Search

BackgroundIn sickle cell disease (SCD), unpredictable episodes of acute severe pain, known as vaso-occlusive crises (VOC), disrupt school, work activities and family life and ultimately lead to multiple hospitalizations. The ability to predict VOCs would allow a timely and adequate intervention. The first step towards this ultimate goal is to use patient-friendly and accessible technology to collect relevant data that helps infer a patient's pain experience during VOC. This study aims to: (1) determine the feasibility of remotely monitoring with a consumer wearable during hospitalization for VOC and up to 30 days after discharge, and (2) evaluate the accuracy of pain prediction using machine learning models based on physiological parameters measured by a consumer wearable.MethodsPatients with SCD (≥18 years) who were admitted for a vaso-occlusive crisis were enrolled at a single academic center. Participants were instructed to report daily pain scores (0–10) in a mobile app (Nanbar) and to continuously wear an Apple Watch up to 30 days after discharge. Data included heart rate (in rest, average and variability) and step count. Demographics, SCD genotype, and details of hospitalization including pain scores reported to nurses, were extracted from electronic medical records. Physiological data from the wearable were associated with pain scores to fit 3 different machine learning classification models. The performance of the machine learning models was evaluated using: accuracy, F1, root-mean-square error and area under the receiver-operating curve.ResultsBetween April and June 2022, 19 patients (74% HbSS genotype) were included in this study and followed for a median time of 28 days [IQR 22–34], yielding a dataset of 2,395 pain data points. Ten participants were enrolled while hospitalized for VOC. The metrics of the best performing model, the random forest model, were micro-averaged accuracy of 92%, micro-averaged F1-score of 0.63, root-mean-square error of 1.1, and area under the receiving operating characteristic curve of 0.9.ConclusionOur random forest model accurately predicts high pain scores during admission for VOC and after discharge. The Apple Watch was a feasible method to collect physiologic data and provided accuracy in prediction of pain scores.

BackgroundSickle cell disease (SCD) is the most common inherited blood disorder affecting millions of people worldwide. Most patients with SCD experience repeated, unpredictable episodes of severe pain. These pain episodes are the leading cause of emergency department visits among patients with SCD and may last for several weeks. Arguably, the most challenging aspect of treating pain episodes in SCD is assessing and interpreting a patient’s pain intensity level. ObjectiveThis study aims to learn deep feature representations of subjective pain trajectories using objective physiological signals collected from electronic health records. MethodsThis study used electronic health record data collected from 496 Duke University Medical Center participants over 5 consecutive years. Each record contained measures for 6 vital signs and the patient’s self-reported pain score, with an ordinal range from 0 (no pain) to 10 (severe and unbearable pain). We also extracted 3 features related to medication: medication type, medication status (given or applied, or missed or removed or due), and total medication dosage (mg/mL). We used variational autoencoders for representation learning and designed machine learning classification algorithms to build pain prediction models. We evaluated our results using an accuracy and confusion matrix and visualized the qualitative data representations. ResultsWe designed a classification model using raw data and deep representational learning to predict subjective pain scores with average accuracies of 82.8%, 70.6%, 49.3%, and 47.4% for 2-point, 4-point, 6-point, and 11-point pain ratings, respectively. We observed that random forest classification models trained on deep represented features outperformed models trained on unrepresented data for all pain rating scales. We observed that at varying Likert scales, our models performed better when provided with medication data along with vital signs data. We visualized the data representations to understand the underlying latent representations, indicating neighboring representations for similar pain scores with a higher resolution of pain ratings. ConclusionsOur results demonstrate that medication information (the type of medication, total medication dosage, and whether the medication was given or missed) can significantly improve subjective pain prediction modeling compared with modeling with only vital signs. This study shows promise in data-driven estimated pain scores that will help clinicians with additional information about the patient’s condition, in addition to the patient’s self-reported pain scores.

Engagement is enhanced by the ability to access the state of flow during a task, which is described as a full immersion experience. We report two studies on the efficacy of using physiological data collected from a wearable sensor for the automated prediction of flow. Study 1 took a two-level block design where activities were nested within its participants. A total of five participants were asked to complete 12 tasks that aligned with their interests while wearing the Empatica E4 sensor. This yielded 60 total tasks across the five participants. In a second study representing daily use of the device, a participant wore the device over the course of 10 unstructured activities over 2 weeks. The efficacy of the features derived from the first study were tested on these data. For the first study, a two-level fixed effects stepwise logistic regression procedure indicated that five features were significant predictors of flow. In total, two were related to skin temperature (median change with respect to the baseline and skewness of the temperature distribution) and three were related to acceleration (the acceleration skewness in the x and y directions and the kurtosis of acceleration in the y direction). Logistic regression and naïve Bayes models provided a strong classification performance (AUC > 0.7, between-participant cross-validation). For the second study, these same features yielded a satisfactory prediction of flow for the new participant wearing the device in an unstructured daily use setting (AUC > 0.7, leave-one-out cross-validation). The features related to acceleration and skin temperature appear to translate well for the tracking of flow in a daily use environment.

Computational fluid dynamics (CFD) is widely employed to predict hemodynamic characteristics in arterial models, while not friendly to clinical applications due to the complexity of numerical simulations. Alternatively, this work proposed a framework to estimate hemodynamics in vessels based on angiography images using machine learning (ML) algorithms. First, the iodine contrast perfusion in blood was mimicked by a flow of dye diffusing into water in the experimentally validated CFD modeling. The generated projective images from simulations imitated the counterpart of light passing through the flow field as an analogy of X-ray imaging. Thus, the CFD simulation provides both the ground truth velocity field and projective images of dye flow patterns. The rough velocity field was estimated using the optical flow method (OFM) based on 53 projective images. ML training with least absolute shrinkage, selection operator and convolutional neural network was conducted with CFD velocity data as the ground truth and OFM velocity estimation as the input. The performance of each model was evaluated based on mean absolute error and mean squared error, where all models achieved or surpassed the criteria of 3 × 10−3 and 5 × 10−7 m/s, respectively, with a standard deviation less than 1 × 10−6 m/s. Finally, the interpretable regression and ML models were validated with over 613 image sets. The validation results showed that the employed ML model significantly reduced the error rate from 53.5% to 2.5% on average for the v-velocity estimation in comparison with CFD. The ML framework provided an alternative pathway to support clinical diagnosis by predicting hemodynamic information with high efficiency and accuracy.

The ability to monitor mental effort during a task using a wearable sensor may improve productivity for both work and study. The use of the electrodermal activity (EDA) signal for tracking mental effort is an emerging area of research. Through analysis of over 92 h of data collected with the Empatica E4 on a single participant across 91 different activities, we report on the efficacy of using EDA features getting at signal intensity, signal dispersion, and peak intensity for prediction of the participant’s self-reported mental effort. We implemented the logistic regression algorithm as an interpretable machine learning approach and found that features related to signal intensity and peak intensity were most useful for the prediction of whether the participant was in a self-reported high mental effort state; increased signal and peak intensity were indicative of high mental effort. When cross-validated by activity moderate predictive efficacy was achieved (AUC = 0.63, F1 = 0.63, precision = 0.64, recall = 0.63) which was significantly stronger than using the model bias alone. Predicting mental effort using physiological data is a complex problem, and our findings add to research from other contexts showing that EDA may be a promising physiological indicator to use for sensor-based self-monitoring of mental effort throughout the day. Integration of other physiological features related to heart rate, respiration, and circulation may be necessary to obtain more accurate predictions.