| From Pedro Domingos
| 1. Title: Echocardiogram Data
| 
| 2. Source Information:
|    -- Donor: Steven Salzberg (salzberg@cs.jhu.edu)
|    -- Collector:
|       -- Dr. Evlin Kinney
|       -- The Reed Institute
|       -- P.O. Box 402603
|       -- Maimi, FL 33140-0603
|    -- Date Received: 28 February 1989
| 
| 3. Past Usage:
|    -- 1. Salzberg, S. (1988).  Exemplar-based learning: Theory and
|          implementation (Technical Report TR-10-88).  Harvard University,
|          Center for Research in Computing Technology, Aiken Computation
|          Laboratory (33 Oxford Street; Cambridge, MA 02138).
|       -- Steve applied his EACH program to predict survival (i.e., life
|          or death), did not use the wall-motion attribute, and recorded 87 
|          correct and 29 incorrect in an incremental application to this
|          database.  He also showed that, by tuning EACH to this domain,
|          EACH was able to derive (non-incrementally) a set of 28 
|          hyper-rectangles that could perfectly classify 119 instances.
|    -- 2. Kan, G., Visser, C., Kooler, J., & Dunning, A. (1986).  Short
|          and long term predictive value of wall motion score in acute 
|          myocardial infarction.  British Heart Journal, 56, 422-427.
|       -- They predicted the same variable (whether patients will live
|          one year after a heart attack) using a different set of 345
|          instances.  Their statistical test recorded a 61% accuracy
|          in predicting that a patient will die (post-hoc fit).
|    -- 3. Elvin Kinney (in communication with Steven Salzberg) reported
|          that a Cox regression application recorded a 60% accuracy
|          in predicting that a patient will die.
| 
| 4. Relevant Information:
|   -- All the patients suffered heart attacks at some point in the past.
|      Some are still alive and some are not.  The survival and still-alive
|      variables, when taken together, indicate whether a patient survived
|      for at least one year following the heart attack.  
| 
|      The problem addressed by past researchers was to predict from the 
|      other variables whether or not the patient will survive at least
|      one year.  The most difficult part of this problem is correctly
|      predicting that the patient will NOT survive.  (Part of the difficulty
|      seems to be the size of the data set.)
| 
| 5. Number of Instances: 132
| 
| 6. Number of Attributes: 13 (all numeric-valued)
| 
| 7. Attribute Information:
|    1. survival -- the number of months patient survived (has survived,
| 		  if patient is still alive).  Because all the patients
| 		  had their heart attacks at different times, it is 
| 		  possible that some patients have survived less than
| 		  one year but they are still alive.  Check the second
| 		  variable to confirm this.  Such patients cannot be 
| 		  used for the prediction task mentioned above.
|    2. still-alive -- a binary variable.  0=dead at end of survival period,
| 		     1 means still alive 
|    3. age-at-heart-attack -- age in years when heart attack occurred
|    4. pericardial-effusion -- binary. Pericardial effusion is fluid
| 			      around the heart.  0=no fluid, 1=fluid
|    5. fractional-shortening -- a measure of contracility around the heart
| 			       lower numbers are increasingly abnormal
|    6. epss -- E-point septal separation, another measure of contractility.  
| 	      Larger numbers are increasingly abnormal.
|    7. lvdd -- left ventricular end-diastolic dimension.  This is
| 	      a measure of the size of the heart at end-diastole.
| 	      Large hearts tend to be sick hearts.
|    8. wall-motion-score -- a measure of how the segments of the left
| 			   ventricle are moving
|    9. wall-motion-index -- equals wall-motion-score divided by number of
| 			   segments seen.  Usually 12-13 segments are seen
| 			   in an echocardiogram.  Use this variable INSTEAD
| 			   of the wall motion score.
|    10. mult -- a derivate var which can be ignored
|    11. name -- the name of the patient (I have replaced them with "name")
|    12. group -- meaningless, ignore it
|    13. alive-at-1 -- Boolean-valued. Derived from the first two attributes.
|                      0 means patient was either dead after 1 year or had
|                      been followed for less than 1 year.  1 means patient 
|                      was alive at 1 year.
| 
| 8. Missing Attribute Values: (denoted by "?")
|    Attribute #:    Number of Missing Values: (total: 132)
|    ------------    -------------------------
|               1    2  
|               2	   1  
|               3	   5  
|               4	   1  
|               5	   8  
|               6	   15 
|               7	   11 
|               8	   4  
|               9	   1  
|              10	   4 
|              11	   0 
|              12	   22
|              13	   58
| 
| 9. Distribution of attribute number 2: still-alive
|    Value   Number of instances with this value
|     ----   -----------------------------------
|       0    88 (dead)
|       1    43 (alive)
|       ?    1
|     Total  132
| 
| 
| 10. Distribution of attribute number 13: alive-at-1
|    Value   Number of instances with this value
|     ----   -----------------------------------
|       0    50
|       1    24
|       ?    58
|     Total  132

0, 1.

age at heart attack: continuous.
pericardial effusion: 0, 1
fractional shortening: continuous.
E-point septal separation: continuous.
left ventricular end-diastolic dimension: continuous.
wall-motion-score: continuous.
wall-motion-index: continuous.