• Creation a new algorithms like Pascal' triangle to diagnosis of Skin lesions
  • aida valizadeh,1,*
    1. Student Research Committee, School of medicine, Shahid Beheshti University of Medical Sciences, Tehran, Iran


  • Introduction: skin lesions can show themselves in a variety of ways. the skin examination begins with taking a history then recognize patterns basic such as shape for example linear, annular, agminate and so on. size, depth, color, symptoms, signs, sun-exposed and other spatial relationship of the primary lesion are important in describing skin lesions and note arrangement of them. data mining methods and analysis is an appropriate tool for knowledge early diagnosis of this disease.
  • Methods: this study was performed on 98 patients and 18 types of skin lesions were examined.all records were entered into excel software and clustering algorithms as a tool of data mining which examine the existing epidemiological data. two algorithms are used for the implementation purpose. we develop SI algorithm by making a triangle like pascal's triangle with this difference we use sequence instead of numbers and we use integrating the string and we enter i instead of 1 at the i th row. then specified symmetric locations for each diagnosis by a binarization process and selected most likely incident.
  • Results: the accuracy achieved from the decision tree c4.5 which was 97.13% and by this patients may be categorized for the treatment purpose
  • Conclusion: data mining techniques are helpful in describing the distribution of skin lesions and to help healthcare to solve problems in the diagnosis and treatment of skin disease.
  • Keywords: data mining skin lesions classifications techniques decision tree c4.5