Direct Electron Transfer-Type Oxidoreductases for Biomedical Applications.

Among the various types of enzyme-based biosensors, sensors utilizing enzymes capable of direct electron transfer (DET) are recognized as the most ideal. However, only a limited number of redox enzymes are capable of DET with electrodes, that is, dehydrogenases harboring a subunit or domain that functions specifically to accept electrons from the redox cofactor of the catalytic site and transfer the electrons to the external electron acceptor. Such subunits or domains act as built-in mediators for electron transfer between enzymes and electrodes; consequently, such enzymes enable direct electron transfer to electrodes and are designated as DET-type enzymes. DET-type enzymes fall into several categories, including redox cofactors of catalytic reactions, built-in mediators for DET with electrodes and by their protein hierarchic structures, DET-type oxidoreductases with oligomeric structures harboring electron transfer subunits, and monomeric DET-type oxidoreductases harboring electron transfer domains. In this review, we cover the science of DET-type oxidoreductases and their biomedical applications. First, we introduce the structural biology and current understanding of DET-type enzyme reactions. Next, we describe recent technological developments based on DET-type enzymes for biomedical applications, such as biosensors and biochemical energy harvesting for self-powered medical devices. Finally, after discussing how to further engineer and create DET-type enzymes, we address the future prospects for DET-type enzymes in biomedical engineering.

The correlation of urea and creatinine concentrations in sweat and saliva with plasma during hemodialysis: an observational cohort study.

OBJECTIVES: Urea and creatinine concentrations in plasma are used to guide hemodialysis (HD) in patients with end-stage renal disease (ESRD). To support individualized HD treatment in a home situation, there is a clinical need for a non-invasive and continuous alternative to plasma for biomarker monitoring during and between cycles of HD. In this observational study, we therefore established the correlation of urea and creatinine concentrations between sweat, saliva and plasma in a cohort of ESRD patients on HD. METHODS: Forty HD patients were recruited at the Dialysis Department of the Catharina Hospital Eindhoven. Sweat and salivary urea and creatinine concentrations were analyzed at the start and at the end of one HD cycle and compared to the corresponding plasma concentrations. RESULTS: A decrease of urea concentrations during HD was observed in sweat, from 27.86 mmol/L to 12.60 mmol/L, and saliva, from 24.70 mmol/L to 5.64 mmol/L. Urea concentrations in sweat and saliva strongly correlated with the concentrations in plasma (ρ 0.92 [p<0.001] and 0.94 [p<0.001], respectively). Creatinine concentrations also decreased in sweat from 43.39 µmol/L to 19.69 µmol/L, and saliva, from 59.00 µmol/L to 13.70 µmol/L. However, for creatinine, correlation coefficients were lower than for urea for both sweat and saliva compared to plasma (ρ: 0.58 [p<0.001] and 0.77 [p<0.001], respectively). CONCLUSIONS: The results illustrate a proof of principle of urea measurements in sweat and saliva to monitor HD adequacy in a non-invasive and continuous manner. Biosensors enabling urea monitoring in sweat or saliva could fill in a clinical need to enable at-home HD for more patients and thereby decrease patient burden.

Continuous cardiac monitoring in epilepsy: an implantable loop manual activation algorithm for improving ECG signal acquisition accuracy.

BACKGROUND: The muscle artifacts, caused by prominent muscle contractions, mimicking cardiac arrhythmias, might compromise the ECG signal quality and the implantable loop recorder memory capacity in patients with epilepsy. We developed an epileptic seizures clinical pattern-based implantable loop recorder manual activation algorithm, presenting its real-world efficacy here. METHODS: One hundred ninety-three patients (18-60 years) with drug-resistant focal epilepsy were consecutively enrolled and underwent a subcutaneous loop recorder implantation. Patients with focal onset-aware seizures and patients with focal impaired awareness seizures /bilateral tonic-clonic seizures without aura were recommended to use the activator once - just after the episode. Patients with focal impaired awareness seizures/bilateral tonic-clonic seizures with aura, the caregivers of patients experiencing status epilepticus, were advised to use the activator twice - during the aura and after the episode/ regaining consciousness. RESULTS: Six thousand four hundred ninety-four ECG traces (4826 - auto-triggered events, 1668 - person-activated events) were recorded and analyzed. The rate of true positive events in the person-activated group was statistically higher than in the autoactivation group (72.5% vs.19.4%, p < 0.0001). Person-activated false-positive events were observed in 30.5% of patients with focal impaired awareness seizures and 27.7% in patients with bilateral tonic-clonic seizures. The highest rate of false-positive events (61.5%) was detected in patients undergoing epileptic status, and the lowest rate (3.8%) - was in patients with focal onset aware seizures. The rate of false-positive events was significantly higher in patients with impaired awareness seizures without aura both in focal impaired awareness (45.5% vs. 19.3%, p < 0.0001) and bilateral tonic-clonic seizure groups (38.8% vs. 5.9%, p < 0.0001). CONCLUSIONS: Arrhythmias with varying clinical outcomes are expected in epilepsy patients and have been monitored continuously. The specified loop recorder external activation algorithm can improve the clinically relevant cardiac arrhythmia detection accuracy in epilepsy patients and the value of future studies.

Evaluation of an algorithm-guided photoplethysmography for atrial fibrillation burden using a smartwatch.

BACKGROUND: Wearable devices based on the PPG algorithm can detect atrial fibrillation (AF) effectively. However, further investigation of its application on long-term, continuous monitoring of AF burden is warranted. METHOD: The performance of a smartwatch with continuous photoplethysmography (PPG) and PPG-based algorithms for AF burden estimation was evaluated in a prospective study enrolling AF patients admitted to Beijing Anzhen Hospital for catheter ablation from September to November 2022. A continuous Electrocardiograph patch (ECG) was used as the reference device to validate algorithm performance for AF detection in 30-s intervals. RESULTS: A total of 578669 non-overlapping 30-s intervals for PPG and ECG each from 245 eligible patients were generated. An interval-level sensitivity of PPG was 96.3% (95% CI 96.2%-96.4%), and specificity was 99.5% (95% CI 99.5%-99.6%) for the estimation of AF burden. AF burden estimation by PPG was highly correlated with AF burden calculated by ECG via Pearson correlation coefficient (R2 = 0.996) with a mean difference of -0.59 (95% limits of agreement, -7.9% to 6.7%). The subgroup study showed the robust performance of the algorithm in different subgroups, including heart rate and different hours of the day. CONCLUSION: Our results showed the smartwatch with an algorithm-based PPG monitor has good accuracy and stability in continuously monitoring AF burden compared with ECG patch monitors, indicating its potential for diagnosing and managing AF.

Improving diagnosis of early complications (<1 week) through continuous vital sign monitoring following oncological gastrointestinal surgical procedures.

BACKGROUND: Patients undergoing major oncological abdominal surgery are prone to postoperative complications, making early recognition crucial. Clinical deterioration is often preceded by changes in vital signs, which are typically measured thrice a day by a nurse. However, intermittent measurements may delay recognizing clinical deterioration. Continuous vital parameter monitoring may lead to earlier recognition and management of complications and reduce nursing workload. OBJECTIVE: To compare vital parameter measurements between ward nurses and a wireless continuous monitoring system (Sensium® wireless patch) and assess whether this patch can detect clinical deterioration earlier in patients with complications in the first postoperative week. METHODS: Vital parameters (heart rate, respiratory rate, and temperature) were collected in patients undergoing an oncological resection of the liver, colorectal, or pancreas. Sensium® patch measurements were compared to nurses' measurements to assess the percentages of discordant measurements. In patients with complications in the first postoperative week, time discrepancies between nurses and Sensium® patch measurements were identified in cases of clinical deterioration (respiratory rate ≥15/min, heart rate ≥100/min, and temperature ≥38°C). RESULTS: Among 227 patients, 22% of the patients experienced complications. Nurse and Sensium® measurements were discrepant in 586/2272 measurements (26%). In 506/586 discrepancies (86%), this was due to the respiratory rate (difference ≥4/min). Compared to nurses, the Sensium® patch detected an elevated respiratory rate 14 h earlier and heart rate 2 h earlier within complications in the first postoperative week. For temperature, no difference was observed. CONCLUSION: Continuous monitoring with the Sensium® wireless patch holds promise for earlier recognition of complications in patients who underwent major oncological abdominal surgery.

Workload associated with manual assessment of vital signs as compared with continuous wireless monitoring.

BACKGROUND: Vital sign monitoring is considered an essential aspect of clinical care in hospitals. In general wards, this relies on intermittent manual assessments performed by clinical staff at intervals of up to 12 h. In recent years, continuous monitoring of vital signs has been introduced to the clinic, with improved patient outcomes being one of several potential benefits. The aim of this study was to determine the workload difference between continuous monitoring and manual monitoring of vital signs as part of the National Early Warning Score (NEWS). METHODS: Three wireless sensors continuously monitored blood pressure, heart rate, respiratory rate, and peripheral oxygen saturation in 20 patients admitted to the general hospital ward. The duration needed for equipment set-up and maintenance for continuous monitoring in a 24-h period was recorded and compared with the time spent on manual assessments and documentation of vital signs performed by clinical staff according to the NEWS. RESULTS: The time used for continuous monitoring was 6.0 (IQR 3.2; 7.2) min per patient per day vs. 14 (9.7; 32) min per patient per day for the NEWS. Median difference in duration for monitoring of vital signs was 9.9 (95% CI 5.6; 21) min per patient per day between NEWS and continuous monitoring (p < .001). Time used for continuous monitoring in isolated patients was 6.6 (4.6; 12) min per patient per day as compared with 22 (9.7; 94) min per patient per day for NEWS. CONCLUSION: The use of continuous monitoring was associated with a significant reduction in workload in terms of time for monitoring as compared with manual assessment of vital signs.

New spin coated multilayer lactate biosensor for acidosis monitoring in continuous flow assisted with an electrochemical pH probe.

A LOx-based electrochemical biosensor for high-level lactate determination was developed. For the construction of the biosensor, chitosan and Nafion layers were integrated by using a spin coating procedure, leading to less porous surfaces in comparison with those recorded after a drop casting procedure. The analytical performance of the resulting biosensor for lactate determination was evaluated in batch and flow regime, displaying satisfactory results in both modes ranging from 0.5 to 20 mM concentration range for assessing the lactic acidosis. Finally, the lactate levels in raw serum samples were estimated using the biosensor developed and verified with a blood gas analyzer. Based on these results, the biosensor developed is promising for its use in healthcare environment, after its proper miniaturization. A pH probe based on common polyaniline-based electrochemical sensor was also developed to assist the biosensor for the lactic acidosis monitoring, leading to excellent results in stock solutions ranging from 6.0 to 8.0 mM and raw plasma samples. The results were confirmed by using two different approaches, blood gas analyzer and pH-meter. Consequently, the lactic acidosis monitoring could be achieved in continuous flow regime using both (bio)sensors.

Advances in microneedles for transdermal diagnostics and sensing applications.

Microneedles, the miniaturized needles, which can pierce the skin with minimal invasiveness open up new possibilities for constructing personalized Point-of-Care (POC) diagnostic platforms. Recent advances in microneedle-based POC diagnostic systems, especially their successful implementation with wearable technologies, enable biochemical detection and physiological recordings in a user-friendly manner. This review presents an overview of the current advances in microneedle-based sensor devices, with emphasis on the biological basis of transdermal sensing, fabrication, and application of different types of microneedles, and a summary of microneedle devices based on various sensing strategies. It concludes with the challenges and future prospects of this swiftly growing field. The aim is to present a critical and thorough analysis of the state-of-the-art development of transdermal diagnostics and sensing devices based on microneedles, and to bridge the gap between microneedle technology and pragmatic applications.

Inline Raman spectroscopy as process analytical technology for SARS-CoV-2 VLP production.

The present work focused on inline Raman spectroscopy monitoring of SARS-CoV-2 VLP production using two culture media by fitting chemometric models for biochemical parameters (viable cell density, cell viability, glucose, lactate, glutamine, glutamate, ammonium, and viral titer). For that purpose, linear, partial least square (PLS), and nonlinear approaches, artificial neural network (ANN), were used as correlation techniques to build the models for each variable. ANN approach resulted in better fitting for most parameters, except for viable cell density and glucose, whose PLS presented more suitable models. Both were statistically similar for ammonium. The mean absolute error of the best models, within the quantified value range for viable cell density (375,000-1,287,500 cell/mL), cell viability (29.76-100.00%), glucose (8.700-10.500 g/), lactate (0.019-0.400 g/L), glutamine (0.925-1.520 g/L), glutamate (0.552-1.610 g/L), viral titer (no virus quantified-7.505 log10 PFU/mL) and ammonium (0.0074-0.0478 g/L) were, respectively, 41,533 ± 45,273 cell/mL (PLS), 1.63 ± 1.54% (ANN), 0.058 ± 0.065 g/L (PLS), 0.007 ± 0.007 g/L (ANN), 0.007 ± 0.006 g/L (ANN), 0.006 ± 0.006 g/L (ANN), 0.211 ± 0.221 log10 PFU/mL (ANN), and 0.0026 ± 0.0026 g/L (PLS) or 0.0027 ± 0.0034 g/L (ANN). The correlation accuracy, errors, and best models obtained are in accord with studies, both online and offline approaches while using the same insect cell/baculovirus expression system or different cell host. Besides, the biochemical tracking throughout bioreactor runs using the models showed suitable profiles, even using two different culture media.

All-Fiber Flexible Electrochemical Sensor for Wearable Glucose Monitoring.

Wearable sensors, specifically microneedle sensors based on electrochemical methods, have expanded extensively with recent technological advances. Today's wearable electrochemical sensors present specific challenges: they show significant modulus disparities with skin tissue, implying possible discomfort in vivo, especially over extended wear periods or on sensitive skin areas. The sensors, primarily based on polyethylene terephthalate (PET) or polyimide (PI) substrates, might also cause pressure or unease during insertion due to the skin's irregular deformation. To address these constraints, we developed an innovative, wearable, all-fiber-structured electrochemical sensor. Our composite sensor incorporates polyurethane (PU) fibers prepared via electrospinning as electrode substrates to achieve excellent adaptability. Electrospun PU nanofiber films with gold layers shaped via thermal evaporation are used as base electrodes with exemplary conductivity and electrochemical catalytic attributes. To achieve glucose monitoring, gold nanofibers functionalized by gold nanoflakes (AuNFs) and glucose oxidase (GOx) serve as the working electrode, while Pt nanofibers and Ag/AgCl nanofibers serve as the counter and reference electrode. The acrylamide-sodium alginate double-network hydrogel synthesized on electrospun PU fibers serves as the adhesive and substance-transferring layer between the electrodes. The all-fiber electrochemical sensor is assembled layer-by-layer to form a robust structure. Given the stretchability of PU nanofibers coupled with a high specific surface area, the manufactured porous microneedle glucose sensor exhibits enhanced stretchability, superior sensitivity at 31.94 µA (lg(mM))-1 cm-2, a broad detection range (1-30 mM), and a significantly low detection limit (1 mM, S/N = 3), as well as satisfactory biocompatibility. Therefore, the novel electrochemical microneedle design is well-suited for wearable or even implantable continuous monitoring applications, thereby showing promising significant potential within the global arena of wearable medical technology.

Inferring ECG Waveforms from PPG Signals with a Modified U-Net Neural Network.

There are two widely used methods to measure the cardiac cycle and obtain heart rate measurements: the electrocardiogram (ECG) and the photoplethysmogram (PPG). The sensors used in these methods have gained great popularity in wearable devices, which have extended cardiac monitoring beyond the hospital environment. However, the continuous monitoring of ECG signals via mobile devices is challenging, as it requires users to keep their fingers pressed on the device during data collection, making it unfeasible in the long term. On the other hand, the PPG does not contain this limitation. However, the medical knowledge to diagnose these anomalies from this sign is limited by the need for familiarity, since the ECG is studied and used in the literature as the gold standard. To minimize this problem, this work proposes a method, PPG2ECG, that uses the correlation between the domains of PPG and ECG signals to infer from the PPG signal the waveform of the ECG signal. PPG2ECG consists of mapping between domains by applying a set of convolution filters, learning to transform a PPG input signal into an ECG output signal using a U-net inception neural network architecture. We assessed our proposed method using two evaluation strategies based on personalized and generalized models and achieved mean error values of 0.015 and 0.026, respectively. Our method overcomes the limitations of previous approaches by providing an accurate and feasible method for continuous monitoring of ECG signals through PPG signals. The short distances between the infer-red ECG and the original ECG demonstrate the feasibility and potential of our method to assist in the early identification of heart diseases.

Advancements in Polymer-Assisted Layer-by-Layer Fabrication of Wearable Sensors for Health Monitoring.

Recent advancements in polymer-assisted layer-by-layer (LbL) fabrication have revolutionized the development of wearable sensors for health monitoring. LbL self-assembly has emerged as a powerful and versatile technique for creating conformal, flexible, and multi-functional films on various substrates, making it particularly suitable for fabricating wearable sensors. The incorporation of polymers, both natural and synthetic, has played a crucial role in enhancing the performance, stability, and biocompatibility of these sensors. This review provides a comprehensive overview of the principles of LbL self-assembly, the role of polymers in sensor fabrication, and the various types of LbL-fabricated wearable sensors for physical, chemical, and biological sensing. The applications of these sensors in continuous health monitoring, disease diagnosis, and management are discussed in detail, highlighting their potential to revolutionize personalized healthcare. Despite significant progress, challenges related to long-term stability, biocompatibility, data acquisition, and large-scale manufacturing are still to be addressed, providing insights into future research directions. With continued advancements in polymer-assisted LbL fabrication and related fields, wearable sensors are poised to improve the quality of life for individuals worldwide.

Real-Time Monitoring of Odour Emissions at the Fenceline of a Waste Treatment Plant by Instrumental Odour Monitoring Systems: Focus on Training Methods.

Waste treatment plants (WTPs) often generate odours that may cause nuisance to citizens living nearby. In general, people are becoming more sensitive to environmental issues, and particularly to odour pollution. Instrumental Odour Monitoring Systems (IOMSs) represent an emerging tool for continuous odour measurement and real-time identification of odour peaks, which can provide useful information about the process operation and indicate the occurrence of anomalous conditions likely to cause odour events in the surrounding territories. This paper describes the implementation of two IOMSs at the fenceline of a WTP, focusing on the definition of a specific experimental protocol and data processing procedure for dealing with the interferences of humidity and temperature affecting sensors' responses. Different approaches for data processing were compared and the optimal one was selected based on field performance testing. The humidity compensation model developed proved to be effective, bringing the IOMS classification accuracy above 95%. Also, the adoption of a class-specific regression model compared to a global regression model resulted in an odour quantification capability comparable with those of the reference method (i.e., dynamic olfactometry). Lastly, the validated models were used to process the monitoring data over a period of about one year.

Human Activity Recording Based on Skin-Strain-Actuated Microfluidic Pumping in Asymmetrically Designed Micro-Channels.

The capability to record data in passive, image-based wearable sensors can simplify data readouts and eliminate the requirement for the integration of electronic components on the skin. Here, we developed a skin-strain-actuated microfluidic pump (SAMP) that utilizes asymmetric aspect ratio channels for the recording of human activity in the fluidic domain. An analytical model describing the SAMP's operation mechanism as a wearable microfluidic device was established. Fabrication of the SAMP was achieved using soft lithography from polydimethylsiloxane (PDMS). Benchtop experimental results and theoretical predictions were shown to be in good agreement. The SAMP was mounted on human skin and experiments conducted on volunteer subjects demonstrated the SAMP's capability to record human activity for hundreds of cycles in the fluidic domain through the observation of a stable liquid meniscus. Proof-of-concept experiments further revealed that the SAMP could quantify a single wrist activity repetition or distinguish between three different shoulder activities.

Excessive Oxygen Administration in High-Risk Patients Admitted to Medical and Surgical Wards Monitored by Wireless Pulse Oximeter.

The monitoring of oxygen therapy when patients are admitted to medical and surgical wards could be important because exposure to excessive oxygen administration (EOA) may have fatal consequences. We aimed to investigate the association between EOA, monitored by wireless pulse oximeter, and nonfatal serious adverse events (SAEs) and mortality within 30 days. We included patients in the Capital Region of Copenhagen between 2017 and 2018. Patients were hospitalized due to acute exacerbation of chronic obstructive pulmonary disease (AECOPD) or after major elective abdominal cancer surgery, and all were treated with oxygen supply. Patients were divided into groups by their exposure to EOA: no exposure, exposure for 1-59 min or exposure over 60 min. The primary outcome was SAEs or mortality within 30 days. We retrieved data from 567 patients for a total of 43,833 h, of whom, 63% were not exposed to EOA, 26% had EOA for 1-59 min and 11% had EOA for ≥60 min. Nonfatal SAEs or mortality within 30 days developed in 24%, 12% and 22%, respectively, and the adjusted odds ratio for this was 0.98 (95% CI, 0.96-1.01) for every 10 min. increase in EOA, without any subgroup effects. In conclusion, we did not observe higher frequencies of nonfatal SAEs or mortality within 30 days in patients exposed to excessive oxygen administration.

Point Sensor Networks Struggle to Detect and Quantify Short Controlled Releases at Oil and Gas Sites.

This study evaluated multiple commercially available continuous monitoring (CM) point sensor network (PSN) solutions under single-blind controlled release testing conducted at operational upstream and midstream oil and natural gas (O&G) sites. During releases, PSNs reported site-level emission rate estimates of 0 kg/h between 38 and 86% of the time. When non-zero site-level emission rate estimates were provided, no linear correlation between the release rate and the reported emission rate estimate was observed. The average, aggregated across all PSN solutions during releases, shows 5% of the mixing ratio readings at downwind sensors were greater than the site's baseline plus two standard deviations. Four of seven total PSN solutions tested during this field campaign provided site-level emission rate estimates with the site average relative error ranging from -100% to 24% for solution D, -100% to -43% for solution E, -25% for solution F (solution F was only at one site), and -99% to 430% for solution G, with an overall average of -29% across all sites and solutions. Of all the individual site-level emission rate estimates, only 11% were within ±2.5 kg/h of the study team's best estimate of site-level emissions at the time of the releases.

Direct Memory Access-Based Data Storage for Long-Term Acquisition Using Wearables in an Energy-Efficient Manner.

This study introduces a lightweight storage system for wearable devices, aiming to optimize energy efficiency in long-term and continuous monitoring applications. Utilizing Direct Memory Access and the Serial Peripheral Interface protocol, the system ensures efficient data transfer, significantly reduces energy consumption, and enhances the device autonomy. Data organization into Time Block Data (TBD) units, rather than files, significantly diminishes control overhead, facilitating the streamlined management of periodic data recordings in wearable devices. A comparative analysis revealed marked improvements in energy efficiency and write speed over existing file systems, validating the proposed system as an effective solution for boosting wearable device performance in health monitoring and various long-term data acquisition scenarios.

Performance Assessment for the Validation of Wireless Communication Engines in an Innovative Wearable Monitoring Platform.

In today's health-monitoring applications, there is a growing demand for wireless and wearable acquisition platforms capable of simultaneously gathering multiple bio-signals from multiple body areas. These systems require well-structured software architectures, both to keep different wireless sensing nodes synchronized each other and to flush collected data towards an external gateway. This paper presents a quantitative analysis aimed at validating both the wireless synchronization task (implemented with a custom protocol) and the data transmission task (implemented with the BLE protocol) in a prototype wearable monitoring platform. We evaluated seven frequencies for exchanging synchronization packets (10 Hz, 20 Hz, 30 Hz, 40 Hz, 50 Hz, 60 Hz, 70 Hz) as well as two different BLE configurations (with and without the implementation of a dynamic adaptation of the BLE Connection Interval parameter). Additionally, we tested BLE data transmission performance in five different use case scenarios. As a result, we achieved the optimal performance in the synchronization task (1.18 ticks as median synchronization delay with a Min-Max range of 1.60 ticks and an Interquartile range (IQR) of 0.42 ticks) when exploiting a synchronization frequency of 40 Hz and the dynamic adaptation of the Connection Interval. Moreover, BLE data transmission proved to be significantly more efficient with shorter distances between the communicating nodes, growing worse by 30.5% beyond 8 m. In summary, this study suggests the best-performing network configurations to enhance the synchronization task of the prototype platform under analysis, as well as quantitative details on the best placement of data collectors.

Artificial Intelligence-Driven Prognosis of Respiratory Mechanics: Forecasting Tissue Hysteresivity Using Long Short-Term Memory and Continuous Sensor Data.

Tissue hysteresivity is an important marker for determining the onset and progression of respiratory diseases, calculated from forced oscillation lung function test data. This study aims to reduce the number and duration of required measurements by combining multivariate data from various sensing devices. We propose using the Forced Oscillation Technique (FOT) lung function test in both a low-frequency prototype and the commercial RESMON device, combined with continuous monitoring from the Equivital (EQV) LifeMonitor and processed by artificial intelligence (AI) algorithms. While AI and deep learning have been employed in various aspects of respiratory system analysis, such as predicting lung tissue displacement and respiratory failure, the prediction or forecasting of tissue hysteresivity remains largely unexplored in the literature. In this work, the Long Short-Term Memory (LSTM) model is used in two ways: (1) to estimate the hysteresivity coefficient η using heart rate (HR) data collected continuously by the EQV sensor, and (2) to forecast η values by first predicting the heart rate from electrocardiogram (ECG) data. Our methodology involves a rigorous two-hour measurement protocol, with synchronized data collection from the EQV, FOT, and RESMON devices. Our results demonstrate that LSTM networks can accurately estimate the tissue hysteresivity parameter η, achieving an R2 of 0.851 and a mean squared error (MSE) of 0.296 for estimation, and forecast η with an R2 of 0.883 and an MSE of 0.528, while significantly reducing the number of required measurements by a factor of three (i.e., from ten to three) for the patient. We conclude that our novel approach minimizes patient effort by reducing the measurement time and the overall ambulatory time and costs while highlighting the potential of artificial intelligence methods in respiratory monitoring.

A machine learning algorithm for detecting abnormal patterns in continuous capnography and pulse oximetry monitoring.

Continuous capnography monitors patient ventilation but can be susceptible to artifact, resulting in alarm fatigue. Development of smart algorithms may facilitate accurate detection of abnormal ventilation, allowing intervention before patient deterioration. The objective of this analysis was to use machine learning (ML) to classify combined waveforms of continuous capnography and pulse oximetry as normal or abnormal. We used data collected during the observational, prospective PRODIGY trial, in which patients receiving parenteral opioids underwent continuous capnography and pulse oximetry monitoring while on the general care floor [1]. Abnormal ventilation segments in the data stream were reviewed by nine experts and inter-rater agreement was assessed. Abnormal segments were defined as the time series 60s before and 30s after an abnormal pattern was detected. Normal segments (90s continuous monitoring) were randomly sampled and filtered to discard sequences with missing values. Five ML models were trained on extracted features and optimized towards an Fß score with ß = 2. The results show a high inter-rater agreement (> 87%), allowing 7,858 sequences (2,944 abnormal) to be used for model development. Data were divided into 80% training and 20% test sequences. The XGBoost model had the highest Fß score of 0.94 (with ß = 2), showcasing an impressive recall of 0.98 against a precision of 0.83. This study presents a promising advancement in respiratory monitoring, focusing on reducing false alarms and enhancing accuracy of alarm systems. Our algorithm reliably distinguishes normal from abnormal waveforms. More research is needed to define patterns to distinguish abnormal ventilation from artifacts.