Quantifying the impact of homophily and influencer networks on song popularity prediction.

Forecasting the popularity of new songs has become a standard practice in the music industry and provides a comparative advantage for those that do it well. Considerable efforts were put into machine learning prediction models for that purpose. It is known that in these models, relevant predictive parameters include intrinsic lyrical and acoustic characteristics, extrinsic factors (e.g., publisher influence and support), and the previous popularity of the artists. Much less attention was given to the social components of the spreading of song popularity. Recently, evidence for musical homophily-the tendency that people who are socially linked also share musical tastes-was reported. Here we determine how musical homophily can be used to predict song popularity. The study is based on an extensive dataset from the last.fm online music platform from which we can extract social links between listeners and their listening patterns. To quantify the importance of networks in the spreading of songs that eventually determines their popularity, we use musical homophily to design a predictive influence parameter and show that its inclusion in state-of-the-art machine learning models enhances predictions of song popularity. The influence parameter improves the prediction precision (TP/(TP + FP)) by about 50% from 0.14 to 0.21, indicating that the social component in the spreading of music plays at least as significant a role as the artist's popularity or the impact of the genre.

Efficient full Monte Carlo modelling and multi-energy generative model development of an advanced X-ray device.

Monte Carlo (MC) simulations of X-ray image devices require splitting the simulation into two parts (i.e. the generation of x-rays and the actual imaging). The X-ray production remains unchanged for repeated imaging and can thus be stored in phase space (PhS) files and used for subsequent MC simulations. Especially for medical images these dedicated PhS files require a large amount of data storage, which is partly why Generative Adversarial Networks (GANs) were recently introduced. We enhanced the approach by a conditional GAN to model multiple energies using one network. This study compares the use of PhSs, GANs, and conditional GANs as photon source with measurements. An X-ray -based imaging system (i.e. ImagingRing) was modelled in this study. half-value layers (HVLs), focal spot, and Heel effect were measured for subsequent comparison. MC simulations were performed with GATE-RTion v1.0 considering the geometry and materials of the imaging system with vendor specific schematics. A traditional GAN model as well as the favourable conditional GAN was implemented for PhS generation. Results of the MC simulation were in agreement with the measurements regarding HVL, focal spot, and Heel effect. The conditional GAN performed best with a non-saturated loss function with R1 regularisation and gave similarly results as the traditional GAN approach. GANs proved to be superior to the PhS approach in terms of data storage and calculation overhead. Moreover, a conditional GAN enabled an energy interpolation to separate the network training process from the final required X-ray energies.

A Monte Carlo based scatter removal method for non-isocentric cone-beam CT acquisitions using a deep convolutional autoencoder.

The primary cone-beam computed tomography (CBCT) imaging beam scatters inside the patient and produces a contaminating photon fluence that is registered by the detector. Scattered photons cause artifacts in the image reconstruction, and are partially responsible for the inferior image quality compared to diagnostic fan-beam CT. In this work, a deep convolutional autoencoder (DCAE) and projection-based scatter removal algorithm were constructed for the ImagingRingTM system on rails (IRr), which allows for non-isocentric acquisitions around virtual rotation centers with its independently rotatable source and detector arms. A Monte Carlo model was developed to simulate (i) a non-isocentric training dataset of ≈1200 projection pairs (primary + scatter) from 27 digital head-and-neck cancer patients around five different virtual rotation centers (DCAENONISO), and (ii) an isocentric dataset existing of ≈1200 projection pairs around the physical rotation center (DCAEISO). The scatter removal performance of both DCAE networks was investigated in two digital anthropomorphic phantom simulations and due to superior performance only the DCAENONISO was applied on eight real patient acquisitions. Measures for the quantitative error, the signal-to-noise ratio, and the similarity were evaluated for two simulated digital head-and-neck patients, and the contrast-to-noise ratio (CNR) was investigated between muscle and adipose tissue in the real patient image reconstructions. Image quality metrics were compared between the uncorrected data, the currently implemented heuristic scatter correction data, and the DCAE corrected image reconstruction. The DCAENONISO corrected image reconstructions of two digital patient simulations showed superior image quality metrics compared to the uncorrected and corrected image reconstructions using a heuristic scatter removal. The proposed DCAENONISO scatter correction in this study was successfully demonstrated in real non-isocentric patient CBCT acquisitions and achieved statistically significant higher CNRs compared to the uncorrected or the heuristic corrected image data. This paper presents for the first time a projection-based scatter removal algorithm for isocentric and non-isocentric CBCT imaging using a deep convolutional autoencoder trained on Monte Carlo composed datasets. The algorithm was successfully applied to real patient data.

Modelling of the focal spot intensity distribution and the off-focal spot radiation in kilovoltage x-ray tubes for imaging.

X-ray tubes for medical applications typically generate x-rays by accelerating electrons, emitted from a cathode, with an interelectrode electric field, towards an anode target. X-rays are not emitted from one point, but from an irregularly shaped area on the anode, the focal spot. Focal spot intensity distributions and off-focal radiation negatively affect the imaging spatial resolution and broadens the beam penumbra. In this study, a Monte Carlo simulation model of an x-ray tube was developed to evaluate the spectral and spatial characteristics of off-focal radiation for multiple photon energies. Slit camera measurements were used to determine the horizontal and vertical intensity profiles of the small and the large focal spot of a diagnostic x-ray tube. First, electron beamlet weighting factors were obtained via an iterative optimization method to represent both focal spot sizes. These weighting factors were then used to extract off-focal spot radiation characteristics for the small and large focal spot sizes at 80, 100, and 120 kV. Finally, 120 kV simulations of a steel sphere (d = 4 mm) were performed to investigate image blurring with a point source, the small focal spot, and the large focal spot. The magnitude of off-focal radiation strongly depends on the anode size and the electric field coverage, and only minimally on the tube potential and the primary focal spot size. In conclusion, an x-ray tube Monte Carlo simulation model was developed to simulate focal spot intensity distributions and to evaluate off-focal radiation characteristics at several energies. This model can be further employed to investigate focal spot correction methods and to improve cone-beam CT image quality.