Standardised image acquisition and comprehensive automated analysis of multi-parametric Magnetic Resonance Imaging (mpMRI) are crucial in prostate cancer for optimal diagnosis and therapy and improving patient outcome. There is a clear need for computer-aided analysis and for software which can support clinical workflow to enable faster, more quality-controlled analysis and the extraction of both functional and anatomical information on suspicious lesions.

Prostate cancer (PCa) is the second most common cancer in men worldwide. The serum level of prostate-specific antigen (PSA) has been widely used for screening since the early 1990s. If after digital rectal examination, a cancer is suspected a biopsy is required. PSA, however, remains a controversial method of choice as it has poor sensitivity and an unacceptably low specificity. In addition, the low-cost 12-core trans-rectal ultrasound (TRUS)-guided biopsy routinely misses and under-stages cancers. As a consequence, too many patients are still being unnecessarily treated for indolent cancers.

Technological advances in MR sequences over the last few decades have resulted in significant improvements in MRI so that it is now a pivotal modality in prostate cancer management. Multiparametric Magnetic Resonance Imaging (mpMRI) combines anatomical images from T2-weighted imaging (T2wI) with functional sequences:

  • diffusion-weighted imaging (DWI), which quantifies the microscopic mobility of water molecules in tissues, and the apparent-diffusion coefficient (ADC) derived from it.
  • dynamic contrast-enhanced (DCE) MRI, which is based on the permeability of blood vessels and extravasation of contrast agent into adjacent tissue.

While high-resolution T2wI provides the best assessment of the prostate’s morphology, margins, and internal structure, DWI brings specificity, and DCE adds sensitivity, together making mpMRI especially effective in revealing anterior prostate cancer in men with negative random TRUS-biopsy.

MpMRI offers considerable information on prostatic lesions including the localisation,

The Dynamika software allows the computation of an ADC map from a DWI dataset if it is not already provided within the study. Furthermore, Regions-Of-Interest (ROIs) defined in a view are instantaneously propagated, mapped, and can be modified in the other views, with the relevant pixel value statistics being available.
Optimising (Structured) Reporting and Communication of Findings. The benefits of proforma type reporting have been widely acknowledged, and the PiRADS standardised graphic prostate scheme and scores aim to harmonise the reading and communication of the findings. Nevertheless, assessing multiple lesions by mapping subjective interpretations of different images into a standardised score and localising each one on pre-defined prostatic regions, requires both a high level of expertise and time.

The introduction of interactive PiRADS report forms in the Dynamika software allows readers to experience a direct connection between the images and the reporting sheet. Images, ROIs and reports are no longer detached, and instant feedback in the report can serve a better understanding and analysis of the images.

Not only are the scans within a study synchronized spatially, but Dynamika also links manipulations such as zoom and pan to retain a coherent presentation of information between the MR sequences. Upon review, the user can open a window of the interactive PiRADS form, in which the synchronised ROIs are listed to be assigned to a specific sector defined by the standardised prostatic regions diagram.

The PiRADS form is regularly updated with the most recent research updates in line with PIRADS v1 and v2 and include PSA, prostate and lesion volume, biopsy results, Gleason score, extracapsular extension, seminal vesicular invasion, lesion scores, DWI results, and 2D visualisations. In addition, the customisable user individual interface allows for adjustments for both PI-RADS updates as well as for user-specific preferences. Advances in computer aided-diagnosis (CADx) could offer decreased reading time and consistent risk assessment of cancer presence. Evaluation of the principal current CADx systems for “Prostate Cancer Diagnosis” has unfortunately shown that they are not fully ready yet. Improvements will be made over the next decade and the wide deployment of prostate CADx systems in the clinical environment will eventually occur.

In the meantime, more focused applications for suspicious lesion detection, localisation and description, based on the combination of T2wI, DWI and DCE, could help readers efficiently grade and report lesions in PiRADS form. This is an area in which the company Image Analysis, developers of the Dynamika software package, is actively involved. Enabling Advanced Quantitative Analysis. Advanced quantitative analysis and colour maps based on DCE (parametric maps, pharmacokinetic parameters, subtraction) are also available and novel methods developed either by in-house research, collaborations, or external innovators – are continually integrated into the Dynamika software to aid lesion classification. These quantitative outcomes may be linked to PiRADS scores for lesions and therefore allow a more precise monitoring for a specific lesion. Facilitating Disease and Treatment Monitoring. The ability to quantify ‘evolution’ over time is key in active surveillance or treatment monitoring. Imaging biomarkers are used to categorise lesions, measure disease progression or estimate doses for focal radiotherapy as well as guide biopsies. Multiple images of multiple time series need to be compared. This is difficult from a viewing perspective as well as from a timing aspect. The time-consuming task of comparison of a current image with a prior is automated. All datasets and reports are stored in one central database, easily and rapidly accessible from any computer connected to the internet. Further, the software organises the arrangement of the individual scans in an intuitive way and allows easy identification and propagation of ROIs between scans and studies.

MpMRI is growing in many areas of PCa management, as a result of the constant efforts of clinical research communities to promulgate and refine standards for acquisition, interpretation and reporting. The value and richness of the multi-parametric approach is undeniable in oncology, and mpMRI offers hopes and opportunities to further expand the understanding of PCa mechanisms, provide better detection and staging of lesions, evaluate therapeutic effects more efficiently and deliver more focused and personalised treatments.

Technology needs to adapt and offer better, easier and faster ways to present and analyse ever-larger and more complex clinical data. In the current era of evidence-based medicine, technology also needs to provide platforms to allow multi-disciplinary and multi-site/national collaborative research to investigate, evaluate and validate new treatments, quantitative imaging markers, or computer aided-diagnosis tools for PCa. Dynamika is such a dedicated cloud platform [9] developed for and in collaboration with researchers, radiologists and urologists. By incorporating the latest standards and keeping abreast of science, Image Analysis aims at improving quality and efficiency, and ensuring reliability and reproducibility of PCa diagnosis and treatment from more and more complex mpMRI data.

1. Andriole GL Prostate cancer screening in the randomized prostate, lung, colorectal,
and ovarian cancer screening trial: mortality results after 13 years of follow-up,. J Natl
Cancer Inst. 2012; 104: 125
2. Serefoglu EC How reliable is 12-core prostate biopsy procedure in the detection of
prostate cancer?”J Can Urol Ass 2013; 5: E293–E298,
3. Komai Y High diagnostic ability of multiparametricmagnetic resonance imaging
to detect anterior prostate cancer missed by transrectal 12-core biopsy. J Urology, 2013;
190: 867
4. Turkbey B Prostate cancer: value of multiparametric MR imaging at 3 T for detection—
histopathologic correlation. Radiology 2010; 255 : 89.
5. Barentsz JO & Richenberg ESUR prostate MR guidelines, J Eur Radiol. 2012; 22: 746.
6. American College of Radiology. PI-RADSProstate Imaging-Reporting and Data System,
2015 version 2, Reston, VA, USA; 2015.
7. Image Analysis, MRI analysis platform Dynamika
8. Wang S et al. Computer Aided-Diagnosis of Prostate Cancer on Multiparametric MRI: A
Technical Review of Current Research. Biomed Res Int. 2014; 789561.
9. Kubassova O “Scaling computer aided detection from workstation to the cloud: medical
image analysis today”, Diagnostic Imaging Europe, Oct 2014

Guidelines on staging and characterization of prostate cancer have been published by the National Comprehensive Cancer Network.

Experience: Scoring Systems
  • RECIST 1.1
  • irRC (Immune-Related Response Criteria)
  • PET-based Standard Uptake Value (SUV)
  • Volumetric Assessment
Experience: Imaging
  • CT
  • anatomical MRI
  • perfusion imaging (DCE-MRI, DSC)
  • diffusion imaging (ADC, DWI, DTI)
  • Multiparametric MRI (mpMRI)
  • PET/CT
  • Multiparametric ultrasound (mpUS)

Since 2007, over 2000 articles were published to cover scientific discoveries, technology break-throughs and special cases. We list here some critically important papers and abstracts.


Combining our technologies and business advisory services with promising life science companies has yielded spectacular results over the past five years. As a trusted partner to many biotech and pharma companies, IAG’s team is proud to share your words and quotes.