Today, 2.7 million patients are treated for bladder cancer each year and there are more than 430 000 new cases which arise every year worldwide. Bladder cancer is a highly recurrent cancer and its 5 years’ survival rate is highly dependent on the precocity of the diagnosis. When bladder cancer progresses to an advanced stage it often requires invasive, debilitating and costly surgical procedures as well as chemotherapy. The method of reference for primary diagnosis is urinary cytology. While this method is efficient in detecting high grade with both a high specificity and sensitivity, it is unsuccessful for low grades, with a sensitivity below 20%.
Hence there is an unmet medical need for a non-invasive, highly sensitive method for all grades, which is also reproducible, reliable, easy to implement and in line with pathologist’s current practice. Early diagnosis would allow better patient care and help reduce healthcare costs, resulting from a late diagnosis.
Pathologists and labs are in need of an innovative solution allowing them to offer better care for their patients, preventing high grade cancer by diagnosing early ones. In addition, labs have the necessity to differentiate themselves between one another and VisioCyt® is giving them an opportunity to do so.
VisioCyt® is an image processing software based on a customized cytological staining process allowing the observation of cells in fluorescence and in white light. VisioCyt® provides both a metabolism evaluation in fluorescence and a cellular morphology analysis in white light, for the detection of tumor cells even at an early stage. To ensure the automaticity of the process, the manual method was adjusted to industrial slide preparation equipment already used in centralized laboratories, followed by a scanner to produce digital slides. The scanned slides are then processed using an image processing software, based on algorithms developed machine learning and deep learning.
VisioCyt® is the only test combining urinary cytology and artificial intelligence, with the following benefits: Reproducibility, speed and reliability, diagnostic performances even for low-grade cancers, continuous improvement of the test’s performance over time with Machine Learning and best performance/price ratio.
More detailed information can be found on Youtube.