Received Signal Strength Fingerprinting-Based Indoor Location Estimation Employing Machine Learning.

The fingerprinting technique is a popular approach to reveal location of persons, instruments or devices in an indoor environment. Typically based on signal strength measurement, a power level map is created first in the learning phase to align with measured values in the inference. Second, the location is determined by taking the point for which the recorded received power level is closest to the power level actually measured. The biggest limit of this technique is the reliability of power measurements, which may lack accuracy in many wireless systems. To this end, this work extends the power level measurement by using multiple anchors and multiple radio channels and, consequently, considers different approaches to aligning the actual measurements with the recorded values. The dataset is available online. This article focuses on the very popular radio technology Bluetooth Low Energy to explore the possible improvement of the system accuracy through different machine learning approaches. It shows how the accuracy-complexity trade-off influences the possible candidate algorithms on an example of three-channel Bluetooth received signal strength based fingerprinting in a one dimensional environment with four static anchors and in a two dimensional environment with the same set of anchors. We provide a literature survey to identify the machine learning algorithms applied in the literature to show that the studies available can not be compared directly. Then, we implement and analyze the performance of four most popular supervised learning techniques, namely k Nearest Neighbors, Support Vector Machines, Random Forest, and Artificial Neural Network. In our scenario, the most promising machine learning technique being the Random Forest with classification accuracy over 99%.

Contrast-enhanced MR angiography utilizing parallel acquisition techniques in renal artery stenosis detection.

Significant renal artery stenosis (RAS) is a potentially curable cause of renovascular hypertension and/or renal impairment. It is caused by either atherosclerosis or fibromuscular dysplasia. Correct and timely diagnosis remains a diagnostic challenge. MR angiography (MRA) as a minimally invasive method seems to be suitable for RAS detection, however, its diagnostic value widely differs in the literature (sensitivity 62-100% and specificity 75-100%). The aim of our prospective study was to compare the diagnostic value of contrast-enhanced MRA utilizing parallel acquisition techniques in the detection of significant RAS with digital subtraction angiography (DSA). A total of 78 hypertensive subjects with suspected renal artery stenosis were examined on a 1.5 Tesla MR system using a body array coil. Bolus tracking was used to monitor the arrival of contrast agent to the abdominal aorta. The MRA sequence parameters were as follows: TR 3.7 ms; TE 1.2 ms; flip angle 25 degrees; acquisition time 18s; voxel size 1.1 mm x1.0 mm x 1.1 mm; centric k-space sampling; parallel acquisition technique with acceleration factor of 2 (GRAPPA). Renal artery stenosis of 60% and more was considered hemodynamically significant. The results of MRA were compared to digital subtraction angiography serving as a standard of reference. Sensitivity and specificity of MRA in the detection of hemodynamically significant renal artery stenosis were 90% and 96%, respectively. Prevalence of RAS was 39% in our study population. Contrast-enhanced MRA with high spatial resolution offers sufficient sensitivity and specificity for screening of RAS.