A regression-based machine learning approach for pH and glucose detection with redox-sensitive colorimetric paper sensors

A stretchable, adhesive, and wearable hydrogel-based patches based on a bilayer PVA composite for online monitoring of sweat by artificial intelligence-assisted smartphones

Real-time monitoring of sweat using wearable devices faces challenges such as limited adhesion, mechanical flexibility, and accurate detection. In this work, we present a stretchable, adhesive, bilayer hydrogel-based patch designed for continuous monitoring of sweat pH and glucose levels using AI-assisted smartphones. The patch is composed of a bottom PVA hydrogel layer functionalized with colorimetric reagents and glucose oxidase enzyme, while the top PVA-sucrose layer enhances skin adhesion and protects against air moisture. The hydrogel demonstrates excellent mechanical properties with a tensile strain of 440 % and an elastic modulus of 157 kPa, providing a strong yet flexible interface with the skin. Machine learning models, including random forest (RF) and convolutional neural network (CNN), enabled accurate sweat analysis, achieving a coefficient of determination (R2) of ∼0.99 for pH (3-9) and glucose concentrations up to 0.5 mM. Validation against standard methods like HPLC confirmed the reliability of the patch. This AI-powered system offers a promising platform for next-generation wearable health monitoring devices.

Perfusable cellulose channels from decellularized leaf scaffolds for modeling vascular amyloidosis

Amyloid infiltration in blood vessels damages them and spreads amyloid to surrounding tissues. Research on amyloid flow and deposition in capillaries is limited due to the lack of suitable models. In this study, we created a decellularized leaf scaffold (DCLS) mimicking complex capillary structures to study vascular amyloidosis. Fluorescent molecules (e.g., Nile red) confirmed the intact cellulose framework of the DCLS. Additionally, DCLS with colorimetric nanoparticles (e.g., polyaniline nanoparticles) showed reversible color changes with pH variations, indicating preserved pore structure. The DCLS's responsiveness and preserved vein structures demonstrate its similarity to human vasculature. Hen egg-white lysozyme amyloid deposition was observed in various areas of the DCLS after perfusion. An amyloid-degrading agent (e.g., trypsin) was then perfused, showing a reduction of 18.3 % after 90 min and 25.5 % after 180 min. This DCLS model offers a more realistic and physiologically meaningful platform for studying intravascular amyloid accumulation and clearance than existing in vitro vascular models.

Transfer learning and data augmentation for glucose concentration prediction from colorimetric biosensor images

A deep learning algorithm is introduced to accurately predict glucose concentrations using colorimetric paper sensor (CPS) images. We used an image dataset from CPS treated with five different glucose concentrations as input for deep learning models. Transfer learning was performed by modifying four established deep learning models-ResNet50, ResNet101, GoogLeNet, and VGG-19-to predict glucose concentrations. Additionally, we attempted to alleviate the challenge of requiring the large amount of training data by applying data augmentation techniques. Prediction performance was evaluated using coefficients of determination (R2), root mean squared error (RMSE), and relative-RMSE (rRMSE). GoogLeNet showed the highest coefficient of determination (R2 = 0.994) and significantly lower prediction errors across all concentration levels compared with a traditional machine learning approach (two-sample t-test, p < 0.001). When data augmentation was performed using 20% of the entire training dataset, the mean prediction error was comparable to that of the original entire training dataset. We presented a novel approach for glucose concentration prediction using deep learning techniques based on transfer learning and data augmentation with image data. Our method uses images from CPS as input and eliminates the need for separate feature extraction, simplifying the prediction process.

Convolutional neural network for colorimetric glucose detection using a smartphone and novel multilayer polyvinyl film microfluidic device

Detecting glucose levels is crucial for diabetes patients as it enables timely and effective management, preventing complications and promoting overall health. In this endeavor, we have designed a novel, affordable point-of-care diagnostic device utilizing microfluidic principles, a smartphone camera, and established laboratory colorimetric methods for accurate glucose estimation. Our proposed microfluidic device comprises layers of adhesive poly-vinyl films stacked on a poly methyl methacrylate (PMMA) base sheet, with micro-channel contours precision-cut using a cutting printer. Employing the gold standard glucose-oxidase/peroxidase reaction on this microfluidic platform, we achieve enzymatic glucose determination. The resulting colored complex, formed by phenol and 4-aminoantipyrine in the presence of hydrogen peroxide generated during glucose oxidation, is captured at various glucose concentrations using a smartphone camera. Raw images are processed and utilized as input data for a 2-D convolutional neural network (CNN) deep learning classifier, demonstrating an impressive 95% overall accuracy against new images. The glucose predictions done by CNN are compared with ISO 15197:2013/2015 gold standard norms. Furthermore, the classifier exhibits outstanding precision, recall, and F1 score of 94%, 93%, and 93%, respectively, as validated through our study, showcasing its exceptional predictive capability. Next, a user-friendly smartphone application named "GLUCOLENS AI" was developed to capture images, perform image processing, and communicate with cloud server containing the CNN classifier. The developed CNN model can be successfully used as a pre-trained model for future glucose concentration predictions.

Role of Machine Learning Assisted Biosensors in Point-of-Care-Testing For Clinical Decisions

Point-of-Care-Testing (PoCT) has emerged as an essential component of modern healthcare, providing rapid, low-cost, and simple diagnostic options. The integration of Machine Learning (ML) into biosensors has ushered in a new era of innovation in the field of PoCT. This article investigates the numerous uses and transformational possibilities of ML in improving biosensors for PoCT. ML algorithms, which are capable of processing and interpreting complicated biological data, have transformed the accuracy, sensitivity, and speed of diagnostic procedures in a variety of healthcare contexts. This review explores the multifaceted applications of ML models, including classification and regression, displaying how they contribute to improving the diagnostic capabilities of biosensors. The roles of ML-assisted electrochemical sensors, lab-on-a-chip sensors, electrochemiluminescence/chemiluminescence sensors, colorimetric sensors, and wearable sensors in diagnosis are explained in detail. Given the increasingly important role of ML in biosensors for PoCT, this study serves as a valuable reference for researchers, clinicians, and policymakers interested in understanding the emerging landscape of ML in point-of-care diagnostics.

Capillary Flow-Based One-Minute Quantification of Amyloid Proteolysis

Quantifying the formation and decomposition of amyloid is a crucial issue in the development of new drugs and therapies for treating amyloidosis. The current technologies for grasping amyloid formation and decomposition include fluorescence analysis using thioflavin-T, secondary structure analysis using circular dichroism, and image analysis using atomic force microscopy or transmission electron microscopy. These technologies typically require spectroscopic devices or expensive nanoscale imaging equipment and involve lengthy analysis, which limits the rapid screening of amyloid-degrading drugs. In this study, we introduce a technology for rapidly assessing amyloid decomposition using capillary flow-based paper (CFP). Amyloid solutions exhibit gel-like physical properties due to insoluble denatured polymers, resulting in a shorter flow distance on CFP compared to pure water. Experimental conditions were established to consistently control the flow distance based on a hen-egg-white lysozyme amyloid solution. It was confirmed that as amyloid is decomposed by trypsin, the flow distance increases on the CFP. Our method is highly useful for detecting changes in the gel properties of amyloid solutions within a minute, and we anticipate its use in the rapid, large-scale screening of anti-amyloid agents in the future.

Glucose Oxidase Activity Colorimetric Assay Using Redox-Sensitive Electrochromic Nanoparticle-Functionalized Paper Sensors

Glucose oxidase (GOx) activity assays are vital for various applications, including glucose metabolism estimation and fungal testing. However, conventional methods involve time-consuming and complex procedures. In this study, we present a colorimetric platform for in situ GOx activity measurement utilizing redox-sensitive electrochromic nanoparticles based on polyaniline (PAni). The glucose-adsorbed colorimetric paper sensor, herein termed Glu@CPS, is created by immobilizing ferrocene and glucose onto paper substrates that have been functionalized with PAni nanoparticles. Glu@CPS not only demonstrated rapid detection (within 5 min) but also exhibited remarkable selectivity for GOx and a limit of detection as low as 1.25 μM. Moreover, Glu@CPS demonstrated consistent accuracy in the measurement of GOx activity, exhibiting no deviations even after being stored at ambient temperature for a duration of one month. To further corroborate the effectiveness of this method, we applied Glu@CPS in the detection of GOx activity in a moldy red wine. The results highlight the promising potential of Glu@CPS as a convenient and precise platform for GOx activity measurement in diverse applications including food quality control, environmental monitoring, and early detection of fungal contamination.