Interpretable machine learning for predicting the fate and transport of pentachlorophenol in groundwater.

Pentachlorophenol (PCP) is a commonly found recalcitrant and toxic groundwater contaminant that resists degradation, bioaccumulates, and has a potential for long-range environmental transport. Taking proper actions to deal with the pollutant accounting for the life cycle consequences requires a better understanding of its behavior in the subsurface. We recognize the huge potential for enhancing decision-making at contaminated groundwater sites with the arrival of machine learning (ML) techniques in environmental applications. We used ML to enhance the understanding of the dynamics of PCP transport properties in the subsurface, and to determine key hydrochemical and hydrogeological drivers affecting its transport and fate. We demonstrate how this complementary knowledge, provided by data-driven methods, may enable a more targeted planning of monitoring and remediation at two highly contaminated Swedish groundwater sites, where the method was validated. We evaluated 6 interpretable ML methods, 3 linear regressors and 3 non-linear (i.e., tree-based) regressors, to predict PCP concentration in the groundwater. The modeling results indicate that simple linear ML models were found to be useful in the prediction of observations for datasets without any missing values, while tree-based regressors were more suitable for datasets containing missing values. Considering that missing values are common in datasets collected during contaminated site investigations, this could be of significant importance for contaminated site planners and managers, ultimately reducing site investigation and monitoring costs. Furthermore, we interpreted the proposed models using the SHAP (SHapley Additive exPlanations) approach to decipher the importance of different drivers in the prediction and simulation of critical hydrogeochemical variables. Among these, sum of chlorophenols is of highest significance in the analyses. Setting that aside from the model, tetra chlorophenols, dissolved organic carbon, and conductivity found to be of highest importance. Accordingly, ML methods could potentially be used to improve the understanding of groundwater contamination transport dynamics, filling gaps in knowledge that remain when using more sophisticated deterministic modeling approaches.

M2D2: Maximum-Mean-Discrepancy Decoder for Temporal Localization of Epileptic Brain Activities.

Recent years have seen growing interest in leveraging deep learning models for monitoring epilepsy patients based on electroencephalographic (EEG) signals. However, these approaches often exhibit poor generalization when applied outside of the setting in which training data was collected. Furthermore, manual labeling of EEG signals is a time-consuming process requiring expert analysis, making fine-tuning patient-specific models to new settings a costly proposition. In this work, we propose the Maximum-Mean-Discrepancy Decoder (M2D2) for automatic temporal localization and labeling of seizures in long EEG recordings to assist medical experts. We show that M2D2 achieves 76.0% and 70.4% of F1-score for temporal localization when evaluated on EEG data gathered in a different clinical setting than the training data. The results demonstrate that M2D2 yields substantially higher generalization performance than other state-of-the-art deep learning-based approaches.

DynSus: Dynamic sustainability assessment in groundwater remediation practice.

Decision-making processes for clean-up of contaminated sites are often highly complex and inherently uncertain. It depends not only on hydrological and biogeochemical site variability, but also on the associated health, environmental, economic, and social impacts of taking, or not taking, action. These variabilities suggest that a dynamic framework is required for promoting sustainable remediation. For this, the decision support system DynSus is presented here for integrating a predeveloped contaminant fate and transport model with a sustainability assessment tool. Implemented within a system dynamics framework, the new tool uses model simulations to provide remediation scenario analysis and handling of uncertainty in various data. DynSus was applied to a site in south Sweden, contaminated with pentachlorophenol (PCP). Simulation scenarios were developed to enable a comparison between alternative remediation strategies and combinations of these. Such comparisons are provided for selected sustainability indicators and remediation performance (in terms of concentration at the recipient). This leads to identifying the most critical variables to ensure that sustainable solutions are chosen. Simulation results indicated that although passive practices, e.g., monitored natural attenuation, were more sustainable at first (5-7 years after beginning remediation measures), they failed to compete with more active practices, e.g., bioremediation, over the entire life cycle of the project (from the beginning of remedial action to achieving the target concentration at the recipient). In addition, statistical tools (clustering and genetic algorithms) were used to further assess the available hydrogeochemical data. Taken together, the results reaffirmed the suitability of the simple analytical framework that was implemented in the contaminant transport model. DynSus outcomes could therefore enable site managers to evaluate different scenarios more quickly and effectively for life cycle sustainability in such a complex and multidimensional problem.

Personalized seizure signature: An interpretable approach to false alarm reduction for long-term epileptic seizure detection.

OBJECTIVE: Long-term automatic detection of focal seizures remains one of the major challenges in epilepsy due to the unacceptably high number of false alarms from state-of-the-art methods. Our aim was to investigate to what extent a new patient-specific approach based on similarly occurring morphological electroencephalographic (EEG) signal patterns could be used to distinguish seizures from nonseizure events, as well as to estimate its maximum performance. METHODS: We evaluated our approach on >5500 h of long-term EEG recordings using two public datasets: the PhysioNet.org Children's Hospital Boston-Massachusetts Institute of Technology (CHB-MIT) Scalp EEG database and the EPILEPSIAE European epilepsy database. We visually identified a set of similarly occurring morphological patterns (seizure signature) seen simultaneously over two different EEG channels, and within two randomly selected seizures from each individual. The same seizure signature was then searched for in the entire recording from the same patient using dynamic time warping (DTW) as a similarity metric, with a threshold set to reflect the maximum sensitivity our algorithm could achieve without false alarm. RESULTS: At a DTW threshold providing no false alarm during the entire recordings, the mean seizure detection sensitivity across patients was 84%, including 96% for the CHB-MIT database and 74% for the European epilepsy database. A 100% sensitivity was reached in 50% of patients, including 79% from the CHB-MIT database and 27% from the European epilepsy database. The median latency from seizure onset to its detection was 17 ± 10 s, with 84% of seizures being detected within 40 s. SIGNIFICANCE: Personalized EEG signature combined with DTW appears to be a promising method to detect ictal events from a limited number of EEG channels with high sensitivity despite low rate of false alarms, high degree of interpretability, and low computational complexity, compatible with its future use in wearable devices.

Personalized Real-Time Federated Learning for Epileptic Seizure Detection.

Epilepsy is one of the most prevalent paroxystic neurological disorders. It is characterized by the occurrence of spontaneous seizures. About 1 out of 3 patients have drug-resistant epilepsy, thus their seizures cannot be controlled by medication. Automatic detection of epileptic seizures can substantially improve the patient's quality of life. To achieve a high-quality model, we have to collect data from various patients in a central server. However, sending the patient's raw data to this central server puts patient privacy at risk and consumes a significant amount of energy. To address these challenges, in this work, we have designed and evaluated a standard federated learning framework in the context of epileptic seizure detection using a deep learning-based approach, which operates across a cluster of machines. We evaluated the accuracy and performance of our proposed approach on the NVIDIA Jetson Nano Developer Kit based on the EPILEPSIAE database, which is one of the largest public epilepsy datasets for seizure detection. Our proposed framework achieved a sensitivity of 81.25%, a specificity of 82.00%, and a geometric mean of 81.62%. It can be implemented on embedded platforms that complete the entire training process in 1.86 hours using 344.34 mAh energy on a single battery charge. We also studied a personalized variant of the federated learning, where each machine is responsible for training a deep neural network (DNN) to learn the discriminative electrocardiography (ECG) features of the epileptic seizures of the specific person monitored based on its local data. In this context, the DNN benefitted from a well-trained model without sharing the patient's raw data with a server or a central cloud repository. We observe in our results that personalized federated learning provides an increase in all the performance metric, with a sensitivity of 90.24%, a specificity of 91.58%, and a geometric mean of 90.90%.

Real-Time EEG-Based Cognitive Workload Monitoring on Wearable Devices.

OBJECTIVE: Cognitive workload monitoring (CWM) can enhance human-machine interaction by supporting task execution assistance considering the operator's cognitive state. Therefore, we propose a machine learning design methodology and a data processing strategy to enable CWM on resource-constrained wearable devices. METHODS: Our CWM solution is built upon edge computing on a simple wearable system, with only four peripheral channels of electroencephalography (EEG). We assess our solution on experimental data from 24 volunteers. Moreover, to overcome the system's memory constraints, we adopt an optimization strategy for model size reduction and a multi-batch data processing scheme for optimizing RAM memory footprint. Finally, we implement our data processing strategy on a state-of-the-art wearable platform and assess its execution and system battery life. RESULTS: We achieve an accuracy of 74.5% and a 74.0% geometric mean between sensitivity and specificity for CWM classification on unseen data. Besides, the proposed model optimization strategy generates a 27.5x smaller model compared to the one generated with default parameters, and the multi-batch data processing scheme reduces RAM memory footprint by 14x compared to a single batch data processing. Finally, our algorithm uses only 1.28% of the available processing time, thus, allowing our system to achieve 28.5 hours of battery life. CONCLUSION: We provide a reliable and optimized CWM solution using wearable devices, enabling more than a day of operation on a single battery charge. SIGNIFICANCE: The proposed methodology enables real-time data processing on resource-constrained devices and supports real-time wearable monitoring based on EEG for applications as CWM in human-machine interaction.

EpilepsyGAN: Synthetic Epileptic Brain Activities With Privacy Preservation.

Epilepsy is a chronic neurological disorder affecting more than 65 million people worldwide and manifested by recurrent unprovoked seizures. The unpredictability of seizures not only degrades the quality of life of the patients, but it can also be life-threatening. Modern systems monitoring electroencephalography (EEG) signals are being currently developed with the view to detect epileptic seizures in order to alert caregivers and reduce the impact of seizures on patients' quality of life. Such seizure detection systems employ state-of-the-art machine learning algorithms that require a large amount of labeled personal data for training. However, acquiring EEG signals during epileptic seizures is a costly and time-consuming process for medical experts and patients. Furthermore, this data often contains sensitive personal information, presenting privacy concerns. In this work, we generate synthetic seizure-like brain electrical activities, i.e., EEG signals, that can be used to train seizure detection algorithms, alleviating the need for sensitive recorded data. Our experiments show that the synthetic seizure data generated with our GAN model succeeds at preserving the privacy of the patients without producing any degradation in performance during seizure monitoring.

Robust Epileptic Seizure Detection on Wearable Systems with Reduced False-Alarm Rate.

Epilepsy affects more than 50 million people and ranks among the most common neurological diseases worldwide. Despite advances in treatment, one-third of patients still suffer from refractory epilepsy. Wearable devices for real-time patient monitoring can potentially improve the quality of life for such patients and reduce the mortality rate due to seizure-related accidents and sudden death in epilepsy. However, the majority of employed seizure detection techniques and devices suffer from unacceptable false-alarm rate. In this paper, we propose a robust seizure detection methodology for a wearable platform and validate it on the Physionet.org CHB-MIT Scalp EEG database. It reaches sensitivity of 0.966 and specificity of 0.925, and reducing the false-alarm rate by 34.7%. We also evaluate the battery lifetime of the wearable system including our proposed methodology and demonstrate the feasibility of using it in real time for up to 40.87 hours on a single battery charge.

Resource-Aware Distributed Epilepsy Monitoring Using Self-Awareness From Edge to Cloud.

The integration of wearable devices in humans' daily lives has grown significantly in recent years and still continues to affect different aspects of high-quality life. Thus, ensuring the reliability of the decisions becomes essential in biomedical applications, while representing a major challenge considering battery-powered wearable technologies. Transferring the complex and energy-consuming computations to fogs or clouds can significantly reduce the energy consumption of wearable devices and result in a longer lifetime of these systems with a single battery charge. In this work, we aim to distribute the complex and energy-consuming machine-learning computations between the edge, fog, and cloud, based on the notion of self-awareness that takes into account the complexity and reliability of the algorithm. We also model and analyze the trade-offs in terms of energy consumption, latency, and performance of different Internet of Things (IoT) solutions. We consider the epileptic seizure detection problem as our real-world case study to demonstrate the importance of our proposed self-aware methodology.

Multi-Modal Acute Stress Recognition Using Off-the-Shelf Wearable Devices.

Monitoring stress and, in general, emotions has attracted a lot of attention over the past few decades. Stress monitoring has many applications, including high-risk missions and surgical procedures as well as mental/emotional health monitoring. In this paper, we evaluate the possibility of stress and emotion monitoring using off-the-shelf wearable sensors. To this aim, we propose a multi-modal machine-learning technique for acute stress episodes detection, by fusing the information careered in several biosignals and wearable sensors. Furthermore, we investigate the contribution of each wearable sensor in stress detection and demonstrate the possibility of acute stress recognition using wearable devices. In particular, we acquire the physiological signals using the Shimmer3 ECG Unit and the Empatica E4 wristband. Our experimental evaluation shows that it is possible to detect acute stress episodes with an accuracy of 84.13%, for an unseen test set, using multi-modal machinelearning and sensor-fusion techniques.

Real-Time Event-Driven Classification Technique for Early Detection and Prevention of Myocardial Infarction on Wearable Systems.

A considerable portion of government health-care spending is allocated to the continuous monitoring of patients suffering from cardiovascular diseases, particularly myocardial infarction (MI). Wearable devices present a cost-effective means of monitoring patients' vital signs in ambulatory settings. A major challenge is to design such ultra-low energy devices for long-term patient monitoring. In this paper, we present a real-time event-driven classification technique based on the random forest classification scheme, which uses a confidence-related decision-making process. The main goal of this technique is to maintain a high classification accuracy while reducing the complexity of the classification algorithm. We validate our approach on a well-established and complete MI database (Physiobank, PTB Diagnostic ECG database). Our experimental evaluation demonstrates that our real-time classification scheme outperforms the existing approaches in terms of energy consumption and battery lifetime by a factor of 2.60, with no classification quality loss.

Online Obstructive Sleep Apnea Detection on Medical Wearable Sensors.

Obstructive Sleep Apnea (OSA) is one of the main under-diagnosed sleep disorder. It is an aggravating factor for several serious cardiovascular diseases, including stroke. There is, however, a lack of medical devices for long-term ambulatory monitoring of OSA since current systems are rather bulky, expensive, intrusive, and cannot be used for long-term monitoring in ambulatory settings. In this paper, we propose a wearable, accurate, and energy efficient system for monitoring obstructive sleep apnea on a long-term basis. As an embedded system for Internet of Things, it reduces the gap between home health-care and professional supervision. Our approach is based on monitoring the patient using a single-channel electrocardiogram signal. We develop an efficient time-domain analysis to meet the stringent resources constraints of embedded systems to compute the sleep apnea score. Our system, for a publicly available database (PhysioNet Apnea-ECG), has a classification accuracy of up to 88.2% for our new online and patient-specific analysis, which takes the distinct profile of each patient into account. While accurate, our approach is also energy efficient and can achieve a battery lifetime of 46 days for continuous screening of OSA.