Why algorithmic speed can be more important than algorithmic accuracy

Abstract

Artificial Intelligence (AI) often outperforms human doctors in terms of decisional speed. For some diseases, the expected benefit of a fast but less accurate decision exceeds the benefit of a slow but more accurate one. In such cases, we argue, it is often justified to rely on a medical AI to maximise decision speed – even if the AI is less accurate than human doctors.

Keywords

Bioethics and medical ethics clinical ethics enhancement technologies philosophical aspects

Get full access to this article

View all access options for this article.

References

Norgeot

Glicksberg

Butte

. A call for deep-learning healthcare. Nat Med. 2019; 25: 14–15.

Grote

. Trustworthy medical AI systems need to know when they don’t know. Jour Med Eth. 2021; 37: 337–338.

Alvarado

. Should we replace radiologists with deep learning? Pigeons, error and trust in medical AI. Bioeth 2021; 36: 121–133.

Bjerring

Busch

. Artificial intelligence and patient centered decision-making. Phil Tech 2021; 34: 349–371.

Owens

Walker

. Those designing healthcare algorithms must become actively anti-racist. Nat Med 2020; 26: 1327–1328.

Santoni de Sio

Mecacci

. Four responsibility gaps with artificial intelligence: Why they matter and how to address them. Phil Tech 2021; 34: 1057–1084.

Birch

Creel

Jha

, et al. Clinical decisions using AI must consider patient values. Nat Med 2022; 28: 229–232.

Sorell

Rajpoot

Verrill

. Ethical issues in computational pathology. Jour Med Eth 2022; 48: 278–284.

Payne

Bettman

. When time Is money: Decision behavior under opportunity-cost time pressure. Org Beh Hum Dec Pro 1996; 66: 131–152.

10.

https://cpr.heart.org/en/resources/bystander-cpr

11.

Gyawali

Ramakrishna

Dhamoon

. Sepsis: The evolution in definition, pathophysiology, and management. Open Med 2019; 7: 1–13.

12.

Komorowski

Celi

Badawi

, et al. The Artificial Intelligence Clinician learns optimal treatment strategies for sepsis in intensive care. Nat Med 2018; 24: 1716–1720.

13.

Vincent

, et al. Assessment of the worldwide burden of critical illness: the intensive care over nations (ICON) audit. The Lanc Respir Med. 2014; 2: 380–386.

14.

Zhao

Shen

Wang

. Early prediction of sepsis based on machine learning algorithm. Comp Int Neu. 2021; 6522633.

15.

Pasa

Golkov

Pfeiffer

, et al. Efficient deep network architectures for fast chest X-ray tuberculosis screening and visualization. Sci Rep. 2019; 9: 6268.

16.

Nho

Saykin

. Deep learning in Alzheimer's disease: Diagnostic classification and prognostic prediction using neuroimaging data. Front Agi Neu. 2019; 11: 220–220.

17.

Islam

, et al.. Prediction of sepsis patients using machine learning approach: A meta-analysis. Comp Met Prog Bio. 2019; 170: 1–9.

18.

Fleuren

, et al. Machine learning for the prediction of sepsis: a systematic review and meta-analysis of diagnostic test accuracy. Int Care Med. 2020; 46: 383–400.

19.

https://dihi.org/project/sepsiswatch/

20.

Sendak , et al. Real-world integration of a sepsis deep learning technology into routine clinical care: Implementation study. JMIR Med Inform. 2020; 8: e15182.

21.

https://dihi.org/outcome/the-human-body-is-a-black-box-supporting-clinical-decision-making-with-deep-learning-2/

22.

Kempt

Nagel

. Responsibility, second opinions and peerdisagreement: ethical and epistemological challenges of using AI in clinical diagnostic contexts. Jour Med Eth. 2022; 48: 222–229.

23.

Lopansri

, et al. Physician agreement on the diagnosis of sepsis in the intensive care unit: estimation of concordance and analysis of underlying factors in a multicenter cohort. Jour Int Care. 2019; 7: 1–17.

24.

Zerilli

Knott

Maclaurin

, et al.. Transparency in algorithmic and human decision-making: Is there a double standard?. Phil Tech. 2019; 32: 661–683.

25.

Günther

Kasirzadeh

. Algorithmic and human decision making: for a double standard of transparency. AI & Soc. 2021; 37: 375–381.

26.

Prescott

, et al. Late mortality after sepsis: propensity matched cohort study. Br Med J. 2016; 353: i2375.