Forecasting Time Series in Healthcare With Gaussian Processes and Dynamic Time Warping Based Subset Selection.

Modelling real-world time series can be challenging in the absence of sufficient data. Limited data in healthcare, can arise for several reasons, namely when the number of subjects is insufficient or the observed time series is irregularly sampled at a very low sampling frequency. This is especially true when attempting to develop personalised models, as there are typically few data points available for training from an individual subject. Furthermore, the need for early prediction (as is often the case in healthcare applications) amplifies the problem of limited availability of data. This article proposes a novel personalised technique that can be learned in the absence of sufficient data for early prediction in time series. Our novelty lies in the development of a subset selection approach to select time series that share temporal similarities with the time series of interest, commonly known as the test time series. Then, a Gaussian processes-based model is learned using the existing test data and the chosen subset to produce personalised predictions for the test subject. We will conduct experiments with univariate and multivariate data from real-world healthcare applications to show that our strategy outperforms the state-of-the-art by around 20%.

Feature selection for unbiased imputation of missing values: A case study in healthcare.

Datasets in healthcare are plagued with incomplete information. Imputation is a common method to deal with missing data where the basic idea is to substitute some reasonable guess for each missing value and then continue with the analysis as if there were no missing data. However unbiased predictions based on imputed datasets can only be guaranteed when the missing mechanism is completely independent of the observed or missing data. Often, this promise is broken in healthcare dataset acquisition due to unintentional errors or response bias of the interviewees. We highlight this issue by studying extensively on an annual health survey dataset on infant mortality prediction and provide a systematic testing for such assumption. We identify such biased features using an empirical approach and show the impact of wrongful inclusion of these features on the predictive performance.Clinical relevance- We show that blind analysis along with plug and play imputation of healthcare data is a potential pitfall that clinicians and researchers want to avoid in finding important markers of disease.

Gestational weight gain prediction using privacy preserving federated learning.

Gestational weight gain prediction in expecting women is associated with multiple risks. Manageable interventions can be devised if the weight gain can be predicted as early as possible. However, training the model to predict such weight gain requires access to centrally stored privacy sensitive weight data. Federated learning can help mitigate this problem by sending local copies of trained models instead of raw data and aggregate them at the central server. In this paper, we present a privacy preserving federated learning approach where the participating users collaboratively learn and update the global model. Furthermore, we show that this model updation can be done incrementally without having the need to store the local updates eternally. Our proposed model achieves a mean absolute error of 4.455 kgs whilst preserving privacy against 2.572 kgs achieved in a centralised approach utilising individual training data until day 140.Clinical relevance- Privacy preserving training of machine learning algorithm for early gestational weight gain prediction with minor tradeoff to performance.

PREgDICT : Early prediction of gestational weight gain for pregnancy care.

Excessive or inadequate Gestational Weight Gain (GWG) is considered to not only put the mothers, but also the infants at increased risks with a number of adverse outcomes. In this paper, we use self-reported weight measurements from the early days of pregnancy to predict and classify the end-of-pregnancy weight gain into an underweight, normal or obese category in accordance with the Institute of Medicine recommended guidelines. Self-reported weight measurements suffer from issues such as lack of enough data and non-uniformity. We propose and compare two novel parametric and non-parametric approaches that utilise self-training data along with population data to tackle limited data availability. We, dynamically find the subset of closest time series from the population weight-gain data to a given subject. Then, a non-parametric Gaussian Process (GP) regression model, learnt on the selected subset is used to forecast the self-reported weight measurements of given subject. Our novel approach produces mean absolute error (MAE) of 2.572 kgs in forecasting end-of-pregnancy weight gain and achieves weight-category-classification accuracy of 63.75% mid-way through the pregnancy, whereas a state-of-the-art approach is only 53.75% accurate and produces high MAE of 16.22 kgs. Our method ensures reliable prediction of the end-of-pregnancy weight gain using few data points and can assist in early intervention that can prevent gaining or losing excessive weight during pregnancy.

Detection of atrial fibrillation and other abnormal rhythms from ECG using a multi-layer classifier architecture.

OBJECTIVE: Atrial fibrillation (AF) and other types of abnormal heart rhythm are related to multiple fatal cardiovascular diseases that affect the quality of human life. Hence the development of an automated robust method that can reliably detect AF, in addition to other non-sinus and sinus rhythms, would be a valuable addition to medicine. The present study focuses on developing an algorithm for the classification of short, single-lead electrocardiogram (ECG) recordings into normal, AF, other abnormal rhythms and noisy classes. APPROACH: The proposed classification framework presents a two-layer, three-node architecture comprising binary classifiers. PQRST markers are detected on each ECG recording, followed by noise removal using a spectrogram power based novel adaptive thresholding scheme. Next, a feature pool comprising time, frequency, morphological and statistical domain ECG features is extracted for the classification task. At each node of the classification framework, suitable feature subsets, identified through feature ranking and dimension reduction, are selected for use. Adaptive boosting is selected as the classifier for the present case. The training data comprises 8528 ECG recordings provided under the PhysioNet 2017 Challenge. F1 scores averaged across the three non-noisy classes are taken as the performance metric. MAIN RESULT: The final five-fold cross-validation score achieved by the proposed framework on the training data has high accuracy with low variance (0.8254 [Formula: see text] 0.0043). SIGNIFICANCE: Further, the proposed algorithm has achieved joint first place in the PhysioNet/Computing in Cardiology Challenge 2017 with a score of 0.83 computed on a hidden test dataset.

Pattern Analysis in Physiological Pulsatile Signals: An Aid to Personalized Healthcare.

We present a system to analyze patterns inside pulsatile signals and discover repetitions inside signals. We measure dominance of the repetitions using morphology and discrete nature of the signals by exploiting machine learning and information theoretic concepts. Patterns are represented as combinations of the basic features and derived features. Consistency of discovered patterns identifies state of physiological stability which varies from one individual to another. Hence it has immense impact on deriving the accurate physiological parameters for personalized health analytics. Proposed mechanism discovers the regular and irregular patterns by performing extensive analysis on several real life cardiac data sets. We have achieved more than 90% accuracy in identifying irregular patterns using our proposed method.

Analysis of phonocardiogram signals through proactive denoising using novel self-discriminant learner.

Phonocardiogram (PCG) records heart sound and murmurs, which contains significant information of cardiac health. Analysis of PCG signal has the potential to detect abnormal cardiac condition. However, the presence of noise and motion artifacts in PCG hinders the accuracy of clinical event detection. Thus, noise detection and elimination are crucial to ensure accurate clinical analysis. In this paper, we present a robust denoising technique, Proclean that precisely detects the noisy PCG signal through pattern recognition, and statistical learning. We propose a novel self-discriminant learner that ensures to obtain distinct feature set to distinguish clean and noisy PCG signals without human-in-loop. We demonstrate that our proposed denoising leads to higher accuracy in subsequent clinical analytics for medical investigation. Our extensive experimentations with publicly available MIT-Physionet datasets show that we achieve more than 85% accuracy for noisy PCG signal detection. Further, we establish that physiological abnormality detection improves by more than 20%, when our proposed denoising mechanism is applied.

An unsupervised learning for robust cardiac feature derivation from PPG signals.

We propose here derivation algorithms for physiological parameters like beat start point, systolic peak, pulse duration, peak-to-peak distance related to heart rate, dicrotic minima, diastolic peak from Photoplethysmogram (PPG) signals robustly. Our methods are based on unsupervised learning mainly following morphology as well as discrete nature of the signal. Statistical learning has been used as a special aid to infer most probable feature values mainly to cope up with presence of noise, which is assumed to be insignificant compared to signal values at each investigation window. Performance of the proposed method is found to be better than other standard methods, yielding precision and sensitivity more than 97% obtained from three real life data sets.