Colorimetric Sensor Reading and Illumination Correction via Multi-Task Deep-Learning.

Colorimetric sensors represent an accessible and sensitive nanotechnology for rapid and accessible measurement of a substance's properties (e.g., analyte concentration) via color changes. Although colorimetric sensors are widely used in healthcare and laboratories, interpretation of their output is performed either by visual inspection or using cameras in highly controlled illumination set-ups, limiting their usage in end-user applications, with lower resolutions and altered light conditions. For that purpose, we implement a set of image processing and deep-learning (DL) methods that correct for non-uniform illumination alterations and accurately read the target variable from the color response of the sensor. Methods that perform both tasks independently vs. jointly in a multi-task model are evaluated. Video recordings of colorimetric sensors measuring temperature conditions were collected to build an experimental reference dataset. Sensor images were augmented with non-uniform color alterations. The best-performing DL architecture disentangles the luminance, chrominance, and noise via separate decoders and integrates a regression task in the latent space to predict the sensor readings, achieving a mean squared error (MSE) performance of 0.811±0.074[°C] and r2=0.930±0.007, under strong color perturbations, resulting in an improvement of 1.26[°C] when compared to the MSE of the best performing method with independent denoising and regression tasks.Clinical Relevance- The proposed methodology aims to improve the accuracy of colorimetric sensor reading and their large-scale accessibility as point-of-care diagnostic and continuous health monitoring devices, in altered illumination conditions.

Reshaping de Novo Protein Switches into Bioresponsive Materials for Biomarker, Toxin, and Viral Detection.

De novo designed protein switches are powerful tools to specifically and sensitively detect diverse targets with simple chemiluminescent readouts. Finding an appropriate material host for de novo designed protein switches without altering their thermodynamics while preserving their intrinsic stability over time would enable the development of a variety of sensing formats to monitor exposure to pathogens, toxins, and for disease diagnosis. Here, a de novo protein-biopolymer hybrid that maintains the detection capabilities induced by the conformational change of the incorporated proteins in response to analytes of interest is generated in multiple, shelf-stable material formats without the need of refrigerated storage conditions. A set of functional demonstrator devices including personal protective equipment such as masks and laboratory gloves, free-standing films, air quality monitors, and wearable devices is presented to illustrate the versatility of the approach. Such formats are designed to be responsive to human epidermal growth factor receptor (HER2), anti-hepatitis B (HBV) antibodies, Botulinum neurotoxin B (BoNT/B), and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). This combination of form and function offers wide opportunities for ubiquitous sensing in multiple environments by enabling a large class of bio-responsive interfaces of broad utility.

Stabilization of Salivary Biomarkers.

The use of saliva as a diagnostic biofluid has been increasing in recent years, thanks to the identification and validation of new biomarkers and improvements in test accuracy, sensitivity, and precision that enable the development of new noninvasive and cost-effective devices. However, the lack of standardized methods for sample collection, treatment, and storage contribute to the overall variability and lack of reproducibility across analytical evaluations. Furthermore, the instability of salivary biomarkers after sample collection hinders their translation into commercially available technologies for noninvasive monitoring of saliva in home settings. The present review aims to highlight the status of research on the challenges of collecting and using diagnostic salivary samples, emphasizing the methodologies used to preserve relevant proteins, hormones, genomic, and transcriptomic biomarkers during sample handling and analysis.

Functionalized Mouth-Conformable Interfaces for pH Evaluation of the Oral Cavity.

Oral health monitoring is highly desired, especially for in home use, however, current methods are not sensitive enough and technically convoluted for this purpose. This paper presents incorporation of bioactive materials and colorimetric chemical sensors into routinely used oral appliances transforming them into bioresponsive, conformable interfaces. Specifically, endodontic paper points and dental floss can be functionalized to locally sense and monitor pH variations within the oral cavity via color changes. Moreover, edible colorimetric indicators are developed and used to make sensing, edible devices in the form factor of candies that can dynamically and visually respond to acidity changes in saliva. These interfaces would enable early detection of caries (e.g., using dental floss and paper points) providing low-cost point of care devices that respond in real-time by detecting pH variations in biological fluids thus bringing monitoring to home settings .

Large-Scale Patterning of Reactive Surfaces for Wearable and Environmentally Deployable Sensors.

Wearable interfaces are central to multiple healthcare and wellness strategies encompassing diet and nutrition, personalized health monitoring, and performance optimization. Specifically, the advent of flexible electronic formats coupled with microfluidic interfaces has resulted in sophisticated conformal devices for biofluid sampling and quantification. Here, a complementary approach is presented to wearable sensing by using a large-scale, conformal, distributed format that relies on the use of biomaterial-based inks to print and stabilize deterministic patterns of biochemical reporters with high resolution. Colorimetric devices can vary in size and a sensing T-shirt based on a colorimetric pattern is developed to illustrate the utility that such formats can add to the wearable interface space. Image analysis allows parameter variation to be tracked in real-time, yielding a map-like format of distributed biophysical response.

A Wearable Device for Monitoring Sweat Rates via Image Analysis.

A feasibility study on a new technique capable of monitoring localized sweat rate is explored in this paper. Wearable devices commonly used in clinical practice for sweat sampling (i.e., Macroducts) were positioned on the body of an athlete whose sweat rate was then monitored during cycling sessions. The position at which the sweat fills the Macroduct was indicated by a contrasting marker and captured via a series of time-stamped photos or a video recording of the device during an exercise period. Given that the time of each captured image/frame is known (either through time stamp on photos or the constant frame rate of the video capture), it was, therefore, possible to estimate the sweat flow rate through a simple calibration model. The importance of gathering such valuable information is described, together with the results from a number of exercise trials to investigate the viability of this approach.

A liquid-junction-free reference electrode based on a PEDOT solid-contact and ionogel capping membrane.

Liquid-junction-free reference electrodes have been prepared on screen printed substrates using poly-3,4-ethylenedioxythiophene (PEDOT) as solid-contact and novel ionogels as capping membrane. The chemico-physical properties of the PEDOT layer were tuned by changing the electropolymerization media and the electrodeposition technique. Electrodepositing PEDOT films potentiostatically or potentiodynamically were found to have a significant impact on the stability of the electrodes during the conditioning step. Optimization of the capping membrane formulation, e.g., acrylate monomers, ionic liquid, cross-linkers and photo-initiators, produced electrodes with properties almost equivalent with a commercial reference electrode. Thus, calibration plots of Na(+) ion-selective electrodes against the optimized solid-contact ionogel reference electrodes (SCI-REs) or against a double-liquid junction Ag/AgCl electrode did not present any significant difference. Such SCI-REs may provide an effective route to the generation of future low-cost components for potentiometric sensing strips.