Best detection for breast cancer found
Research led by Professor Phoebe Chen has analysed algorithmic data used in two of the most common breast screening techniques: Microarray DNA tissue sampling and diagnostic medical imaging, such as ultrasound, X-ray, Computed Tomography (CT) scan and Magnetic Resonance Imaging (MRI).
The results of the La Trobe research suggest an algorithmic formula known as the SMO (Sequential Minimal Optimisation) formula performs best in early detection of breast cancer.
The study found that SMO achieved better results for the majority of test data than another widely used computational method, the Synthetic Minority Over-sampling Technique (SMOTE), especially for data derived from microarray based DNA sampling diagnostics.
Professor Chen, Chair and Director of Research in the Department of Computer Science and Computer Engineering said: "From a clinical view, detecting early-stage breast cancer is difficult.
"Early detection requires accurate and reliable diagnostic processes and the use of robust prediction techniques," she said.
"Our aim is for these findings to help guide clinicians to the best suited formula to identify breast cancer in its early stages.
"With early identification of tumour types and treatment we hope these findings will help reduce breast cancer-related death."
Breast cancer is the most common cancer for Australian women, with recent research showing the chance of a woman being affected by invasive breast cancer at some point in her lifetime is about one in eight.
The chance of death is one in 35. It is the second leading cause of cancer-related death for Australian women.
Difficult to treat in advanced stages, early diagnosis of breast cancer means patients have a much greater chance of being treated successfully.