Automatic computer-aided detection of prostate cancer based on multiparametric magnetic resonance image analysis

Abstract

In this paper, a fully automatic computer-aided detection (CAD) method is proposed for the detection of prostate cancer. The CAD method consists of multiple sequential steps in order to detect locations that are suspicious for prostate cancer. In the initial stage, a voxel classification is performed using a Hessian-based blob detection algorithm at multiple scales on an apparent diffusion coefficient map. Next, a parametric multi-object segmentation method is applied and the resulting segmentation is used as a mask to restrict the candidate detection to the prostate. The remaining candidates are characterized by performing histogram analysis on multiparametric MR images. The resulting feature set is summarized into a malignancy likelihood by a supervised classifier in a two-stage classification approach. The detection performance for prostate cancer was tested on a screening population of 200 consecutive patients and evaluated using the free response operating characteristic methodology. The results show that the CAD method obtained sensitivities of 0.41, 0.65 and 0.74 at false positive (FP) levels of 1, 3 and 5 per patient, respectively. In conclusion, this study showed that it is feasible to automatically detect prostate cancer at a FP rate lower than systematic biopsy. The CAD method may assist the radiologist to detect prostate cancer locations and could potentially guide biopsy towards the most aggressive part of the tumour.

A fully automatic computer aided diagnosis system for peripheral zone prostate cancer detection using multi-parametric magnetic resonance imaging

“…at least one PCa lesion was detected in 54 of the 56 patients -and the median number of FP per exam was 3 (1st-3rd quartile; 1-4). Other CAD systems relying on mp-MRI for PCa detection showed slightly lower AUROC values (0.89 vs 0.91) [5,6,10]. Besides, CAD scheme accuracy may be influenced by the method used for drawing ROIs.…”

“…One good reason why mp-MRI has not yet progressed to becoming a front-line imaging modality to detect PCa is because it is a labour-intensive examination and has a steep learning curve. Indeed, interpretation requires experienced radiologists capable of analysing data extracted from the different MR sequences [5,6].…”

“…Chan et al first implemented in 2003 a CAD system for the diagnosis of peripheral zone (PZ) PCa, using a support vector machine (SVM) classifier [9]. Vos et al [6] developed in 2012 a CAD scheme using multiple sequential steps, including initial blob detection on apparent diffusion coefficient (ADC) maps followed by a local feature analysis by a supervised classifier. Similarly, Niaf et al [10] used a SVM classifier combined with a t-test featureselection method, and achieved an area under the ROC curve (AUROC) equal to 0.82.…”

Quantitative identification of magnetic resonance imaging features of prostate cancer response following laser ablation and radical prostatectomy

“…The temporal resolution of the DCE time series was 4 s. To capture characteristics on the microvasculature, we included a total of three curve features (time to peak, washout rate, and maximum enhancement) 22 and three pharmacokinetic features (K trans , K ep , V e ). 14 The microvasculature might be affected due to local tissue death and inflammation effects in the ablated area, as has been observed in the ablation of liver lesions. 2 Furthermore, as cancer also tends to have a focal appearance on DCE MRI, 16 we also calculate the Li blobness filter on the K trans , K ep , V e , maximum enhancement, time-to-peak and wash-out rate images.…”

Automatic segmentation of subcutaneous mouse tumors by multiparametric MR analysis based on endogenous contrast

“…Multiple methods for (semi)automatic tumor segmentation based on MRI images have been described for different tumors, including breast [2][3][4][5][6], prostate [7,8], brain [9][10][11], and head and neck [12] tumors. A variety of segmentation algorithms were used in these studies, including volume growing, threshold-based methods,…”