Creators: Marek Lubicz (1), Konrad Pawelczyk (2), Adam Rzechonek (2), Jerzy Kolodziej (2)
 -- (1) Wroclaw University of Technology, wybrzeze Wyspianskiego 27, 50-370, Wroclaw, Poland
 -- (2) Wroclaw Medical University, wybrzeze L. Pasteura 1, 50-367 Wroclaw, Poland

Donor: Maciej Zieba (maciej.zieba '@' pwr.wroc.pl), Jakub M. Tomczak (jakub.tomczak '@' pwr.wroc.pl), (+48) 71 320 44 53  

Date: November, 2013

Data Set Information:

The data was collected retrospectively at Wroclaw Thoracic Surgery Centre for patients who underwent major lung resections for primary lung cancer in the years 2007a€“2011. The Centre is associated with the Department of Thoracic Surgery of the Medical University of Wroclaw and Lower-Silesian Centre for Pulmonary Diseases, Poland, while the research database constitutes a part of the National Lung Cancer Registry, administered by the Institute of Tuberculosis and Pulmonary Diseases in Warsaw, Poland.

Attribute Information:

1. DGN: Diagnosis - specific combination of ICD-10 codes for primary and secondary as well multiple tumours if any (DGN3,DGN2,DGN4,DGN6,DGN5,DGN8,DGN1)
2. PRE4: Forced vital capacity - FVC (numeric)
3. PRE5: Volume that has been exhaled at the end of the first second of forced expiration - FEV1 (numeric)
4. PRE6: Performance status - Zubrod scale (PRZ2,PRZ1,PRZ0)
5. PRE7: Pain before surgery (T,F)
6. PRE8: Haemoptysis before surgery (T,F)
7. PRE9: Dyspnoea before surgery (T,F)
8. PRE10: Cough before surgery (T,F)
9. PRE11: Weakness before surgery (T,F)
10. PRE14: T in clinical TNM - size of the original tumour, from OC11 (smallest) to OC14 (largest) (OC11,OC14,OC12,OC13)
11. PRE17: Type 2 DM - diabetes mellitus (T,F)
12. PRE19: MI up to 6 months (T,F)
13. PRE25: PAD - peripheral arterial diseases (T,F)
14. PRE30: Smoking (T,F)
15. PRE32: Asthma (T,F)
16. AGE: Age at surgery (numeric)
17. Risk1Y: 1 year survival period - (T)rue value if died (T,F)

Class Distribution: the class value (Risk1Y) is binary valued.
  Risk1Y Value:   Number of Instances:
T                  70
N                  400

Summary Statistics:

Binary Attributes Distribution:
  PRE7 Value:   Number of Instances:
T               31
N             439
  PRE8 Value:   Number of Instances:
T               68
N             402
  PRE9 Value:   Number of Instances:
T               31
N             439
  PRE10 Value:   Number of Instances:
T               323
N             147
  PRE11 Value:   Number of Instances:
T               78
N             392
  PRE17 Value:   Number of Instances:
T               35
N             435
  PRE19 Value:   Number of Instances:
T               2
N             468
  PRE25 Value:   Number of Instances:
T               8
N             462
  PRE30 Value:   Number of Instances:
T               386
N             84
  PRE32 Value:   Number of Instances:
T               368
N             2

Nominal Attributes Distribution:
  DGN Value:   Number of Instances:
DGN3           349
DGN2           52
DGN4           47
DGN6           4
DGN5           15
DGN8           2
DGN1           1
  PRE6 Value:   Number of Instances:
PRZ2           27
PRZ1           313
PRZ0           130
  PRE14 Value:   Number of Instances:
OC11           177
OC14           17
OC12           257
OC13           19

Numeric Attributes Statistics:
Min   Max   Mean    SD      
   PRE4:    1.4   6.3   3.3     0.9  
   PRE5:    0.96  86.3  4.6     11.8  
   AGE:     21    87    52.5    8.7

Relevant Papers:

Zi??ba, M., Tomczak, J. M., Lubicz, M., & ??wi?…tek, J. (2013). Boosted SVM for extracting rules from imbalanced data in application to prediction of the post-operative life expectancy in the lung cancer patients. Applied Soft Computing. [Web link]
- Results:
 -- Boosted SVM for for imbalanced data gained the Gmean value equal 0.657,
 -- Decision rules induced using Boosted SVM as an oracle gained the Gmean value equal 0.648.

Citation Request:

Zi??ba, M., Tomczak, J. M., Lubicz, M., & ??wi?…tek, J. (2013). Boosted SVM for extracting rules from imbalanced data in application to prediction of the post-operative life expectancy in the lung cancer patients. Applied Soft Computing. [Web link]

BibTeX:

@article{zieba2013boosted,
 title={Boosted SVM for extracting rules from imbalanced data in application to prediction of the post-operative life expectancy in the lung cancer patients},
 author={Zi{k{e}}ba, Maciej and Tomczak, Jakub M and Lubicz, Marek and {'S}wi{k{a}}tek, Jerzy},
 journal={Applied Soft Computing},
 year={2013},
 publisher={Elsevier},
 doi={[Web link]}
}