An AI toolbox for pathology

Nizan Sargiv, the co-founder of the DeePathology Studio from Israel, stepped onto the stage in trade fair hall 13 to represent his company. “We truly believe in what we do” was their confident motto, announcing that they intend to bring artificial intelligence (AI) for research and diagnostics into every single pathology laboratory as a result. Pathologists and researchers can produce AI solutions for their work easily using the DeePathology-developed AI algorithms. This new tool essentially gives them an AI toolbox. They can thus teach their intelligent helpers various tasks quickly, such as counting certain cells and different types of objects, how to segment tumor regions and much more. Sagiv underscores this point: “The advantage of our system is that it’s super fast and super simple”. First, the user keeps training the AI software tool and determines how the searched-for cells or similar structures, such as amyloid plaques which show Alzheimer’s disease, look. After a short training period, the AI will proffer its own suggestions, and these become more accurate as its degree of experience increases. During their final presentation, Sagiv emphasized that access to large amounts of data is not necessary for this. Instead, the system works independently on a laptop and has been conceived so that it can be implemented in any pathology laboratory.

Easy blood analysis via smartphone

The German developer team from midge medical won 3rd place in the MEDICA APP COMPETITION 2019 with a simple hardware and software solution for taking and analyzing blood samples. To use this system, the user pricks their hand with a lancet. The blood that is produced seeps into a test strip integrated into a device. The test strip then changes color. This color is recorded using a smartphone, analyzed to give a diagnosis and digitalized. The lancet is a single-use product, thus ensuring that the required hygiene levels are maintained. “Though it’s very easy to use, our system is actually very complex”, explains Laura Henrich, the Chief Marketing Officer of midge medical. CRP levels in blood can be measured using this device. C-reactive protein is initially a non-specific laboratory parameter for acute inflammatory conditions that can be caused by an infectious or non-infectious pathology. Together with other patient information, the doctor can use this data to determine, for example, whether antibiotics should be given or not. The doctor can be physically present or they can provide consultation via telemedicine. This system could also expand the possibilities for pre-existing telemedical processes. Henrich gave the example of being able to control processes. If the CRP level responds appropriately, the therapy has probably been successful. There are also plans to include other blood parameters in the system’s measuring function. Specifically, the application could also be attractive to German health insurance bodies with a view to telediagnostics. Laura Henrich is equally optimistic in welcoming the German Digital Care Act (DVG) that has recently been enacted by the German government. This act is intended to enable doctors to prescribe apps.

Association Viewpoints Homepage