Mammography Examination

hayongjoe1994 Leave a Comment
Image 1: The procedure of mammography examination and anatomy of breast.

Mammography examination is a low cost non-invasive x-ray examination that is specialised for breast only. According Ren et al. (2022), it is recommended that women who is age 40 years old and above should undergo mammography examination annually. In spite of its deterministic and stochastic effects, mammography examination remain as an important method to rule out breast cancer because it is able to give an accurate diagnosis within a short period of time. During mammography examination, there are two series of mammogram will be produced for the same patient: mediolateral oblique (MLO) view and cranial caudal (CC) view. Breast is an organ that is three dimension which is located anterior to the thoracic wall. Thus, radiologist requires two series of mammogram to classify between normal, benign and malignant and also to locate the location of the tumour if there is any in the mammogram. In my project, I will utilised the dataset from Khaled et al (2021) which consists of 1003 high-resolution Contrast-enhanced spectral mammography (CESM) images.

Image 2: Image 1 and 3 illustrate the acquire method for CC view mammogram and Image 2 and 4 illustrates the acquire method for MLO view mammogram
Case courtesy of Frank Gaillard, <a href=”https://radiopaedia.org/?lang=gb”>Radiopaedia.org</a>. From the case <a href=”https://radiopaedia.org/cases/12608?lang=gb”>rID: 12608</a>
https://www.verywellhealth.com/mammogram-what-to-expect-430283
In order to have a better understanding of the project, better understanding of the characteristics of the mammogram is needed. The definition of normal mammogram is the mammogram only consist of normal breast tissues. On the other hand, the benign mammogram means the mammogram consists of only benign tumour or calcification. However, the malignant mammogram is a mammogram that consists malignant tumour.
Image 3: The different shade of grey is represented by different number from 0 to 255
https://radiologykey.com/digital-mammography/
Digital mammogram is an image that is generated by x-ray that is passing through the breast and received by the detector. Due to the different in density of the breast tissues, the amount of the radiation that reaches the detector will be different. Thus, the different in the amount of radiation will be represented by 256 shades of grey in the mammogram where zero represent black and 255 represent white. Convolutional neural networks (CNN) enable to utilised the different in the shade of grey (pixel) to compare and contrast and determine the different between the normal, benign and malignant.
Image 4: Sample images of benign and malignancy tumour
Li et al. (2020)
Besides, the different in shape of the tumour will be used to justify either benign or malignant. According to Bassett et al. (2003), benign tumour tends to be round and oval. However, malignant tumour tends to be in irregular size. CNN is an algorithm that is mainly used in image recognition and processing. There is a likelihood that the algorithm manages to recognise the different and classifies accordingly.

References,

  1. Bassett, L. W., Conner, K., & Ms, I. (2003). The Abnormal Mammogram. Holland-Frei Cancer Medicine. 6th Edition. https://www.ncbi.nlm.nih.gov/books/NBK12642/

  2. Khaled R., Helal M., Alfarghaly O., Mokhtar O., Elkorany A., El Kassas H., Fahmy A. Categorized Digital Database for Low energy and Subtracted Contrast Enhanced Spectral Mammography images [Dataset]. (2021) The Cancer Imaging Archive. DOI:  10.7937/29kw-ae92
  3. Li, H., Niu, J., Li, D., & Zhang, C. (2020). Classification of breast mass in two‐view mammograms via deep learning. IET Image Processing, 15(2), 454–467. https://doi.org/10.1049/ipr2.12035
  4. Ren, W., Chen, M., Qiao, Y., & Zhao, F. (2022). Global guidelines for breast cancer screening: A systematic review. The Breast, 64, 85–99. https://doi.org/10.1016/j.breast.2022.04.003

Acknowledgements

I would like to acknowledge the individuals and institutions that have provided data for this collection:

  • National Cancer Institute, Cairo University, Cairo, Egypt : Special thanks to Dr. Rana Khaled, M.Sc, Prof. Maha Helal, MD, Prof. Omnia Mokhtar, MD and Dr. Hebatalla El Kassas, MD from the Department of Radiology.
  • Faculty of Computers and Artificial Intelligence, Cairo University, Cairo, Egypt – Special thanks to Omar Alfarghaly, Prof. Abeer Elkorany, and Prof. Aly Fahmy from the Department of Computer Science.

hayongjoe1994

More Posts

Leave a Reply

Your email address will not be published. Required fields are marked *