DEMA: A Deep Learning-Enabled Model for Non-Invasive Human Vital Signs Monitoring Based on Optical Fiber Sensing.

Optical fiber sensors are extensively employed for their unique merits, such as small size, being lightweight, and having strong robustness to electronic interference. The above-mentioned sensors apply to more applications, especially the detection and monitoring of vital signs in medical or clinical. However, it is inconvenient for daily long-term human vital sign monitoring with conventional monitoring methods under the uncomfortable feelings generated since the skin and devices come into direct contact. This study introduces a non-invasive surveillance system that employs an optical fiber sensor and advanced deep-learning methodologies for precise vital sign readings. This system integrates a monitor based on the MZI (Mach-Zehnder interferometer) with LSTM networks, surpassing conventional approaches and providing potential uses in medical diagnostics. This could be potentially utilized in non-invasive health surveillance, evaluation, and intelligent health care.

Compressed sensing framework for BCG signals based on the optical fiber sensor.

A compressed sensing (CS) framework is built for ballistocardiography (BCG) signals, which contains two parts of an optical fiber sensor-based heart monitoring system with a CS module and an end-to-end deep learning-based reconstruction algorithm. The heart monitoring system collects BCG data, and then compresses and transmits the data through the CS module at the sensing end. The deep learning-based algorithm reconstructs compressed data at the received end. To evaluate results, three traditional CS reconstruction algorithms and a deep learning method are adopted as references to reconstruct the compressed BCG data with different compression ratios (CRs). Results show that our framework can reconstruct signals successfully when the CR grows from 50% to 95% and outperforms other methods at high CRs. The mean absolute error (MAE) of the estimated heartbeat rate (HR) is lower than 1 bpm when the CR is below 95%. The proposed CS framework for BCG signals can be integrated into the IoMT system, which has great potential in health care for both medical and home use.

Deep learning-based ballistocardiography reconstruction algorithm on the optical fiber sensor.

Ballistocardiography (BCG) is a vibration signal related to cardiac activity, which can be obtained in a non-invasive way by optical fiber sensors. In this paper, we propose a modified generative adversarial network (GAN) to reconstruct BCG signals by solving signal fading problems in a Mach-Zehnder interferometer (MZI). Based on this algorithm, additional modulators and demodulators are not needed in the MZI, which reduces the cost and hardware complexity. The correlation between reconstructed BCG and reference BCG is 0.952 in test data. To further test the model performance, we collect special BCG signals including sinus arrhythmia data and post-exercise cardiac activities data, and analyze the reconstructed results. In conclusion, a BCG reconstruction algorithm is presented to solve the signal fading problem in the optical fiber interferometer innovatively, which greatly simplifies the BCG monitoring system.

Enhancing Clinical Predictive Modeling through Model Complexity-Driven Class Proportion Tuning for Class Imbalanced Data: An Empirical Study on Opioid Overdose Prediction.

Class imbalance issues are prevalent in the medical field and significantly impact the performance of clinical predictive models. Traditional techniques to address this challenge aim to rebalance class proportions. They generally assume that the rebalanced proportions are derived from the original data, without considering the intricacies of the model utilized. This study challenges the prevailing assumption and introduces a new method that ties the optimal class proportions to model complexity. This approach allows for individualized tuning of class proportions for each model. Our experiments, centered on the opioid overdose prediction problem, highlight the performance gains achieved by this approach. Furthermore, rigorous regression analysis affirms the merits of the proposed theoretical framework, demonstrating a statistically significant correlation between hyperparameters controlling model complexity and the optimal class proportions.

Sleep condition detection and assessment with optical fiber interferometer based on machine learning.

The prevalence of sleep disorders has increased because of the fast-paced and stressful modern lifestyle, negatively impacting the quality of human life and work efficiency. It is crucial to address sleep problems. However, the current practice of diagnosing sleep disorders using polysomnography (PSG) has limitations such as complexity, large equipment, and low portability, hindering its practicality for daily use. To overcome these challenges, in this article an optical fiber sensor is proposed as a viable solution for sleep monitoring. This device offers benefits like low power consumption, non-invasiveness, absence of interference, and real-time health monitoring. We introduce the sensor with an optical fiber interferometer to capture ballistocardiography (BCG) and electrocardiogram (ECG) signals from the human body. Furthermore, a new machine learning method is proposed for sleep condition detection. Experimental results demonstrate the superior performance of this architecture and the proposed model in monitoring and assessing sleep quality.

An integrated LSTM-HeteroRGNN model for interpretable opioid overdose risk prediction.

Opioid overdose (OD) has become a leading cause of accidental death in the United States, and overdose deaths reached a record high during the COVID-19 pandemic. Combating the opioid crisis requires targeting high-need populations by identifying individuals at risk of OD. While deep learning emerges as a powerful method for building predictive models using large scale electronic health records (EHR), it is challenged by the complex intrinsic relationships among EHR data. Further, its utility is limited by the lack of clinically meaningful explainability, which is necessary for making informed clinical or policy decisions using such models. In this paper, we present LIGHTED, an integrated deep learning model combining long short term memory (LSTM) and graph neural networks (GNN) to predict patients' OD risk. The LIGHTED model can incorporate the temporal effects of disease progression and the knowledge learned from interactions among clinical features. We evaluated the model using Cerner's Health Facts database with over 5 million patients. Our experiments demonstrated that the model outperforms traditional machine learning methods and other deep learning models. We also proposed a novel interpretability method by exploiting embeddings provided by GNNs to cluster patients and EHR features respectively, and conducted qualitative feature cluster analysis for clinical interpretations. Our study shows that LIGHTED can take advantage of longitudinal EHR data and the intrinsic graph structure of EHRs among patients to provide effective and interpretable OD risk predictions that may potentially improve clinical decision support.

A Multimodal Transformer: Fusing Clinical Notes with Structured EHR Data for Interpretable In-Hospital Mortality Prediction.

Deep-learning-based clinical decision support using structured electronic health records (EHR) has been an active research area for predicting risks of mortality and diseases. Meanwhile, large amounts of narrative clinical notes provide complementary information, but are often not integrated into predictive models. In this paper, we provide a novel multimodal transformer to fuse clinical notes and structured EHR data for better prediction of in-hospital mortality. To improve interpretability, we propose an integrated gradients (IG) method to select important words in clinical notes and discover the critical structured EHR features with Shapley values. These important words and clinical features are visualized to assist with interpretation of the prediction outcomes. We also investigate the significance of domain adaptive pretraining and task adaptive fine-tuning on the Clinical BERT, which is used to learn the representations of clinical notes. Experiments demonstrated that our model outperforms other methods (AUCPR: 0.538, AUCROC: 0.877, F1:0.490).

Ballistocardiography monitoring system based on optical fiber interferometer aided with heartbeat segmentation algorithm.

An optical fiber interferometer-based ballistocardiography (BCG) monitoring system aided with the IJK complex detection algorithm is proposed in this paper. A new phase modulation method based on a moving-coil transducer is developed to address the problem of signal fading in the optical fiber interferometer and keep the system in quadrature by the closed loop controller. As a result, a stable BCG signal without baseline drift can be obtained. This BCG monitor based on optical fiber interferometer using phase modulation method owns the advantages of compact, low-cost, portable, and user-friendly. In addition, an end-to-end modified U-net is developed to conduct pixel-wise classification in the BCG signal. This network can achieve high accuracy and shows its capability to segment IJK complex and body movement in the BCG signal. In conclusion, the proposed BCG monitoring system with IJK complex segmentation algorithm is potential and promising in healthcare applications.

Non-invasive human vital signs monitoring based on twin-core optical fiber sensors.

Twin-core fiber (TCF)-based sensor was proposed for non-invasive vital sign monitoring, including respiration and heartbeat. The TCF was homemade and the corresponding sensor was fabricated by sandwiching single-mode fiber (SMF) on both ends. The offset distance between SMF and TCF was optimized while the length of TCF was identified from preliminary vital sign measurement results. Then, the TCF-based sensor was attached under a mattress to realize non-invasive vital sign monitoring. Both respiration and heartbeat signal can be obtained simultaneously, which is consistent with the reference signals. For further application, post-exercise physiological activitity characterization were realized based on this vital sign monitoring system. In discussion, mode coupling in TCF was analyzed and utilized for curvature sensing with achieved sensitivity as high as 18 nm/m-1, which supported its excellent performance for vital signs monitoring. In conclusion, the TCF-based vital signs monitors can be a promising candidate for healthcare and biomedical applications.