A Revolution in Workflow

Watson Elementary® has a huge range of benefits, many of which stem from it’s well designed workflow. Not only is it easy to use but it also saves an incredible amount of time. Below is an outline of the Watson Elementary® process and where these benefits come to life.

Import Multiple MRI Types

Simply start by importing the various MRI series representing the multiple parameters to be analysed from your pacs server or from portable media, such as CD, DVD or USB memory stick.

In an instant, Watson Elementary® generates screen representations of these series, which can be navigated in all spatial dimensions. And of course, DCE image series can also be traversed temporally.

Watson elementary® analyses the following types of MR images

– T2-weighted (T2WI)
– Diffusion-weighted images (DWI)
– Dynamic Contrast Enhanced (DCE) images

Automatic Co-registration

Next, Watson elementary® automatically co-registers (‘fuses’) the various series so that you are able to observe features in the different images types in a common coordinate system and in a common spatial resolution.

All three series can be conveniently viewed ‘in lockstep’(synchronised frames), including colour overlays of the various relevant parameters that correspond with malignancy.

Already, you have at your disposal all practical tools and components for a preliminary visual assessment of multiple parameters in relation to each other, such as ADC maps, wash-out, Ve, Kep and K-trans.

Automatic Analysis

Better still, within seconds Watson Elementary® produces information for an in-depth image examination, including an automated pharmacokinetic analysis of DCE sequences, an ADC analysis as well as a fully automated analysis of features that correlate positively with malignancy.

By applying colour overlays, the CAD system provides efficient overviews of the various parameters of interest in relation to the prostate anatomy.

Most prominently, Watson Elementary® produces colour maps that show a malignancy correlation factor for each available image voxel. This so-called Malignancy Attention Index, or MAI, greatly facilitates the identification and localisation of ROIS.

DICOM Structured Reporting

When delineated, each ROI is automatically included into an individual record. Additionally and again automatically, Watson Elementary®, subsequently generates a comprehensive standardised report that contains all specific records, including MAI maps, lesion location maps, relevant images as well as added observer (e.g. PI-RADS scores and free format text) and CAD annotations.

Of course, this complete report can be printed or exported in PDF or XPS format. Similarly, you can send it to a PACS server as an encapsulated PDF object. Moreover, the complete analysis can be stored on PACS or portable media as a DICOM Structured Report, this being the most inclusive standard in imaging communications and connectivity to date.

Also, this standard ensures easy access to relevant data for other purposes, such as biopsy targeting, active surveillance, treatment and clinical studies.

Crucial Benefits

The multi-faceted approach delivers unprecedented accuracy and the Watson Elementary technology results in huge time savings both during and after analysis. We designed Watson Elementary with the user experience in mind.

Multiple Analysis Options

Hugely facilitates analysis of various multiparametric MR images.


Multiple Analysis = Greater Accuracy

Helps identify potential malignancies by automatically combining all available information


Synchronised Analysis

Automatically co-registers multiple MRI series


Highlights Potential Malignancy

Automatically generates malignancy attention maps and shows all parameters in colour overlays


Better Targeting

 Biopsy target generation with direct export to image-guided biopsy systems


Information At Your Fingertips

DICOM Structured Reporting and full connectivity to PACS


Reduced Analysis Time

Exceptionally fast co-registration and fully automated analysis. Real-time access to end-result as well as all underlying data


Eliminate Errors

No operator dependence and intra-observer variability