No abstract

The World Health Organization (WHO) estimated in 2012 that globally 285 million people have low vision or are blind. A similar number of an estimated 253 million people was reported by the World Blind Union (WBU) in 2015. Most visually impaired or blind persons rely on white canes or service dogs to improve their daily mobility. However, the use of white canes limits the detection distance of objects to the length of the cane or approximately 1.5m. On the other hand, the number of guide dogs is limited, and the training of a service animal is very expensive. Therefore, there is a real need for systems that increase the mobility of the visually impaired, but at the same time are affordable, comfortable, and easy to use.This article presents a complete assistive navigation system, called SENSATION: Sidewalk Environment Detection System for Assistive Navigation, which is inexpensive, easy to wear and does not interfere with the additional usage of a white cane. It is a standalone system that does not depend on cloud computing and uses deep learning models on the device for the immediate detection of the environment along with image segmentation and an algorithm to correct drifting while walking.

The purpose of this work was to automatically classify BPE using T1-weighted subtraction volumes and diffusion-weighted imaging volumes in breast MRI. The dataset consisted of 621 routine breast MRI examination acquired at University Hospital Erlangen. 2D MIP and 3D T1-subtraction volumes were used for the automatic detection of BPE classes. Multi-b-value DWI (up to1500s/mm2) DWI images were used for automatic prediction. ResNet and DenseNet models were used for 2D and 3D data respectively. The study demonstrated an AUROC of 0.8107 on the test set using the T1-subtraction volumes. With DWI volumes, a slightly decreased AuROC of 0.78 was achieved.