social Science & Medicine, 210, 2–21. https://doi.org/10.1016/j.socscimed.2017.12.005
  13. Folkert, E. D., & Khan, M. A. (1976). Essential hypertension: A review of pathophysiology. American Heart Journal, 92(6), 1021–1028. https://doi.org/10.1016/S0002-8703%2876%2980132-0
  14. Hacking, I. (1999). The social construction of what? Harvard University Press.
  15. Harrer, S., Shah, P., Antony, B., & Hu, J. (2019). Artificial intelligence for clinical trial design. Trends in Pharmacological Sciences, 40(8), 1–4. https://doi.org/10.1016/j.tips.2019.06.003
  16. He, F. J., Li, J., & MacGregor, G. A. (2014). Effect of longer term modest salt reduction on blood pressure. Cochrane Database of Systematic Reviews, 4, CD004937. https://doi.org/10.1002/14651858.CD004937.pub3
  17. Hesslow, G. (1993). Do we need a concept of disease? Theoretical Medicine, 14(1), 1–14. https://doi.org/10.1007/BF00993957
  18. Ioannidis, J. P. A. (2008). Effectiveness of antidepressants: An evidence myth constructed from a thousand randomized trials? Philosophy of Science, 75(5), 731–743. https://doi.org/10.1086/594535
  19. Jiang, F., Jiang, Y., Zhi, H., Dong, Y., Li, H., Ma, S., ... & Wang, Y. (2017). Artificial intelligence in healthcare: Past, present and future. Stroke and Vascular Neurology, 2(4), 230–243. https://doi.org/10.1136/svn-2017-000101
  20. Kesselheim, A. S., Avorn, J., & Sarpatwari, A. (2016). The high cost of prescription drugs in the United States. JAMA, 316(8), 858–871. https://doi.org/10.1001/jama.2016.11237
  21. Krittanawong, C., Zhang, H., Wang, Z., Aydar, M., & Kitai, T. (2020). Artificial intelligence in precision cardiovascular medicine. Journal of the American College of Cardiology, 69(13), 1317–1327. https://doi.org/10.1016/j.jacc.2016.12.044
  22. Laragh, J. H. (1973). Vasoconstriction volume analysis in hypertension. Circulation, 47(2), 267–272. https://doi.org/10.1161/01.CIR.47.2.267
  23. Laragh, J. H. (2001). Laragh's lessons in pathophysiology and clinical pearls for treating hypertension. American Journal of Hypertension, 14(3), 260–269. https://doi.org/10.1016/S0895-7061%2800%2901233-9
  24. Lindeman, N. I., Cagle, P. T., Aisner, D. L., Arcila, M. E., Beasley, M. B., Berniel, E., ... & Yatabe, Y. (2018). Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors. Journal of Thoracic Oncology, 13(3), 323–358. https://doi.org/10.1016/j.jtho.2017.12.004
  25. McMurray, J. J., Packer, M., Desai, A. S., Gong, J., Lefkowitz, M. P., Rizkala, A. R., ... & Zile, M. R. (2014). Angiotensin neprilysin inhibition versus enalapril in heart failure. New England Journal of Medicine, 371(11), 1547–1557. https://doi.org/10.1056/NEJMoa1409077
  26. Messina, J., Hall, D., & Diamond, J. (2021). Clinical trial design for hypertension. Current Hypertension Reports, 23(7), 1–9. https://doi.org/10.1007/s11906-021-01158-9
  27. Oparil, S., Acelajado, M. C., Bakris, G. L., Berlowitz, D. R., Cífková, R., Dominiczak, A. F., ... & Victor, R. G. (2018). Hypertension. Nature Reviews Disease Primers, 4(1), 1–21. https://doi.org/10.1038/nrdp.2018.32
  28. Padmanabhan, S., Melnyk, O., & Curtis, A. M. (2018). The genomics of hypertension. Current Hypertension Reports, 20(10), 1–9. https://doi.org/10.1007/s11906-018-0887-0
  29. Pang, R. (2003). Chinese medicine and the problem of disease categories. Asian Bioethics Review, 1(2), 110–125. https://link.springer.com/journal/41649
  30. Peters, J., Janzing, D., & Schölkopf, B. (2017). Elements of causal inference: Foundations and learning algorithms. MIT Press.
  31. Scannell, J. W., Blanckley, A., Boldon, H., & Warrington, B. (2012). Diagnosing the decline in pharmaceutical R&D efficiency. Nature Reviews Drug Discovery, 11(3), 569–580. https://doi.org/10.1038/nrd3681
  32. SPRINT Research Group. (2015). A randomized trial of intensive versus standard blood pressure control. New England Journal of Medicine, 373(22), 2265–2266. https://doi.org/10.1056/NEJMoa1511939
  33. Tangwa, G. B. (2004). Genetic engineering, ethics and the environment. University of Yaoundé Press.
  34. Topol, E. J. (2019). Deep medicine: How artificial intelligence can make healthcare human again. Basic Books.
  35. Turnbull, F., Neal, B., Ninomiya, T., Algert, C., Woodward, M., Chalmers, J., & MacMahon, S. (2008). Effects of different blood pressure lowering regimens on major cardiovascular events. The Lancet, 371(9623), 1751–1761. https://doi.org/10.1016/S0140-6736%2808%2960790-5
  36. Vamathevan, J., Clark, D., Czodrowski, P., Dunham, I., Ferran, E., Lee, G., ... & Zhao, S. (2019). Applications of machine learning in drug discovery and development. Nature Reviews Drug Discovery, 18(6), 437–444. https://doi.org/10.1038/s41573-019-0024-5
  37. Weinberger, M. H. (1996). Salt sensitivity of blood pressure in humans. Hypertension, 27(3), 481–490. https://doi.org/10.1161/01.HYP.27.3.481
  38. Whelton, P. K., Carey, R. M., Aronow, W. S., Casey, D. E., Collins, K. J., Dennison Himmelfarb, C., ... & Wright, J. T. (2018). 2017 ACC/AHA guideline for high blood pressure in adults. Journal of the American College of Cardiology, 71(19), e13–e115. https://doi.org/10.1016/j.jacc.2017.11.006
  39. Wiens, J., Saria, S., Sendak, M., Ghassemi, M., Liu, V. X., Doshi-Velez, F., Jung, K., Heller, K., Kale, D., Saeed, K., Ossorio, P. N., Thadaney-Israni, S., & Goldenberg, A. (2019). Do no harm: A roadmap for responsible machine learning for health care. Nature Medicine, 25(9), 1337–1340. https://doi.org/10.1038/s41591-019-0548-6
  40. Woodcock, J., & LaVange, L. M. (2017). Master protocols to study multiple therapies, multiple diseases, or both. New England Journal of Medicine, 377(1), 771–774. https://doi.org/10.1056/NEJMra1510062
  41. Zhavoronkov, A., Ivanenkov, Y. A., Aliper, A., Veselov, M. S., Aladinskiy, V. A., Aladinskaya, A. V., Terentiev, V. A., Polykovskiy, D. A., Mamoshina, P., Zhebrak, A., & Aspuru-Guzik, A. (2019). Deep learning enables rapid identification of potent DDR1 kinase inhibitors. Nature Biotechnology, 37(9), 1038–1040. https://doi.org/10.1038/s41587-019-0224-x
Scroll to Top