Monitoring Applications Research Articles

Ultrahighly Sensitive Flexible Pressure Sensors with Dual-Mode Response Based on Monolayer Films of Calcium Niobate Nanosheets.

Flexible pressure sensors present enormous potential for applications in health monitoring, human-machine interfacing, and electronic skins (e-skin). However, at the cost of flexibility, the design of flexible pressure sensors has been facing the trade off between high sensitivity and wide sensing range. Herein, we propose a sandwiched structure composed of monolayer films of calcium niobate nanosheets to endow the device with both ultrahigh sensitivity and a wide sensing range. Tunable contact between the two electrodes of the pressure sensor through the gaps in the insulative monolayer film and precise thickness modulation of the monolayer films at the nanoscale result in an ultrahigh sensitivity and wide sensing range of the sensors. By virtue of these two traits, the pressure sensor distinguishes itself with unprecedented performances of ultrahigh sensitivity (6.43 × 104 kPa-1), a wide sensing range (1.94-60.00 kPa), a fast response time (<165 ms), and reliable repeatability. In addition, the sensor has a sensing mechanism transition from capacitive mode to piezoresistive mode from low pressure to high pressure. The sensors demonstrates the ability for motion detection of the human body. This work sheds light on the development of highly sensitive flexible pressure sensors.

Humidity Sensing Properties of Different Atomic Layers of Graphene on the SiO2/Si Substrate.

Graphene has great potential to be used for humidity sensing due to its ultrahigh surface area and conductivity. However, the impact of different atomic layers of graphene on the SiO2/Si substrate on humidity sensing has not been studied yet. In this paper, we fabricated three types of humidity sensors on the SiO2/Si substrate based on one to three atomic layers of graphene, in which the sensing areas of graphene are 75 μm × 72 μm and 45 μm × 72 μm, respectively. We studied the impact of both the number of atomic layers of graphene and the sensing areas of graphene on the responsivity and response/recovery time of the prepared graphene-based humidity sensors. We found that the relative resistance change of the prepared devices decreased with the increase of number of atomic layers of graphene under the same change of relative humidity. Further, devices based on tri-layer graphene showed the fastest response/recovery time, while devices based on double-layer graphene showed the slowest response/recovery time. Finally, we chose devices based on double-layer graphene that have relatively good responsivity and stability for application in respiration monitoring and contact-free finger monitoring.

Synergistic surface ligand modification of Ni-Pt bimetallic nanozymes: Enhanced catalytic activity and versatile detection of penicillin

Monitoring and control of penicillin (PCN) in the environment are critically important for reducing the emergence and dissemination of antibiotic-resistant bacteria, ensuring food safety, and safeguarding human health. In this study, we introduce a novel and facile strategy for modulating the catalytic activity of bimetallic nanozymes through surface ligand engineering. Experimental activity assays and theoretical model calculations confirm that Ni-Pt nanozymes coated with polyallylamine hydrochloride (PAM) exhibit significantly enhanced peroxidase (POD)-like activity compared to their counterparts coated with conventional polyvinyl pyrrolidone (PVP) and polystyrene sulfonate (PSS). Leveraging the potent reactivity between PCN and the hydroxyl radicals (∙OH) generated by Ni-Pt@PAM catalytic system, we have developed a colorimetric sensor that directly detects PCN, demonstrating a linear response range from 0.5 to 10 μM and detection limit as low as 13.5 nM. Additionally, a fluorescent sensing platform utilizing self-synthesized fluorophore Rho-Si based on inner filter effect (IFE), has been successfully established. These two sensing platforms, capable of generating both colorimetric and fluorescent signals, enable sensitive, rapid, and straightforward detection of PCN in water samples. This work not only offers an innovative approach for tuning the POD-like activity of Ni-Pt bimetallic nanozymes but also presents a promising avenue for their application in PCN monitoring.

A small-period long-period fiber grating biosensor based on immobilization of concanavalin A on polydopamine nanospheres for simultaneous detection of D-glucose and temperature

The accurate and sensitive monitoring of glucose levels plays a crucial role in diagnosing diabetes in clinical settings. In this work, a novel small-period long-period fiber grating (SP-LPG) biosensor was first developed for D-glucose detection based on the immobilization of concanavalin A (Con A) on polydopamine (PDA) nanospheres. SP-LPG was fabricated by using femtosecond laser direct writing technology and the obtained grating region was only 2.4 mm. The recognition element of D-glucose, Con A, was attached to the surface of PDA nanospheres. Due to the high specific surface area of PDA nanospheres, a large number of binding sites were available for Con A. The as-prepared sensor showed selective and sensitive to D-glucose measurement in the range of 10 nM to 10 mM and a low detection limit of 1.95 nM (S/N = 3) was achieved. Moreover, the small grating period of SP-LPG allows for the observation of Bragg reflection, which can be utilized for temperature sensing. This property eliminates the issue of temperature cross-sensitivity present in refractive index sensing. And the sensor exhibited a thermal sensitivity of 10 pm/°C within the range of 30 ∼ 100 °C. The developed sensor offers a convenient and efficient platform of simultaneously measuring multiple parameters, including target molecules and temperature, for real-time medical monitoring applications.

AI-Enhanced Edge Device for Real-Time Snoring Detection

This paper presents the development of a cutting-edge, non-invasive edge device designed to monitor snoring and provide timely, moderate haptic feedback to users. Utilizing the Qualcomm Snapdragon 8cx Gen 3 processor, the device offers robust computing power and AI capabilities for real-time processing, making it a versatile tool for health monitoring applications. The system integrates a high-fidelity MEMS microphone array capable of capturing nuanced audio signals and a TDK piezoelectric haptic actuator, which delivers precise alerts through customized vibrations. The research explores the potential of this advanced hardware in detecting and managing obstructive sleep apnea (OSA), a condition often underdiagnosed due to a lack of patient awareness. By leveraging state-of-the-art digital signal processing and deep learning techniques, the device aims to enhance user awareness and intervention in sleep-related disorders, offering a promising new avenue for improving patient outcomes and quality of life.

Advancements in triboelectric nanogenerator applications for health monitoring

Advancements in triboelectric nanogenerator applications for health monitoring

In situ assembly enabling adhesive-free bonding of large area electronic sensors to concrete for structural health monitoring

Abstract Cracks developed in concrete infrastructure are one of the primary mechanisms that degrade their structural integrity, which may result in structural failures. Previous research on soft elastomeric capacitors (SEC) has shown their viability for structural health monitoring of structural materials, including concrete, steel, and fiberglass composites. The SEC, or its derivative version with a corrugated geometry termed corrugated SEC or cSEC, is a parallel plate capacitor. Prior work demonstrated that it was possible to directly paint the electrode interfacing with the structural material onto the structure and adhere the rest of the pre-fabricated sensor onto the wet interface, thereby eliminating the need for a joining epoxy. This demonstration was conducted on steel and fiberglass. The study on concrete was left to future work as concrete exhibits a much rougher surface and structure/sensor capacitive coupling, causing a significant amplification in signal noise. This study advances structural health monitoring in concrete applications by investigating the in situ assembly of the SEC on concrete where the carbon black (CB)-based electrode plate of the cSEC is directly painted onto the concrete surface and serves as both the adhesive and electrode for the sensor. A series of free-vibration and compression tests were designed and conducted to evaluate the sensing performance of in situ assembled cSEC compared to that of an epoxy-bonded cSEC. Additionally, the bonding strength of the in situ assembled and epoxy-bonded cSEC is evaluated through a peel test. Results show good strain sensing capabilities of the in situ assembled SEC with an R 2 value of 0.986 and a resolution of 45 ± μ ε . Even though the epoxied cSEC demonstrated a higher bonding strength than the in situ assembled cSEC, the in situ assembled cSEC demonstrated adequate bonding strength for the application. This research contributes to the scientific understanding of sensor adhesion and opens avenues for practical applications in infrastructure monitoring, potentially leading to more resilient and sustainable urban environments.

Self-doped Na-carbon materials derived from a lyocell fiber for a high-performance trimethylamine gas sensor at room temperature

In this study, biomass-derived carbon material obtained from Lyocell fibers was first utilized as a gas sensor. The impacts of varying pyrolytic carbonization temperatures and pregrinding treatments on the structure, surface morphology, elemental composition, and gas sensitivity of the samples were thoroughly examined. The CL-500 sensor can realize rapid detection of trimethylamine with a high response (12.79k%, 500 ppm) and high selectivity at room temperature; the response/recovery times are 10 s and 2 s, respectively, and the theoretical detection limit is 3.96 ppm. Moreover, after four months, the response of the CL-500 sensor to trimethylamine fluctuated by less than 9.7 % compared with that of the fresh sensor, indicating good stability. It also shows good recovery after seven consecutive response-recovery cycles. Additionally, the CL-500 sensor has promising applications in real-life fish freshness monitoring. Theoretical calculations indicate that the introduction of trace amounts of Na enhances the sensing performance of this sensor for target gases. This study serves as a guide for developing cost-effective, high-performance gas sensors, promoting the efficient and high-value utilization of biomass waste.

Edible Ultralong Organic Phosphorescent Excipient for Afterglow Visualizing the Quality of Tablets.

Stimuli-responsive ultralong organic phosphorescence (UOP) materials that in response to external factors such as light, heat, and atmosphere have raised a tremendous research interest in fields of optoelectronics, anticounterfeiting labeling, biosensing, and bioimaging. However, for practical applications in life and health fields, some fundamental requirements such as biocompatibility and biodegradability are still challenging for conventional inorganic and aromatic-based stimuli-responsive UOP systems. Herein, an edible excipient, sodium carboxymethyl cellulose (SCC), of which UOP properties exhibit intrinsically multistimuli responses to excited wavelength, pressure, and moisture, is reported. Impressively, as a UOP probe, SCC enables nondestructive detection of hardness with superb contrast (signal-to-background ratio up to 120), while exhibiting a response sensitivity to moisture that is more than 5.0 times higher than that observed in conventional fluorescence. Additionally, its applicability for hardness monitoring and high-moisture warning for tablets containing a moisture-sensitive drug, with the quality of the drug being determinable through the naked-eye visible UOP, is demonstrated. This work not only elucidates the reason for stimulative corresponding properties in SCC but also makes a major step forward in extending the potential applications of stimuli-responsive UOP materials in manufacturing high-quality and safe medicine.

Benzothiazole-quinoline based probe for simultaneous colorimetric detection of CN- and Cu2+ ions, with fluorescence sensing for Cu2+: Mechanistic insights and practical applications in environmental monitoring and cellular analysis

The development and synthesis of notably targeted and colorimetric sensor based on an azomethine compound for the distinct recognition of Cu2+ and CN− ion individually in an aqueous dimethyl formamide solution is performed. In the presence of CN− and Cu2+, the sensor BTZ showed impressive colorimetric changes, going from pale yellow to orange and pale yellow to dark yellow, respectively. In the meantime, FL spectrum (Cu2+) and UV–vis spectroscopy (CN−/Cu2+) were used to assess the sensing features. The plausible binding mechanisms of CN− and Cu2+ ions with sensor BTZ have been studied using the 1H NMR titration, Job's plot and DFT technique. The bathochromic shift produced by the intramolecular charge transfer (ICT) transition may have been the source of the phenomenon. Furthermore, CN− ion in the commercial substance is quickly identified and measured with the naked eye by using sensor BTZ. It was found that the BTZ's LOD for CN− and Cu2+ was 0.280 × 10−7 M and 1.153 × 10−7 M, respectively. Moreover, 1:1 binding ratio for the reaction of CN− and Cu2+ ions with sensor BTZ were demonstrated by Job's plot, which was dependent on analytical data. The findings show that BTZ is an easy-to-use and practical probe for concurrently sensing cyanide and copper ions in environmental samples and living cells that have less cytotoxicity.

V2CTx MXene-derived porphyrin-based MOF modified with ZIF-67 as an ultra-stretchable and deformable double-network hydrogel for room temperature detection of dimethylamine

Stretchable and self-adhesive chemical sensors are highly appealing for wearable applications in health and environmental monitoring. Here, for the first time, we report V2CTx MXene-derived Porphyrin-based MOF (V-PMOF) modified with Zeolite imidazole framework-67 (ZIF) for the room temperature detection of dimethylamine. V-PMOF is synthesized from V2CTx MXene using tetrakis(4-carboxyphenyl)porphyrin (TCPP) as a ligand via a solvothermal route. Thereafter, an optimized wt% ratio (1:1) of ZIF-67 and V-PMOF crosslinked by PVA/polypyrrole (Ppy) is put for lyophilisation which yields the V-PMOF@ZIF-67 double-network (DN) hydrogel. Detailed morphological assessment reveals dodecahedron shaped ZIF-67 particles are evenly distributed on the V-PMOF sheets which are encapsulated in the polymer matrix. This three-dimensional polymeric structure of hydrogel largely enhances the active sites on the surface with abundant oxygen vacancies. The sensor exhibits a dynamic range of 1 ppm to 13 ppm with a low limit of detection (LOD = 3 s/σ) of 902 ppb towards dimethylamine detection. Further, quick response/recovery times (3 s/7s) are observed along with high robustness tosignificant mechanical deformation, including twist (270°), large-range bending, and upto 500 % strain without affecting the sensing performance. The remarkable sensitivity is accredited to the abundance of O2 functional groups in polymer chains and the formation of schottky heterojunction between V-PMOF and ZIF-67, facilitating charge transfer from the heterostructure to dimethylamine during sensing. In addition, the self-adhesive hydrogel based sensor shows outstanding selectivity against other typical volatile organic compounds (VOCs) such acetone, ethanol, ammonia, trimethylamine, aniline, and methanol. In overall, this work provides insight into designing extremely flexible sensitive dimethylamine gas sensor as innovative material.

New paradigms shift in buildings: experimental application of Digital Twin for safety and well-being

Abstract Enhancing durability and service life of buildings and components is pivotal for sustainable development. It constitutes an opportunity to reduce energy consumption, carbon emissions and life cycle impact of buildings in a climate neutral perspective. Issues strongly introduced and set as priorities by EU policies that highlight the urgent need to tackle them also through deeply heritage renovation and digital transformation in the building sector. Digitalization is a cornerstone of this transformation, instrumental in facilitating and enabling it sustainably. The Key Enabling Technologies (KETs) are directly associated with new technological potential and emerging technologies. Digital Twin (DT) approach appears alongside these. Its experimentation and widespread application are gaining prominence in innovating Life Cycle Management (LCM) and sustainability practices of buildings. In this scenario the paper explores the role and potentialities of DT approach presenting the experimental application of the DT4SEM. A digital infrastructure that, by integrating Internet of Things (IoT), synchronizes a physical building with its virtual counterpart. The two realities (physical and virtual) remain interconnected through the mutual exchange of data, both in real-time and asynchronously enabling proactive monitoring and analysis of seismic behaviour. The experimental setup involves simultaneously Big Data analytics, simulation tools and deploying sensors within the sample building. Data is then fed into the virtual DT model, allowing for continuous comparison and analysis to detect anomalies and predict potential risks. This approach facilitates enhanced decision-making, performance optimization and sustainability improvements across the building’s lifecycle: a new vision for built environment management that evolves from a process resulting in a sequence of operative phases into a complex digital infrastructure. The design of the DT4SEM and the current application for seismic monitoring in buildings, results from a collaborative effort involving the academic spinoff BIG srl, the startup Sysdev, the company Berna Engineering srl, and ACCA Software spa.

Advanced electrochemical detection of carbendazim in staple food samples using cobalt hydroxide/titanium dioxide nanocomposite

This research introduces a novel composite material, cobalt hydroxide/titanium dioxide (Co(OH)2/TiO2), engineered for the detection of the fungicide carbendazim (CBZ) in environmental samples. A glassy carbon electrode (GCE) was modified with Co(OH)2/TiO2 to create an electrochemical sensor capable of highly sensitive CBZ detection. Differential pulse voltammetry was utilized to quantify various CBZ concentrations, revealing that the Co(OH)2/TiO2-GCE sensor provided robust response signals across a concentration range from 0.039 μM to 2630.1 μM, with a detection limit of 0.007 μM. The sensor demonstrated notable stability, reproducibility, and selectivity towards CBZ, attributable to the synergistic effects of Co(OH)2/TiO2. Additionally, the Co(OH)2/TiO2-GCE sensor proved effective for analyzing CBZ in fruit and vegetable samples. Thus, this sensor presents significant potential for widespread application in environmental monitoring and food safety assessments.

Optical resonant cavities carving pathways in tunable wavelength sensitive visible-NIR organic photodetectors

Enhancing the photodetection capabilities of organic photodetectors (OPDs) is crucial for advancing applications in medical monitoring, optical communications, image sensing, and robotics, where a strong, focused peak response at a specific designed wavelength is essential for improving sensitivity, wavelength selectivity, and resolution in imaging systems. By controlled integration of ZnO layers within PBDBT:BTP-4F-based OPDs to form a Fabry-Pérot optical cavity, we developed a cost-effective approach to fabricating highly sensitive OPDs by utilizing PBDBT:BTP-4F organic bulk heterojunctions, and extended its detection wavelengths into the near-infrared (NIR) range. Our design integrates a single silver (Ag) layer that significantly enhances peak detection at a wavelength of 830 ​nm, resulting in a remarkably narrow full-width at half maximum (FWHM) wavelength of 30 ​nm and yielding a photoresponse ten times greater than that of non-resonant devices. Furthermore, by varying the thickness of the ZnO layer from 77 ​nm to 620 ​nm, we achieve high spectral tunability, allowing fine adjustments of the resonant peak across a spectrum ranging from ultraviolet (UV) and visible to NIR wavelengths. This sensitive photodetector is also well-suited for applications in photoplethysmography (PPG), effectively detecting pulse signals in the NIR spectrum which has significant potential in medical diagnostics. This work advances the integration of cost-effective, wavelength-selective spectroscopic visible-NIR OPDs, paving the way for the next generation of sensitive photodetectors.

Individualized medication of venetoclax based on therapeutic drug monitoring in Chinese acute myeloid leukemia patients using an HPLC method.

The aim of this study was to establish a simple and sensitive high-performance liquid chromatography method for therapeutic drug monitoring of venetoclax (VEN) and optimize regimens. The analysis required the extraction of a 50 μl plasma sample and the precipitation of proteins using acetonitrile extraction. The chromatographic method employed a mobile phase of acetonitrile: 0.5% KH 2 PO 4 (pH 3.5) (60/40, v/v) on a Diamond C 18 (4.6 mm × 250 mm, 5 μm) column at a flow rate of 1.0 ml/min. The quantitative method was validated based on standards described in 'Bioanalytical Method Validation: Guidance for Industry' published by the US Food and Drug Administration (FDA). The calibration curve was linear ( R2 = 0.9998) over the range of 75-4800 ng/ml, with limits of quantification of 25 ng/ml. The coefficients of intraday and interday validation, specificity, recovery, and stability all met the criteria of FDA guidance. The method was successfully applied to analyze VEN concentrations in 30 cases of acute myeloid leukemia patients. The peak concentration ( Cmax ) was 1881.19 ± 756.61 ng/ml, while the trough concentration ( Cmin ) was 1212.69 ± 767.92 ng/ml in acute myeloid leukemia patients. Our study establishes a simple, precise, and sensitive high-performance liquid chromatography method for monitoring VEN and confirms its applicability for therapeutic drug monitoring of VEN in hematological cancers.

Hybrid FCMG-OP-FIS model approach to convert regression into classification data for machine learning-based AQI prediction

Air pollution from vehicle emissions, industrial activities, and medical facilities poses significant health risks in urban areas, underscoring the necessity for robust air quality index (AQI) monitoring. This paper presents a novel method for AQI prediction by integrating a fuzzy centre merge graph with an optimal value-based fuzzy inference system (FCMG-OP-FIS) and machine learning (ML). Traditional ML techniques encounter difficulties when converting regression datasets into classification formats, particularly when unable to label the dataset using the traditional method. The proposed FCMG-OP-FIS model efficiently converts regression data into a classification framework. Unlike traditional AQI prediction methods that rely solely on pollutant data, this approach incorporates both pollutant and meteorological data to improve prediction accuracy. The innovative fuzzy centre merge graph (FCMG) balances the dataset for optimal solutions and facilitates input grouping for Simulink, simplifying rule management. The FCMG-OP-FIS model generates a regression output for AQI, which is subsequently classified into levels (healthy, moderate, or unhealthy) using IF-THEN rules. To enhance accuracy further, a random forest classifier (RFC) is trained on the FCMG-OP-FIS classified output data. The regression output of the FCMG-OP-FIS model is validated using metrics such as RMSE (0.48), MSE (0.23), MAE (0.23), and MAPE (1.77%). Additionally, the classification output from the RFC model employs advanced validation techniques including stratified shuffle validation, grid search cross-validation, and confusion matrix analysis, achieving an accuracy rate of 99%, with the F1 score, precision, and recall over all at 99%. These results demonstrate the effectiveness of the proposed model in accurately labelling data for classification and predicting AQI through ML, highlighting its potential for practical application in environmental monitoring and management.

MnO4−-triggered wavelength-changeable and rapid-response fluorescence sensor for paper-based on-site sensing of tyrosinase activity in potato

Rapid-response in situ fluorogenic reactions in aqueous solution are important for designing sensitive and stable sensing platforms. Herein, a wavelength-changeable and rapid-response (within 5 s) fluorescence sensing platform for monitoring tyrosinase (TYR) activity is constructed. The developed assay is based on TYR catalyzing the hydroxylation of mono-phenol to o-diphenol and MnO4−-triggered fluorogenic between dopamine (DA) and phenol derivatives in aqueous solution. The fluorescence wavelength can be changeable from 470 to 550 nm with strong fluorescence according to different phenol derivatives. Our proposed sensor not only exhibits a good recovery for TYR in high serum concentration (20 %), but also has been successfully applied to the screening of TYR inhibitors modeled on kojic acid. Furthermore, a paper-based wavelength-changeable fluorescence sensor was developed for on-site detection of TYR activity in potatoes with high recovery, which is consistent with our previously reported method. Consequently, the proposed sensing system has broad prospects in the practical application of TYR-associated food monitoring and clinical diagnosis.

Z-scheme heterostructure of ZnO@NPC/Au/ZnIn2S4 for chemical replacement reaction-mediated signal-on photoelectrochemical assay of mercury ion

Herein, an advanced Zn-based metal organic framework (Zn-MOF) derived ZnO embedded in a nitrogen-doped porous carbon (ZnO@NPC) polyhedron/Au/ three-dimensional (3D) chrysanthemum-like ZnIn2S Z-scheme heterojunction was employed for the chemical replacement reaction-mediated signal-on photoelectrochemical (PEC) sensing detection of mercury ions (Hg2+). The p-type ZnO@NPC polyhedron was initially prepared by carbonising the zeolitic imidazolate framework (ZIF)-8 polyhedron, followed by modification with Au nanoparticles (NPs) via an in situ photocatalytic reduction method. Subsequently, the ZnO@NPC/Au polyhedron was combined with n-type 3D chrysanthemum-like ZnIn2S4 through physical adsorption, forming the ZnO@NPC polyhedron/Au/3D chrysanthemum-like ZnIn2S4 Z-scheme heterostructure. The newly constructed heterostructure exhibited strong visible light–harvesting capacity, and considerably improved photoelectric conversion efficiency. The 3D chrysanthemum-like ZnIn2S4 functioned not only as a photosensitiser but also as an Hg2+ recognition probe. Importantly, the PEC sensor exhibited a considerably increased photocurrent response to Hg2+. This is presumably because the introduction of Hg2+ triggered a selective Zn-to-Hg chemical replacement reaction, leading to the in-situ formation of a ZnO@NPC/Au/ZnIn2S4/HgS heterojunction, which further improved charge separation. Consequently, Hg2+ was sensitively detected with a wide linear range from 0.5 pM to 2 μM and a low detection limit of 0.07 pM. Furthermore, the proposed sensor was validated by detecting Hg2+ in food and aqueous environments, indicating its potential application in food safety and environmental monitoring.

Rapid Advancements in Large Language Models for Quantitative Remote Sensing: The Case of Water Depth Inversion

This study presents a comparative analysis of two advanced AI models, ChatGPT and ERNIE, in the context of water depth inversion. Utilizing satellite spectral data and in-situ bathymetric measurements collected from Rushikonda Beach, India, we processed and analyzed the data to generate high-resolution bathymetric maps. Both models demonstrated significant accuracy, with ChatGPT slightly outperforming ERNIE in terms of mean absolute error. The study highlights the advantages of AI models, such as efficient data processing and the ability to integrate multi-modal inputs, while also discussing challenges related to data quality, interpretability, and computational demands. The findings suggest that while both models are highly effective for water depth inversion, ongoing improvements in data handling and model transparency are essential for their broader application in environmental monitoring. This research contributes to the understanding of AI capabilities in geospatial analysis and sets the stage for future enhancements in the field.

PrivacyLens: On-Device PII Removal from RGB Images using Thermally-Enhanced Sensing

Internet-connected cameras support many useful home monitoring and health applications. However, these same cameras indiscriminately capture sensitive and Personally Identifiable Information (PII), limiting their acceptance in certain settings, such as the home. Prior works removed Region of Interest (ROI) to secure images and improve privacy. However, the methods that rely solely on RGB information to find persons are susceptible to environmental and lighting conditions, causing them to fail and leak PII. From our deployment study, nearly half of the images containing persons had a PII leakage when using RGB-only methods. Furthermore, ROI removal is often performed off-device, requiring the server performing these operations to be trustworthy. This work presents the PrivacyLens system, where with the addition of thermal sensing, our system has a significantly enhanced ability to find persons in RGB images and video and efficiently remove them on the device before any data is stored or transmitted, all while staying under typical IoT power constraints. From our aforementioned deployment study in an office-building atrium, family home, and outdoor park environment, the PrivacyLens prototype effectively removes PII with a sanitization rate of 99.1%. Additionally, PrivacyLens can use its embedded GPU to generate on-device features for downstream CV/ML tasks, as shown in three illustrative applications, further reducing the collection and storage of PII.