Time-Signal Curve Evaluation from Dynamic Contrast-Enhanced MRI for Prostate Cancer Detection

Prostate cancer is one of the most commonly diagnosed cancers globally. The diagnosis of prostate cancer is done using several approaches. Prostate-specific antigen (PSA) screening is straightforward, but prone to false-positive and false-negative results. Biopsy is a more reliable method, but due to being an invasive procedure it is primarily used for confirmation. Magnetic resonance imaging (MRI) is the third approach, which produces several imaging sequences and provides critical information to a radiologist for diagnosis. Typically, three MRI sequences – diffusion-weighted imaging (DWI), T2-weighted (T2W), dynamic contrast-enhancement (DCE) – are used in conjunction, as each sequence characterizes the tissue in its own way. This paper explores the possibility of using only DCE imaging for malignant tissue detection in the prostate. As the DCE sequence shows both spatial and temporal change in a patient's body, a set of time-signal intensity curves are constructed. The form of such curves is then analyzed to find patterns which would help separating healthy tissue from cancerous lesions. Three out of eight patients showed clear separation between curves of clinically significant cancer and benign curves, while five patients showed varying behavior inconsistent with common diagnostic methodology.