Overview

Medicsight LungCAD™ is an image analysis software tool designed to be used with CT imaging to assist radiologists in searching for and measuring potential lung nodules.

The solution is available as an application programming interface (API). This design allows Medicsight’s visualization partner companies to seamlessly integrate the CAD capability into their existing product offerings. Medicsight’s multiple integration partner approach means that many radiologists are able to access the valuable additional information supplied by CAD on a workstation with which they are familiar.

The LungCAD™ software is integrated into the leading 3D advanced visualization imaging workstations of the following partner distribution networks:

TeraRecon Inc. of San Mateo, CA.

Viatronix Inc. of Stony Brook, NY.

3mensio Medical Imaging BV of Bilthoven, Netherlands.

Medicsight LungCAD™ – allowing digital radiography to live up to its promise

Medicsight’s software solution, addresses many of the difficulties being experienced by radiologists and is set to transform lung cancer detection using CT by helping improve productivity of detection of nodules in the lung.

LungCAD™ is an image analysis software tool designed to be used with scanned images from CT examinations and assists radiologists in detecting and measuring potential lung nodules. It uses a series of filters deployed against image CT data. The filters highlight spherical regions as small as 5mm (about the size of a small pea) allowing radiologists to examine these areas in 3D in order to judge their potential as true nodules.

Radiologists are also able to manually highlight any irregularities for closer inspection. Once suspect nodules are found, the software can determine the shape and features, precisely identifying the boundaries and showing the lesion in 3D, with automatic diameter and volume measurements. This allows radiologists to accurately review and monitor nodules over time.

Medicsight LungCAD™:

Incorporates a validated CAD algorithm.

Automatically highlights specific regions of interest (ROI).

Allows accurate measurement of candidate nodules.

Validated against one of the world’s largest CT scan databases

Unlike some commercially available CAD systems, that have only been validated on a small number of patients, Medicsight’s LungCAD™ has been validated against one of the largest and most population diverse databases of verified CT scan data in the world. This solid validation of the LungCAD not only helps you accurately detect and analyse potential abnormalities within the lung, it gives you the reassurance and confidence to make a more reliable, accurate diagnosis.

Lung cancer is on the increase

Lung cancer is still the world’s most common cancer, accounting for over 17 per cent of all cancer deaths worldwide. In 2005, 1.3m people died of lung cancer worldwide (source: World Health Organisation). Furthermore, it is becoming more common, with a worldwide increase in incidence of 16 per cent since 1975, and more growth is predicted.

Early diagnosis is the best hope for cure

The prognosis for a patient diagnosed with lung cancer is dismal, with an average five year survival rate after diagnosis of only ten per cent. Prognosis varies widely depending on how advanced the cancer is – early lung cancer (stage I) can usually be cured by surgery, resulting in a five-year survival rate of 85 per cent compared with only one per cent for advanced (stage IV) disease. Stage I cancers, however cause no symptoms and currently only ten per cent of lung cancers are diagnosed at this early stage. It is hoped that lung cancer screening programmes being adopted worldwide may improve the statistic.

Growth rate is the key to early diagnosis

The challenge for radiologists is to find a way to distinguish malignant nodules from their much more common benign nodule counterparts. Growth rate is the key to distinguishing between the two; this can vary from a volume doubling time as short as four weeks for a very aggressive lung cancer, to significantly lower growth rates for benign nodules.

With a variety of treatment programmes available for symptomatic patients from drugs to radiotherapy, accurate tumour tracking is absolutely essential to recommend the correct course of treatment and monitor treatment effectiveness.

Screenshots of partner integrations

Medicsight LungCAD™ in Viatronix V3D™ workstation.

Medicsight LungCAD™ in TeraRecon Aquarius workstation.

Intended use statements

LungCAD™ API 3.0 is a non-invasive software application for image analysis designed to assist radiologists and other clinicians in searching for non-elongated objects with spherical elements within the lung using CT image data. These objects may be potential lesions (e.g. nodules). It is delivered as an API which can be integrated into other medical imaging solutions.

The API provides tools that allow the user to fully adjust a variety of characteristics used in highlighting objects. The identified objects are returned to the host medical imaging application, allowing the user to view the images with or without the highlighted view.

Additional imaging tools are also provided that perform segmenting and measuring objects of interest so that characteristics may be reported back to the user. Characteristic information on the segmented region includes the longest axis of the three dimensional region ('diameter'), and cubic volume (mm3).

When used by a skilled physician, this software provides information that may be useful in interpreting studies of the lung. The device should not be used to analyze CT lung scans acquired on patients who have had intravenous contrast administered. Patient management decisions should not be made solely on the results of LungCAD™ analysis.

It is essential that the radiologist examines all images in the CT examination, not only the images with identified objects.