Counterfactual measure for medical diagnosis

1. A method for providing a computer implemented medical diagnosis, the method comprising:

receiving an input from a user comprising at least one symptom of the user;

providing the at least one symptom as an input to a medical model, the medical model being retrieved from memory, the medical model comprising: a probabilistic graphical model comprising probability distributions and causal relationships between symptoms and diseases;

performing counterfactual inference on the probabilistic graphical model to obtain a prediction of the probability that the user has a disease; and

outputting a counterfactual measure determined from the counterfactual inference,

wherein the counterfactual measure is the expected number of symptoms that would not be present if a disease was treated,

wherein the probabilistic graphical model is a twin network, the twin network formed by:

creating a first graphical representation from a set of data, the first graphical representation comprising a plurality of first nodes indicating symptoms, risk factors and diseases, the first graphical representation indicating a relationship between the symptoms, the risk factors and the diseases, the first graphical representation indicating real variables;

creating a second graphical representation, wherein the second graphical representation is a copy of the first graphical representation, the second graphical representation comprising a plurality of second nodes indicating the symptoms, the risk factors and the diseases, the second graphical representation indicating counterfactual states of the real variables; and

combining the first graphical representation and the second graphical representation by the sharing of exogenous causes to create said twin network.

2. The method of claim 1 , wherein a counterfactual measure is an expected disablement.

3. The method of claim 1 , wherein the probabilistic graphical model is a Noisy-OR model.

4. The method of claim 1 , wherein a counterfactual measure is an expected disablement and the probabilistic graphical model is a Noisy-OR model.

5. The method of claim 1 , wherein performing counterfactual inference comprises using a discriminative model pre-trained to approximate the probabilistic graphical model, the discriminative model being trained using samples from said probabilistic graphical model;

deriving estimates, from the discriminative model, that the user has a disease; and

performing approximate inference on the probabilistic graphical model to obtain an indication that the user has that disease using the estimate from the discriminative model.

6. The method of claim 1 , wherein the results of predictions of the diseases from the probabilistic graphical model are ranked using a counterfactual measure.

7. The method of claim 1 , wherein the probabilistic graphical model is a twin network, a counterfactual measure is an expected disablement and the probabilistic graphical model is a Noisy-OR model.

8. A system for providing a computer implemented medical diagnosis, the system comprising a processor and a memory, the processor being adapted to:

receive an input from a user comprising at least one symptom of the user;

provide the at least one symptom as an input to a medical model, the medical model being retrieved from memory, the medical model comprising:

a probabilistic graphical model being stored in memory comprising probability distributions and causal relationships between symptoms and diseases,

perform counterfactual inference on the probabilistic graphical model to obtain a prediction of the probability that the user has a disease; and

output a counterfactual measure determined from the counterfactual inference,

wherein, the counterfactual measure is the expected number of symptoms that would not be present if a disease was treated,

wherein the probabilistic graphical model is a twin network, the twin network formed by:

creating a first graphical representation from a set of data, the first graphical representation comprising a plurality of first nodes indicating symptoms, risk factors and diseases, the first graphical representation indicating a relationship between the symptoms, the risk factors and the diseases, the first graphical representation indicating real variables;

creating a second graphical representation, wherein the second graphical representation is a copy of the first graphical representation, the second graphical representation comprising a plurality of second nodes indicating the symptoms, the risk factors and the diseases, the second graphical representation indicating counterfactual states of the real variables; and

combining the first graphical representation and the second graphical representation by the sharing of exogenous causes to create said twin network.

9. A non-transitory carrier medium comprising computer-readable instructions being adapted to cause a computer to perform:

receiving an input from a user comprising at least one symptom of the user;

providing the at least one symptom as an input to a medical model, the medical model being retrieved from memory, the medical model comprising: a probabilistic graphical model comprising probability distributions and causal relationships between symptoms and diseases;

performing counterfactual inference on the probabilistic graphical model to obtain a prediction of the probability that the user has a disease; and

outputting a counterfactual measure determined from the counterfactual inference,

wherein, the counterfactual measure is the expected number of symptoms that would not be present if a disease was treated,

wherein the probabilistic graphical model is a twin network, the twin network formed by:

creating a first graphical representation from a set of data, the first graphical representation comprising a plurality of first nodes indicating symptoms, risk factors and diseases, the first graphical representation indicating a relationship between the symptoms, the risk factors and the diseases, the first graphical representation indicating real variables;

creating a second graphical representation, wherein the second graphical representation is a copy of the first graphical representation, the second graphical representation comprising a plurality of second nodes indicating the symptoms, the risk factors and the diseases, the second graphical representation indicating counterfactual states of the real variables; and

combining the first graphical representation and the second graphical representation by the sharing of exogenous causes to create said twin network.

10. The non-transitory carrier medium of claim 9 , wherein a counterfactual measure is an expected disablement.

11. The non-transitory carrier medium of claim 9 , wherein the probabilistic graphical model is a Noisy-OR model.

12. The method of claim 1 , wherein the counterfactual inference further comprises:

updating a distribution of exogenous latent variables in the twin network based on the at least one symptom:

applying a do-operation to a node indicating the disease in the second graphical representation of the updated twin network; and

computing a probability that the user would have the disease based on the updated twin network.