Normalization and Classification of Histopathology Electronic Health Records on Breast Cancer using NLP and ML Approaches

Abstract

Breast cancer is a prevalent form of cancer that poses a significant threat to women globally, being one of the major causes of cancer-related fatalities among this population. Detecting the disease at an early stage plays a vital role in reducing both the number of cases and the mortality rate associated with it. To achieve this, valuable information for clinical and academic research is typically found within pathology reports, which provide essential insights into the nature and characteristics of the disease. However, these reports are often complex and detailed, making it difficult to extract relevant and qualitative data efficiently. Consequently, there is a need for keyword extraction techniques specifically designed for pathology reports, allowing for the effective summarization of educational content and minimizing the laborious and time-consuming process of report analysis. One such approach is the utilization of natural language processing, which involves the application of computational algorithms to extract keywords from histopathological reports, facilitating the identification and organization of critical information.