A brilliant Ghanaian Senior High School student has broken the mold by successfully creating a system that uses predictive analytics model to predict breast cancer. The student named Mustapha Diyaol Haqq attends T.I Ahmadiyya Senior High School in Kumasi and he has been working on an artificial intelligence model that can diagnose and make predictions on breast cancer.
The teenager is also a content developer, code instructor and volunteer at Ghana Code Club. He has proven in all ways to be an intelligent student who is intent on surpassing all expectations. Mustapha who is passionate about solving practical problems has created a hypothetical algorithm that makes it easier for women to quickly detect breast cancer. He said: “I work as a code instructor and mentor at Ghana Code Club. I taught myself how to code. I have also written my first research paper on Using Predictive analytics to diagnose breast cancer.”
The Machine Learning/Artificial Intelligence system uses two major classifications – namely the malignant (cancerous) and benign (not cancerous) to classify a breast cancer tumor. Mustapha explained this in an easier way: “The goal is to classify whether the breast cancer is benign or malignant and predict the recurrence and non-recurrence of malignant cases after a certain period. To achieve this we used machine learning classification methods to fit a function that can predict the discrete class of new input.” Ghanaian teenager creates an artificial intelligence system that predicts and diagnoses breast cancer The model was trained and tested using the Breast Cancer datasets available on the machine learning repository maintained by the University of California, Irvine.
“We’ve shown it is possible to predict and diagnose Breast Cancer with Machine Learning/ Artificial Intelligence. We are confident with larger data of Ghanaians, models could be developed with higher accuracy scores that could be used in the real world by doctors to diagnose breast cancer efficiently, with ease and a higher accuracy,” he added. The model has been tested using the Breast Cancer database available on the machine learning repository maintained by the University of California in the USA.