RESUMO
Personalized medicine in cancer treatment aims to treat each individual's cancer tumor uniquely based on the genetic sequence of the cancer patient and is a much more effective approach compared to traditional methods which involve treating each type of cancer in the same, generic manner. However, personalized treatment requires the classification of cancer-related genes once profiled, which is a highly labor-intensive and time-consuming task for pathologists making the adoption of personalized medicine a slow progress worldwide. In this paper, we propose an intelligent multi-class classifier system that uses a combination of Natural Language Processing (NLP) techniques and Machine Learning algorithms to automatically classify clinically actionable genetic mutations using evidence from text-based medical literature. The training data set for the classifier was obtained from the Memorial Sloan Kettering Cancer Center and the Random Forest algorithm was applied with TF-IDF for feature extraction and truncated SVD for dimensionality reduction. The results show that the proposed model outperforms the previous research in terms of accuracy and precision scores, giving an accuracy score of approximately 82%. The system has the potential to revolutionize cancer treatment and lead to significant improvements in cancer therapy.
