An approach for disease prediction that combines clustering, Markov models and association analysis techniques is proposed. Patient medical records are first clustered, and then a Markov model is generated for each cluster to perform predictions about illnesses a patient could likely be affected in the future. However, when the probability of the most likely state in the Markov models is not sufficiently high, the framework resorts to the association analysis. High confidence rules generated by recurring to sequential disease patterns are considered, and items induced by these rules are predicted. Experimental results show that the combination of different mining models gives good predictive accuracy and it is a feasible way to diagnose diseases. © 2014 Springer-Verlag Berlin Heidelberg.
A recommendation engine for disease prediction
Folino F;Pizzuti C
2015
Abstract
An approach for disease prediction that combines clustering, Markov models and association analysis techniques is proposed. Patient medical records are first clustered, and then a Markov model is generated for each cluster to perform predictions about illnesses a patient could likely be affected in the future. However, when the probability of the most likely state in the Markov models is not sufficiently high, the framework resorts to the association analysis. High confidence rules generated by recurring to sequential disease patterns are considered, and items induced by these rules are predicted. Experimental results show that the combination of different mining models gives good predictive accuracy and it is a feasible way to diagnose diseases. © 2014 Springer-Verlag Berlin Heidelberg.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
