Applying Artificial Neural Network for the Classification of Breast Cancer Using Infrared Thermographic Images

The second type of cancer that kills more women in the world is breast cancer. If the prognosis is done at an early stage of the disease, women can have a better chance of cure. However, the access to medical exams in poor countries is usually precarious. This work describes the study of a computer-assisted diagnostic system using thermal imaging. The images are generated by a thermographic camera that has a lower cost than the equipment used in conventional exams. We propose a system that classifies the thermographic breasts images in “normal” and “abnormal”. We have analyzed 8 statistical characteristics: mean, variance, standard deviation (SD), skewness, kurtosis, entropy, range and median. The classification used an Artificial Neural Network (ANN) and got a result of 87 % in sensitivity, 83 % in specificity and 85 % in accuracy.

[1]  M. Frize,et al.  Analysis of breast thermography with an artificial neural network , 2004, The 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[2]  Christopher M. Bishop,et al.  Neural networks for pattern recognition , 1995 .

[3]  Qing Wang,et al.  Morphological measurement of localized temperature increase amplitudes in breast infrared thermograms and its clinical application , 2008, Biomed. Signal Process. Control..

[4]  E Y Ng,et al.  Numerical computation as a tool to aid thermographic interpretation , 2001, Journal of medical engineering & technology.

[5]  R. Simmons,et al.  Effectiveness of a noninvasive digital infrared thermal imaging system in the detection of breast cancer. , 2008, American journal of surgery.

[6]  V. Umadevi,et al.  Interpreter for breast thermogram characterization , 2010, 2010 IEEE EMBS Conference on Biomedical Engineering and Sciences (IECBES).

[7]  David M. Skapura,et al.  Neural networks - algorithms, applications, and programming techniques , 1991, Computation and neural systems series.

[8]  Richard Graham Knowles,et al.  Nitric oxide synthase activity in human breast cancer. , 1995, British Journal of Cancer.

[9]  Adhi Susanto,et al.  Principal Component Analysis combined with First Order Statistical Method for Breast Thermal Images Classification , 2011 .

[10]  Jake K. Aggarwal,et al.  Boundary extraction in thermal images by edge map , 2004, SAC '04.

[11]  G. Wishart,et al.  The accuracy of digital infrared imaging for breast cancer detection in women undergoing breast biopsy. , 2010, European journal of surgical oncology : the journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology.

[12]  C. Mathers,et al.  Cancer incidence and mortality worldwide: Sources, methods and major patterns in GLOBOCAN 2012 , 2015, International journal of cancer.

[13]  Aura Conci,et al.  A New Database for Breast Research with Infrared Image , 2014 .

[14]  U. Rajendra Acharya,et al.  Thermography Based Breast Cancer Detection Using Texture Features and Support Vector Machine , 2012, Journal of Medical Systems.