Search

Abstract Drug response prediction is hampered by uncertainty in the measures of response and selection of doses. In this study, we propose a probabilistic multi-output model to simultaneously predict all dose–responses and uncover their biomarkers. By describing the relationship between genomic features and chemical properties to every response at every dose, our multi-output Gaussian Process (MOGP) models enable assessment of drug efficacy using any dose–response metric. This approach was tested across two drug screening studies and ten cancer types. Kullback-leibler divergence measured the importance of each feature and identified EZH2 gene as a novel biomarker of BRAF inhibitor response. We demonstrate the effectiveness of our MOGP models in accurately predicting dose–responses in different cancer types and when there is a limited number of drug screening experiments for training. Our findings highlight the potential of MOGP models in enhancing drug development pipelines by reducing data requirements and improving precision in dose–response predictions.

Adjusting the stimulation parameters is a challenge in deep brain stimulation (DBS) therapy due to the vast number of different configurations available. As a result, systems based on the visualization of the volume of tissue activated (VTA) produced by a particular stimulation setting have been developed. However, the medical specialist still has to search, by trial and error, for a DBS set-up that generates the desired VTA. Therefore, our goal is developing a DBS parameter tuning strategy for current clinical devices that allows defining a target VTA under biophysically viable constraints. We propose a machine learning approach that allows estimating the DBS parameter values for a given VTA, which comprises two main stages: i) A K-nearest neighbors-based deformation to define a target VTA preserving biophysically viable constraints. ii) A parameter estimation stage that consists of a data projection using metric learning to highlight relevant VTA properties, and a regression/classification algorithm to estimate the DBS parameters that generate the target VTA. Our methodology allows setting a biophysically compliant target VTA and accurately predicts the required configuration of stimulation parameters. Also, the performance of our approach is stable for both isotropic and anisotropic tissue conductivities. Furthermore, the computational time of the trained system is acceptable for real-world implementations.

Identification of brain signals from microelectrode recordings (MER) is a key procedure during deep brain stimulation (DBS) applied in Parkinson’s disease patients. The main purpose of this research work is to identify with high accuracy a brain structure called subthalamic nucleus (STN), since it is the target structure where the DBS achieves the best therapeutic results. To do this, we present an approach for optimal representation of MER signals through method of frames. We obtain coefficients that minimize the Euclidean norm of order two. From optimal coefficients, we extract some features from signals combining the wavelet packet and cosine dictionaries. For a comparison frame with the state of the art, we also process the signals using the discrete wavelet transform (DWT) with several mother functions. We validate the proposed methodology in a real data base. We employ simple supervised machine learning algorithms, as the K-Nearest Neighbors classifier (K-NN), a linear Bayesian classifier (LDC) and a quadratic Bayesian classifier (QDC). Classification results obtained with the proposed method improves significantly the performance of the DWT. We achieve a positive identification of the STN superior to 97,6%. Identification outcomes achieved by the MOF are highly accurate, as we can potentially get a false positive rate of less than 2% during the DBS.

Heart rate variability (HRV) has received considerable attention for many years, since it provides a quantitative marker for examining the sinus rhythm modulated by the autonomic nervous system (ANS). The ANS plays an important role in clinical and physiological fields. HRV analysis can be performed by computing several time and frequency domain measurements. However, the computation of such measurements can be affected by the presence of artifacts or ectopic beats in the electrocardiogram (ECG) recording. This is particularly true for ECG recordings from Holter monitors. The aim of this work was to study the performance of several robust Kalman filters for artifact correction in Inter-beat (RR) interval time series. For our experiments, two data sets were used: the first data set included 10 RR interval time series from a realistic RR interval time series generator. The second database contains 10 sets of RR interval series from five healthy patients and five patients suffering from congestive heart failure. The standard deviation of the RR interval was computed over the filtered signals. Results were compared with a state of the art processing software, showing similar values and behavior. In addition, the proposed methods offer satisfactory results in contrast to standard Kalman filtering.

In this paper we present a new methodology for detecting sound events in music signals using Gaussian Processes. Our method firstly takes a time-frequency representation, i.e. the spectrogram, of the input audio signal. Secondly the spectrogram dimension is reduced translating the linear Hertz frequency scale into the logarithmic Mel frequency scale using a triangular filter bank. Finally every short-time spectrum, i.e. every Mel spectrogram column, is classified as “Event” or “Not Event” by a Gaussian Processes Classifier. We compare our method with other event detection techniques widely used. To do so, we use MATLAB® to program each technique and test them using two datasets of music with different levels of complexity. Results show that the new methodology outperforms the standard approaches, getting an improvement by about 1.66 % on the dataset one and 0.45 % on the dataset two in terms of F-measure.