Improving the accuracy of medical diagnosis with causal machine learning

• Published in: Nature communications Vol. 11; no. 1; pp. 3923 - 9
• Main Authors: Richens, Jonathan G., Lee, Ciarán M., Johri, Saurabh
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 11.08.2020; Nature Publishing Group; Nature Portfolio
• Subjects: 631/114/2413; 639/705/531; 692/700/139; Accuracy; Algorithms; Bayes Theorem; Causation; Data Accuracy; Data Collection; Decision Making; Diagnosis; Diagnosis, Computer-Assisted; Diagnostic systems; Disease; Humanities and Social Sciences; Humans; Inference; Learning algorithms; Machine Learning; Medical diagnosis; Models, Statistical; multidisciplinary; Patients; Physicians; Science; Science (multidisciplinary); Signs and symptoms
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-020-17419-7

Machine learning promises to revolutionize clinical decision making and diagnosis. In medical diagnosis a doctor aims to explain a patient’s symptoms by determining the diseases causing them. However, existing machine learning approaches to diagnosis are purely associative, identifying diseases that are strongly correlated with a patients symptoms. We show that this inability to disentangle correlation from causation can result in sub-optimal or dangerous diagnoses. To overcome this, we reformulate diagnosis as a counterfactual inference task and derive counterfactual diagnostic algorithms. We compare our counterfactual algorithms to the standard associative algorithm and 44 doctors using a test set of clinical vignettes. While the associative algorithm achieves an accuracy placing in the top 48% of doctors in our cohort, our counterfactual algorithm places in the top 25% of doctors, achieving expert clinical accuracy. Our results show that causal reasoning is a vital missing ingredient for applying machine learning to medical diagnosis. In medical diagnosis a doctor aims to explain a patient’s symptoms by determining the diseases causing them, while existing diagnostic algorithms are purely associative. Here, the authors reformulate diagnosis as a counterfactual inference task and derive new counterfactual diagnostic algorithms.