Improving the accuracy of medical diagnosis with causal machine learning

Abstract

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.

Toward Causal Representation Learning

“…However, if we train a clinical system on doctors' actions in controlled settings, the system will likely provide little additional insight compared to the doctors' knowledge and may fail in surprising ways when deployed [19]. While it may be useful to automate certain decisions, an understanding of causality may be necessary to recommend treatment options that are personalized and reliable [3], [6], [31], [164], [201], [224], [242], [273].…”

AI for radiographic COVID-19 detection selects shortcuts over signal

“…By illuminating the construction of radiographic datasets in greater detail, these data will make it easier for domain experts to predict likely sources of confounding. Additionally, these metadata enable the construction of models that explicitly control for confounds, providing a route to AI systems that generalize well even in the context of confounded training data [22][23][24]. In contrast, we note that a popular set of approaches to improve generalization performance, known as "unsupervised domain adaptation," are precluded by the presence of worst-case confounding because these methods rely on learning models invariant to data-source labels, which will be perfectly correlated with the pathology labels [25].…”

Causal Research on Soil Temperature and Moisture Content at Different Depths

“…In recent years, causal analysis has become one of the most challenging fields in machine learning [8], [9], [10], [11], [12], [13], [14]. For examples, some researchers proposed causal analysis methods in medical diagnosis [15], [16] and social sciences [17], [18]. Causal analysis plays an important role in revealing the essential relationship of things and identifying causal relationship is important for effective management recommendations on climate, agriculture, epidemiology, financial regulation, and much else [6].…”