15 Oct 2012
Investment of €10 million proves that image analysis techniques have potential to transform cancer diagnosis, company believes.Definiens, a Munich-based developer of image analysis techniques and specialists in the growing field of digital pathology, recently closed a private funding round worth €10 million.
The funds will primarily be used for commercial expansion of its current business, as well as development and commercialization of products for use in clinical digital pathology procedures.
Gilde Healthcare Partners, a new investor, led the funding round, which also saw participation from existing investors and Definiens management.
Digital pathology involves the acquisition, management, and interpretation of pathology information generated from digitized tissue specimens and scanned microscopy images. As pathologists embrace digital workflows, digital pathology is rapidly becoming an essential technology for increasing productivity and improving treatment decisions and patient care, according to Definiens.
"Digitization of data is increasing across the whole field of medicine," noted Markus Rinecker of Definiens speaking to Optics.org. "Most pictures in radiology are digitized; MRIs and CT scans all are digital; the same is becoming true for the patient records and other kinds of visual information. Pathology has been one of the remaining 'analog islands'."
One reason for that has been the formidable demands on the hardware involved, since digitizing pathology images with a high enough resolution to allow meaningful analysis leads to very large files.
"The images can be enormous, with single files in the gigapixel or gigabyte arena," agreed Rinecker. "And you are unlikely to have one single picture per patient, so a patient's complete data set can easily total 1 terabyte of data. This is one reason why the field has taken a long time to evolve. Only in the last three years or so have digital scanners able to perform the task become available, from vendors such as Leica and Hamamatsu. Philips has introduced its own scanner, GE is likely to do the same and are investing heavily in the field, so big companies are starting to chase this market."
A developing value chain
Once adequately scanned pathology images are available, whether from standard or confocal microscopy or other sources, Definiens digital pathology image analysis is designed to detect regions of interest. These can include specific morphologies, rare events, cells, and sub-cellular components.
"Our approach is completely different from our competitors," said Rinecker. "We analyze the pictures in a similar way to how a human being would look at it, examining objects in context, rather than strictly looking for pixels and patterns and matches. This makes the image analysis much more robust, and Definiens holds patents relating to this approach."
Cancer diagnosis is a particular focus, and an area singled out for mention in the announcement of the recent funding. "Digital Pathology combined with the right image analysis tools has the potential to transform cancer diagnosis," commented Janke Dittmer, partner at Gilde Healthcare Partners, at the time. Dittmer has now joined Definiens' supervisory board following the investment round.
"There is a new value chain evolving in digital pathology," said Rinecker. "Definiens has worked in the life sciences and non-regulated space previously, but is now entering the regulated space to work with clinics and hospitals, building the use of digital pathology in patient care. Digitization of pathology images is becoming a commodity, and the interesting activity now is in what you can do with this digitized data."