Goals:

We are developing the product in three phases were,

• Provide doctors with a software aid that reports a probability of DVT vulnerability based on a patient's medical history using deep learning technologies (Phase 1).

• Data-driven analysis of ultrasound images and videos to accurately detect and segment out blood clots leading to DVT/PE (Phase 2).

• Full cardio-vascular simulations of specific patients to probabilistically track the clot’s likely progress within the human body, in order to aid doctor’s diagnostics (Phase 3).

Vision

• To achieve a demonstrable reduction in the incidence of Deep Vein Thrombosis (DVT), and in the severity of its effects, such as pulmonary embolism (PE), once it has occurred.

Mission

• Combine state-of-the-art computational, simulation, and machine-learning technologies in DVT/PE avoidance and mitigation strategies.

• Reduce the incidence of misdiagnosis of DVT/PE (false positives and negatives) by using improved data processing methods and image analysis techniques.

• Broaden the accessibility of the results to a wider range of stakeholders, including those without specialist knowledge.