Healthcare

With respect to the healthcare domain, the objective is to implement and integrate edge-based DL implementations of medical image analysis algorithms on digital neuromorphic HW and evaluate their performance in multiple healthcare applications, including two medical imaging use cases (4.1 and 4.2), as well as personal health monitoring using wearable devices (use case 4.3).

For the imaging use case 4.1 Multi-modal image processing and device tracking in medical X-ray and ultrasound images, intelligent image analysis will lead to intuitive visualization of complex medical data and a simplified clinical workflow. Results will be demonstrated and benchmarked in minimally invasive treatment of patients with structural heart disease.

In medical imaging use case 4.2 Ultrasound acquisition or processing, Philips Research (PRE) will evaluate the potential of a spiking neural network to detect healthy and unhealthy lungs (e.g. pneumonia or Covid-19) from ultrasound lung images. Important validation criteria are high accuracy and power consumption.

For the healthcare devices use case 4.3 Glucose Monitoring, Eesy-Innovation  will develop a prototype for glucose monitoring able to transform high frequency signal into glucose level using novel neuromorphic algorithms, such as Spiking Neural Network (SNN). In this development power consumption and size will be key characteristics.