Developing Predictive Models Using Electronic Medical Records: Challenges and Pitfalls

By Chris Paxton, Suchi Saria, Alexandru Niculescu-Mizil. 2013

106 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Deconfounding Reinforcement Learning in Observational Settings (2018, methodology): “Paxton et al. (2013) developed predictive models based on electronic medical records without using causal inference.”
Clinical artificial intelligence quality improvement: towards continual monitoring and updating of AI algorithms in healthcare (2022, background): “Although confounding occurs in the absence of AI-based predictions 62,63 , the CMIs becomes much more severe when clinicians utilize AI algorithms in their decision-making process 64 – 66 .”
A Non-parametric Learning Method for Confidently Estimating Patient's Clinical State and Dynamics (2016, background): “patient models are not unique), can not detect anomalous patient dynamics, and most importantly, are biased due to therapeutic intervention censoring [16].”
DeepCare: A Deep Dynamic Memory Model for Predictive Medicine (2016, background): “A wide range of information can be stored in EMRs, such as detailed records of symptoms, data from monitoring devices, clinicians’ observations [18].”
A targeted real-time early warning score (TREWScore) for septic shock (2015, methodology): “First, if a patientwith severe sepsis received characteristic treatment for septic shock but never developed shock, it is unknown whether he or she would have developed septic shock in the absence of such treatment (38).”
Mining Electronic Health Records: A Survey (2017, background): “…Pakhomov et al. 2011; Fung et al. 2008; Wells et al. 2008; Vinzamuri et al. 2014; Li et al. 2015; van der Heijden et al. 2014; Jackson et al. 2011; Paxton et al. 2013; Zhao et al. 2011; Maroco et al. 2011; Breault et al. 2002; Wang et al. 2015b; Sun et al. 2014] Time Aware Case-Control or Cohort…”
Learning (predictive) risk scores in the presence of censoring due to interventions (2015, methodology): “The choice of leaving out individuals who receive ﬂuid-bolus but do not experience septic shock is owing to the fact that their outcome is censored due to treatment (Paxton et al, 2013).”
Big data in health care: using analytics to identify and manage high-risk and high-cost patients. (2014, background): “Thus, integrating data about mental health, socioeconomic status, or other issues such as marital and living status from various sources 16 may significantly change the quality of the predictions that can be made.”