Federated clustered multi-domain learning for health monitoring.

Wearable Internet of Things (WIoT) and Artificial Intelligence (AI) are rapidly emerging technologies for healthcare. These technologies enable seamless data collection and precise analysis toward fast, resource-abundant, and personalized patient care. However, conventional machine learning workflow requires data to be transferred to the remote cloud server, which leads to significant privacy concerns. To tackle this problem, researchers have proposed federated learning, where end-point users collaboratively learn a shared model without sharing local data. However, data heterogeneity, i.e., variations in data distributions within a client (intra-client) or across clients (inter-client), degrades the performance of federated learning. Existing state-of-the-art methods mainly consider inter-client data heterogeneity, whereas intra-client variations have not received much attention. To address intra-client variations in federated learning, we propose a federated clustered multi-domain learning algorithm based on ClusterGAN, multi-domain learning, and graph neural networks. We applied the proposed algorithm to a case study on stress-level prediction, and our proposed algorithm outperforms two state-of-the-art methods by 4.4% in accuracy and 0.06 in the F1 score. In addition, we demonstrate the effectiveness of the proposed algorithm by investigating variants of its different modules.

Timeline Registration for Electronic Health Records.

Electronic Health Record (EHR) data are captured over time as patients receive care. Accordingly, variations among patients, such as when a patient presents for care during the course of a disease, introduce bias into standard longitudinal EHR data analysis methods. We, therefore, aim to provide an alignment method that reduces this bias. We structure this task as a registration problem. While limited prior research on longitudinal EHR data considered registration, we propose a robust registration method to provide better data alignment by estimating the optimum time shift at each time point. We validate the proposed method for mortality prediction. We utilize a Recurrent Neural Network (RNN), time-varying Cox regression model, and Logistic Regression (LR) for mortality prediction. Results suggest our proposed registration method enhances mortality prediction with at least a 1-2% increase in major evaluation metrics utilized.

Material-level countermeasures for securing microfluidic biochips.

Flow-based microfluidic biochips (FMBs) have been rapidly commercialized and deployed in recent years for biological computing, clinical diagnostics, and point-of-care-tests (POCTs). However, outsourcing FMBs makes them susceptible to material-level attacks by malicious actors for illegitimate monetary gain. The attacks involve deliberate material degradation of an FMB's polydimethylsiloxane (PDMS) components by either doping with reactive solvents or altering the PDMS curing ratio during fabrication. Such attacks are stealthy enough to evade detection and deteriorate the FMB's function. Furthermore, material-level attacks can become prevalent in attacks based on intellectual property (IP) theft, such as counterfeiting, overbuilding, etc., which involve unscrupulous third-party manufacturers. To address this problem, we present a dynamic material-level watermarking scheme for PDMS-based FMBs with microvalves using a perylene-labeled fluorescent dye. The dyed microvalves show a unique excimer intensity peak under 405 nm laser excitation. Moreover, when pneumatically actuated, the peak shows a predetermined downward shift in intensity as a function of mechanical strain. We validated this protection scheme experimentally using fluorescence microscopy, which showed a high correlation (R2 = 0.971) between the normalized excimer intensity change and the maximum principal strain of the actuated microvalves. To detect curing ratio-based attacks, we adapted machine learning (ML) models, which were trained on the force-displacement data obtained from a mechanical punch test method. Our ML models achieved more than 99% accuracy in detecting curing ratio anomalies. These countermeasures can be used to proactively safeguard FMBs against material-level attacks in the era of global pandemics and diagnostics based on POCTs.

Sensor-based evaluation of a Urine Trap toilet in a shared bathroom.

Urine diversion in a No-Mix Toilet is a promising approach for sustainable fertilizers and reduction of the nutrient load for wastewater treatment; however, user adoption remains a challenge. This study evaluates the Urine Trap, a passive No-Mix toilet design based on the teapot effect, wherein the urine stream inlet is invisible to the user and therefore it does not impact the user experience for increased adoption. This study evaluated the nutrient separation performance of a Urine Trap flush toilet in a bathroom shared by women in two sites in India. Over three different testing periods, 841 uses of this squat plate were recorded in 50 days. Analytical measurements found 36 % separation efficiency for total nitrogen (TN). While effective, the Urine Trap under test by users did not yield a 70-80 % TN separation efficiency observed under engineering characterization. High temporal resolution data from sensors on waste collection tanks, the opening of the bathroom door, and cleansing water flow were used to gain insights into hygiene practices. The data showed a frequent habit of wetting the squat plate during physiological excretion, a hygienic practice that eases cleaning but degrades the teapot separation effect of the Urine Trap design. By using sensors, we demonstrate a method to non-invasively gain quantitative insights into hygiene practices to inform sanitation technologies deployment strategies for improved outcomes.

Structural Attacks and Defenses for Flow-Based Microfluidic Biochips.

Flow-based microfluidic biochips (FMBs) have seen rapid commercialization and deployment in recent years for point-of-care and clinical diagnostics. However, the outsourcing of FMB design and manufacturing makes them susceptible to susceptible to malicious physical level and intellectual property (IP)-theft attacks. This work demonstrates the first structure-based (SB) attack on representative commercial FMBs. The SB attacks maliciously decrease the heights of the FMB reaction chambers to produce false-negative results. We validate this attack experimentally using fluorescence microscopy, which showed a high correlation ( R2 = 0.987) between chamber height and related fluorescence intensity of the DNA amplified by polymerase chain reaction. To detect SB attacks, we adopt two existing deep learning-based anomaly detection algorithms with  âˆ¼ 96% validation accuracy in recognizing such deliberately introduced microstructural anomalies. To safeguard FMBs against intellectual property (IP)-theft, we propose a novel device-level watermarking scheme for FMBs using intensity-height correlation. The countermeasures can be used to proactively safeguard FMBs against SB and IP-theft attacks in the era of global pandemics and personalized medicine.

Acoustohydrodynamic tweezers via spatial arrangement of streaming vortices.

Acoustics-based tweezers provide a unique toolset for contactless, label-free, and precise manipulation of bioparticles and bioanalytes. Most acoustic tweezers rely on acoustic radiation forces; however, the accompanying acoustic streaming often generates unpredictable effects due to its nonlinear nature and high sensitivity to the three-dimensional boundary conditions. Here, we demonstrate acoustohydrodynamic tweezers, which generate stable, symmetric pairs of vortices to create hydrodynamic traps for object manipulation. These stable vortices enable predictable control of a flow field, which translates into controlled motion of droplets or particles on the operating surface. We built a programmable droplet-handling platform to demonstrate the basic functions of planar-omnidirectional droplet transport, merging droplets, and in situ mixing via a sequential cascade of biochemical reactions. Our acoustohydrodynamic tweezers enables improved control of acoustic streaming and demonstrates a previously unidentified method for contact-free manipulation of bioanalytes and digitalized liquid handling based on a compact and scalable functional unit.

Acoustoelectronic nanotweezers enable dynamic and large-scale control of nanomaterials.

The ability to precisely manipulate nano-objects on a large scale can enable the fabrication of materials and devices with tunable optical, electromagnetic, and mechanical properties. However, the dynamic, parallel manipulation of nanoscale colloids and materials remains a significant challenge. Here, we demonstrate acoustoelectronic nanotweezers, which combine the precision and robustness afforded by electronic tweezers with versatility and large-field dynamic control granted by acoustic tweezing techniques, to enable the massively parallel manipulation of sub-100 nm objects with excellent versatility and controllability. Using this approach, we demonstrated the complex patterning of various nanoparticles (e.g., DNAs, exosomes, ~3 nm graphene flakes, ~6 nm quantum dots, ~3.5 nm proteins, and ~1.4 nm dextran), fabricated macroscopic materials with nano-textures, and performed high-resolution, single nanoparticle manipulation. Various nanomanipulation functions, including transportation, concentration, orientation, pattern-overlaying, and sorting, have also been achieved using a simple device configuration. Altogether, acoustoelectronic nanotweezers overcome existing limitations in nano-manipulation and hold great potential for a variety of applications in the fields of electronics, optics, condensed matter physics, metamaterials, and biomedicine.

Hardware Design and Fault-Tolerant Synthesis for Digital Acoustofluidic Biochips.

A digital microfluidic biochip (DMB) is an attractive platform for automating laboratory procedures in microbiology. To overcome the problem of cross-contamination due to fouling of the electrode surface in traditional DMBs, a contactless liquid-handling biochip technology, referred to as acoustofluidics, has recently been proposed. A major challenge in operating this platform is the need for a control signal of frequency 24 MHz and voltage range ±10/±20 V to activate the IDT units in the biochip. In this paper, we present a hardware design that can efficiently activate/de-activated each IDT, and can fully automate an bio-protocol. We also present a fault-tolerant synthesis technique that allows us to automatically map biomolecular protocols to acoustofluidic biochips. We develop and experimentally validate a velocity model, and use it to guide co-optimization for operation scheduling, module placement, and droplet routing in the presence of IDT faults. Simulation results demonstrate the effectiveness of the proposed synthesis method. Our results are expected to open new research directions on design automation of digital acoustofluidic biochips.

Sensor-Array Optimization Based on Time-Series Data Analytics for Sanitation-Related Malodor Detection.

There is an unmet need for a low-cost instrumented technology for detecting sanitation-related malodor as an alert for maintenance around shared toilets and emerging technologies for onsite waste treatment. In this article, our approach to an electronic nose for sanitation-related malodor is based on the use of electrochemical gas sensors, and machine-learning techniques for sensor selection and odor classification. We screened 10 sensors from different vendors with specific target gases and recorded their response to malodor from fecal specimens and urine specimens, and confounding good odors such as popcorn. The analysis of 180 odor exposures data by two feature-selection methods based on mutual information indicates that, for malodor detection, the decision tree (DT) classifier with seven features from four sensors provides 88.0% balanced accuracy and 85.1% macro F1 score. However, the k-nearest-neighbors (KNN) classifier with only three features (from two sensors) obtains 83.3% balanced accuracy and 81.3% macro F1 score. For classification of urine against feces malodor, a balanced accuracy of 94.0% and a macro F1 score of 92.9% are achieved using only four features from three sensors and a logistic regression (LR) classifier.

Acoustic streaming vortices enable contactless, digital control of droplets.

Advances in lab-on-a-chip technologies are driven by the pursuit of programmable microscale bioreactors or fluidic processors that mimic electronic functionality, scalability, and convenience. However, few fluidic mechanisms allow for basic logic operations on rewritable fluidic paths due to cross-contamination, which leads to random interference between "fluidic bits" or droplets. Here, we introduce a mechanism that allows for contact-free gating of individual droplets based on the scalable features of acoustic streaming vortices (ASVs). By shifting the hydrodynamic equilibrium positions inside interconnected ASVs with multitonal electrical signals, different functions such as controlling the routing and gating of droplets on rewritable fluidic paths are demonstrated with minimal biochemical cross-contamination. Electrical control of this ASV-based mechanism allows for unidirectional routing and active gating behaviors, which can potentially be scaled to functional fluidic processors that can regulate the flow of droplets in a manner similar to the current in transistor arrays.

Micro-Electrode-Dot-Array Digital Microfluidic Biochips: Technology, Design Automation, and Test Techniques.

Digital microfluidic biochips (DMFBs) are being increasingly used for DNA sequencing, point-of-care clinical diagnostics, and immunoassays. DMFBs based on a micro-electrode-dot-array (MEDA) architecture have recently been proposed, and fundamental droplet manipulations, e.g., droplet mixing and splitting, have also been experimentally demonstrated on MEDA biochips. There can be thousands of microelectrodes on a single MEDA biochip, and the fine-grained control of nanoliter volumes of biochemical samples and reagents is also enabled by this technology. MEDA biochips offer the benefits of real-time sensitivity, lower cost, easy system integration with CMOS modules, and full automation. This review paper first describes recent design tools for high-level synthesis and optimization of map bioassay protocols on a MEDA biochip. It then presents recent advances in scheduling of fluidic operations, placement of fluidic modules, droplet-size-aware routing, adaptive error recovery, sample preparation, and various testing techniques. With the help of these tools, biochip users can concentrate on the development of nanoscale bioassays, leaving details of chip optimization and implementation to software tools.

Contactless, programmable acoustofluidic manipulation of objects on water.

Contact-free manipulation of small objects (e.g., cells, tissues, and droplets) using acoustic waves eliminates physical contact with structures and undesired surface adsorption. Pioneering acoustic-based, contact-free manipulation techniques (e.g., acoustic levitation) enable programmable manipulation but are limited by evaporation, bulky transducers, and inefficient acoustic coupling in air. Herein, we report an acoustofluidic mechanism for the contactless manipulation of small objects on water. A hollow-square-shaped interdigital transducer (IDT) is fabricated on lithium niobate (LiNbO3), immersed in water and used as a sound source to generate acoustic waves and as a micropump to pump fluid in the ±x and ±y orthogonal directions. As a result, objects which float adjacent to the excited IDT can be pushed unidirectionally (horizontally) in ±x and ±y following the directed acoustic wave propagation. A fluidic processor was developed by patterning IDT units in a 6-by-6 array. We demonstrate contactless, programmable manipulation on water of oil droplets and zebrafish larvae. This acoustofluidic-based manipulation opens avenues for the contactless, programmable processing of materials and small biosamples.

Droplet Size-Aware and Error-Correcting Sample Preparation Using Micro-Electrode-Dot-Array Digital Microfluidic Biochips.

Sample preparation in digital microfluidics refers to the generation of droplets with target concentrations for on-chip biochemical applications. In recent years, digital microfluidic biochips (DMFBs) have been adopted as a platform for sample preparation. However, there remain two major problems associated with sample preparation on a conventional DMFB. First, only a (1:1) mixing/splitting model can be used, leading to an increase in the number of fluidic operations required for sample preparation. Second, only a limited number of sensors can be integrated on a conventional DMFB; as a result, the latency for error detection during sample preparation is significant. To overcome these drawbacks, we adopt a next generation DMFB platform, referred to as micro-electrode-dot-array (MEDA), for sample preparation. We propose the first sample-preparation method that exploits the MEDA-specific advantages of fine-grained control of droplet sizes and real-time droplet sensing. Experimental demonstration using a fabricated MEDA biochip and simulation results highlight the effectiveness of the proposed sample-preparation method.

Droplet Size-Aware High-Level Synthesis for Micro-Electrode-Dot-Array Digital Microfluidic Biochips.

A digital microfluidic biochip (DMFB) is an attractive technology platform for automating laboratory procedures in biochemistry. In recent years, DMFBs based on a microelectrode-dot-array (MEDA) architecture have been demonstrated. However, due to the inherent differences between today's DMFBs and MEDA, existing synthesis solutions for biochemistry mapping cannot be utilized for MEDA biochips. We present the first synthesis approach that can be used for MEDA biochips. We first present a general analytical model for droplet velocity and validate it experimentally using a fabricated MEDA biochip. We then present the proposed synthesis method targeting reservoir placement, operation scheduling, module placement, routing of droplets of various sizes, and diagonal movement of droplets in a two-dimensional array. Simulation results using benchmarks and experimental results using a fabricated MEDA biochip demonstrate the effectiveness of the proposed synthesis technique.

Security Assessment of Cyberphysical Digital Microfluidic Biochips.

A digital microfluidic biochip (DMFB) is an emerging technology that enables miniaturized analysis systems for point-of-care clinical diagnostics, DNA sequencing, and environmental monitoring. A DMFB reduces the rate of sample and reagent consumption, and automates the analysis of assays. In this paper, we provide the first assessment of the security vulnerabilities of DMFBs. We identify result-manipulation attacks on a DMFB that maliciously alter the assay outcomes. Two practical result-manipulation attacks are shown on a DMFB platform performing enzymatic glucose assay on serum. In the first attack, the attacker adjusts the concentration of the glucose sample and thereby modifies the final result. In the second attack, the attacker tampers with the calibration curve of the assay operation. We then identify denial-of-service attacks, where the attacker can disrupt the assay operation by tampering either with the droplet-routing algorithm or with the actuation sequence. We demonstrate these attacks using a digital microfluidic synthesis simulator. The results show that the attacks are easy to implement and hard to detect. Therefore, this work highlights the need for effective protections against malicious modifications in DMFBs.

Digital Microfluidic Logic Gates and Their Application to Built-in Self-Test of Lab-on-Chip.

Dependability is an important system attribute for microfluidic lab-on-chip. Robust testing methods are therefore needed to ensure correct results. Previously proposed techniques for reading test outcomes and for pulse-sequence analysis are cumbersome and error prone. We present a built-in self-test (BIST) method for digital microfluidic lab-on-chip. This method utilizes digital microfluidic logic gates to implement the BIST architecture; AND, OR and NOT gates are used to compress multiple test-outcome droplets into a single droplet to facilitate detection with low overhead. These approaches obviate the need for capacitive sensing test-outcome circuits for analysis. We also apply the BIST architecture to a pin-constrained biochip design. A multiplexed bioassay protocol is used to evaluate the effectiveness of the proposed test method.