Extracting the cerebral anterior vessel tree of patients with an intracranial large vessel occlusion (LVO) is relevant to investigate potential biomarkers that can contribute to treatment decision making. The purpose of our work is to develop a method that can achieve this from routinely acquired computed tomography angiography (CTA) and computed tomography perfusion (CTP) images. To this end, we regard the anterior vessel tree as a set of bifurcations and connected centerlines. The method consists of a proximal policy optimization (PPO) based deep reinforcement learning (DRL) approach for tracking centerlines, a convolutional neural network based bifurcation detector, and a breadth-first vessel tree construction approach taking the tracking and bifurcation detection results as input. We experimentally determine the added values of various components of the tracker. Both DRL vessel tracking and CNN bifurcation detection were assessed in a cross validation experiment using 115 subjects. The anterior vessel tree formation was evaluated on an independent test set of 25 subjects, and compared to interobserver variation on a small subset of images. The DRL tracking result achieves a median overlapping rate until the first error (1.8 mm off the reference standard) of 100, [46, 100] % on 8032 vessels over 115 subjects. The bifurcation detector reaches an average recall and precision of 76% and 87% respectively during the vessel tree formation process. The final vessel tree formation achieves a median recall of 68% and precision of 70%, which is in line with the interobserver agreement.

We consider the problem of detecting commissioning changes in connected lighting systems. Commissioning changes can occur due to repositioning of luminaires/sensors or space renovation. This results in incorrect commissioning mapping of devices to areas, thereby impacting analysis and interpretation of data from such devices. We propose an automated method to detect changes in commissioning mapping using occupancy sensor data. We use similarity features across occupancy sensors and employ a random forest binary classifier to detect changes. The proposed method is evaluated using data from a simulated office environment and an experimental testbed, and is shown to have high accuracy.

We consider the problem of classifying anomalous occupancy sensor behavior in connected indoor lighting systems. Anomalous occupancy sensor behavior may occur in the form of either a high number of false alarms (type-1 anomalies) or missed detections (type-2 anomalies). Two anomaly discovery scenarios are considered: one, in which no anomalies exist post-deployment, and two, in which both anomaly types are found together with normally functional sensors. We address the problem of classifying anomalies that may occur subsequently using a machine learning approach. Under scenario 1, we consider a one class random forest classifier to determine whether an occupancy signal is normal or not. In scenario 2, we consider a supervised random forest classifier to determine whether the detection signal of an occupancy sensor is normal, or exhibits type-1 or type-2 anomalies. We devise occupancy signal features in time and frequency domains to perform 2-class classification in scenario 1, and 3-class classification in scenario 2. The proposed method is evaluated using motion sensor data from an office building, and is shown to have higher true positive rate and a lower false positive rate in comparison to an unsupervised k -means method and a random forest classifier with a single signal energy feature.

Visible light communications (VLC) with light-emitting diodes (LEDs) has attracted applications, such as data communications, lighting control, and light interaction. In this paper, we propose a system by which two LED devices are used for two-way VLC while also providing illumination. We consider Manchester encoding with on-off keying for transmission. A reception scheme in which the LEDs themselves are used as receivers by sensing in the OFF periods is considered. A synchronization scheme to achieve frame- and symbol-level synchronization is also proposed. A prototype of the proposed system is designed and the communication performance is evaluated.

Visible light communications (VLC) with light emitting diodes (LEDs) has attracted applications like data communications, lighting control and interaction. In this paper, we propose a system in which two LED devices are used for two-way VLC while also providing illumination. We consider a redundant run-length limited encoding for data transmission. A reception scheme in which the LEDs themselves are used as receivers by sensing in the off period is considered. A prototype of the proposed system is designed and the communication performance is evaluated. An advantage of the proposed system is that no additional light sensors are needed as the LEDs themselves are used as communication transceivers.

Luminaires based on light emitting diode (LEDs) offer ease of control for light adaptation. In particular, adapting light output of the LED luminaires to daylight leads to energy-efficient lighting solutions. In a daylight-adaptive lighting system, the dimming levels are determined such that the artificial illuminance together with the daylight illuminance meets specified illuminance constraints. Closed-loop illumination control is traditionally used based on feedback obtained by measurements using light sensors, adding to cost and installation complexity. In this paper, we present a solution wherein the LED luminaire is used as a light source as well as a light sensor. We propose a protocol wherein the LEDs are driven in forward-bias by a pulse width modulation signal, thus emitting light, and in reverse-bias over a duration where they sense light. Under the proposed solution, the sensed illuminance is solely the daylight contribution. This enables use of simple open-loop illumination control schemes for daylight adaptation. We demonstrate the proof-of-principle with an LED array prototype.