Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
clinical-informatics [2018/02/08 22:31]
shyam [Clinical Informatics]
clinical-informatics [2019/10/13 10:23] (current)
shyam
Line 1: Line 1:
 ===== Clinical Informatics ===== ===== Clinical Informatics =====
  
-Clinical informatics is the applications ​of informatics principles and methods in the delivery healthcare services. ​This includes advanced electronic medical record (EMR) systems, computerized ​clinical decision support ​systems, wearable technology and mobile health ​(mHealthapplications,​ and reuse of EMR, study and molecular data for improving healthcare delivery. In [[http://​www.dbmi.pitt.edu/​|Department of Biomedical Informatics]], the focus is on development ​and application ​of artificial intelligence, computational and data science methods ​to advance ​clinical ​informatics. Two departmental centers provide training, staff, ​and resources. ​The [[http://​www.decic.thevislab.com/​|Data Enabled Clinical Informatics Center ​(DECIC)]] is focused on doing  +Clinical informatics is the application ​of computing ​methods ​including artificial intelligence ​in the delivery healthcare services. ​Predictive and causal models using big data and artificial intelligence will be increasingly used in clinical decision-making. These models will power a new generation of clinical decision support ​tools in the coming decade. The [[http://​www.ccri.thevislab.com/​|Center for Clinical Research Informatics ​(CCRI)]] in the [[http://​www.dbmi.pitt.edu/​|Department of Biomedical Informatics]] ​focuses ​on the development of clinical decision support tools that aid in very specific clinical tasks and are powered by predictive and causal models. 
-innovative research and healthcare delivery ​that is enabled ​by clinical, mobile health, molecular and research data+ 
 +==== Pipeline ==== 
 +  * Step 1: develop models - Develop ​artificial intelligence ​models with excellent performance for a tightly specified clinical task  
 +  * Step 2: engineer decision support tool - Build decision support tool that applies models developed in step 1 to clinical ​data in real time 
 +  * Step 3: evaluate prospectively - Evaluate decision support tool prospectively ​to establish efficacy and effect on clinical ​outcomes 
 +  * Step 4: obtain certification ​and deploy clinically - Obtain FDA certification and license and deploy clinically 
 + 
 +==== Example projects ==== 
 +  * The [[http://​www.thevislab.com/​lab/​doku.php?​id=lemr|Learning Electronic Medical Record ​(LEMRsystem]] is an intelligent EMR system that highlights relevant patient data in the EMR in the Intensive Care Unit. The LEMR system uses predictive models to draw the physician’s attention to the right data of the right patient at the right time.  
 + 
 +  * Deciding whether to attempt salvage of an infected central venous catheter (CVC) can be challenging. ​ We are developing predictive models to predict retention of CVC lines. A clinical decision support tool that is powered ​by predictive models will help clinicians in making decisions about infected CVC lines. 
 + 
 +  * Realtime intraoperative neurophysiological monitoring is used to identify adverse brain events during surgical procedures. We are developing predictive models to identify adverse events using intraoperative neurophysiological monitoring. A clinical ​decision support tool that is powered by predictive models will aid neurophysiologists in intraoperative monitoring.
  
-==== Areas of research include ==== 
-  * developing intelligent electronic medical record systems 
-  * developing clinical decision support systems for alerting on anomalous physician orders, prediction of readmission,​ 
-  * personalized modeling for precision medicine that includes developing algorithms for predicting clinical outcomes and sub-phenotyping 
-  * providing explanations for predictions by statistical and machine learning models ​ 
-  * developing decision support using wearable technology and mobile health (mHealth) applications 
-  * approaches to close the Learning Health System loop 
  
 ==== Opportunities for research are available at various stages of training ==== ==== Opportunities for research are available at various stages of training ====
Line 20: Line 25:
   * For graduate students in [[http://​www.dbmi.pitt.edu/​training-programs|biomedical informatics]],​ the [[http://​www.isp.pitt.edu/​|Intelligent System Program (ISP)]], and [[https://​cs.pitt.edu/​|computer science]]   * For graduate students in [[http://​www.dbmi.pitt.edu/​training-programs|biomedical informatics]],​ the [[http://​www.isp.pitt.edu/​|Intelligent System Program (ISP)]], and [[https://​cs.pitt.edu/​|computer science]]
   * For post-docs with graduate training in biomedical informatics,​ computer science, artificial intelligence,​ statistics, machine learning and related disciplines   * For post-docs with graduate training in biomedical informatics,​ computer science, artificial intelligence,​ statistics, machine learning and related disciplines
 +
  
 ==== Knowledge and skills in the following areas are desired ==== ==== Knowledge and skills in the following areas are desired ====
   * Probability and statistics   * Probability and statistics
   * Computer science and programming   * Computer science and programming
-  * Machine ​learning ​and data science +  * Artificial intelligence and machine ​learning 
-  * Experience with medical data (especially EMR data)+  * Experience with medical data 
  
 ==== Contact ==== ==== Contact ====