Project Abstract

Breast cancer is a common and life-threatening illness that needs excellent detection and diagnostic techniques to enhance patient outcomes. Deep learning (DL) and machine learning (ML) approaches have emerged as useful tools for detecting breast cancer, including advantages such as increased accuracy and efficiency. However, present approaches have scalability and performance limits, highlighting the need for more study. We must offer a hybrid trustworthy breast cancer detection methodology that combines the capabilities of deep learning with a pre-trained Resnet50v2 model and ensemble-based machine learning techniques. The use of DL allows the technique to learn and extract hidden patterns from complicated breast cancer pictures, while ML algorithms add interpretability and generalization capabilities. We have to conduct lengthy tests utilizing a publicly available invasive ductal breast histopathology picture.Feature extraction is a critical step in breast cancer picture analysis. It entails finding and extracting distinguishing patterns and structures from medical pictures in order to describe complicated data more formally. Relevant aspects, such as statistical measurements, textural qualities, or shape attributes, can be extracted to expose hidden information inside photographs. This procedure is critical because it allows for the identification of image-based biomarkers, enhances tumor characterisation, and assists in discriminating between healthy and malignant tissues. Accurate feature extraction is critical for increasing breast cancer diagnostic and treatment planning accuracy, speed, and dependability.

Keywords: Machine learning, Artificial intelligence, CNN,deep learning, Transfer learning

 Conference Details

Session: Presentation Stream 10 at Presentation Slot 9

Location: CoFo 002 at Tuesday 7th 13:30 – 17:00

Markers: Randell Gaya, George Brooks (GTA)

Course: MSc Advanced Computer Science, Masters PG

Future Plans: I’m looking for an industry placement