An extensible framework for recurrent breast cancer prognosis using deep learning techniques

Reddy Shiva Shankar, Ravi Swaroop Chigurupati, Priyadarshini Voosala, Neelima Pilli

Abstract


Due to population growth, early illness detection is getting more challenging. Breast cancer is the second-deadliest malignancy. An estimated one million people are newly diagnosed with the disease annually in India. Most cases are never diagnosed because they are either ignored or not reported. Also, secondary malignancies may develop after a breast cancer recurrence, including those of the brain, lungs, and bones. Early detection and treatment of people with recurring breast cancer may help prevent secondary cancers and other disorders. By examining cell and tumour data as well as data from other diseases, this project hopes to overcome this obstacle and more accurately diagnose breast cancer. Accurate diagnosis of breast cancer may be achieved with the use of machine learning techniques. The effort focuses on recurring breast cancer and aims to efficiently identify it. In ensemble learning, decision trees filter out non-essential qualities. Cancer recurrences and non-recurrences are distinguished using voting classifiers. The soft voting classifier classifies a variety of data sets with 98.24% accuracy. The proposed model has an accuracy of 0.97, a recall of 0.97, an F1-Score of 0.969, and a Choen kappa score of 0.9655, as stated by the recommended model.

Keywords


Adaboost classifier; Breast cancer; Histogram-based gradient boosting classifier; Machine learning; Multi-layer perceptron; Wisconsin breast cancer dataset

Full Text:

PDF


DOI: http://doi.org/10.11591/ijeecs.v29.i2.pp931-941

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

shopify stats IJEECS visitor statistics