Artificial Sensors Research Articles

Artificial Intelligence to Support the Training and Assessment of Professionals: A Systematic Literature Review

Advances in Artificial Intelligence (AI) and sensors are significantly impacting multiple areas, including education and workplaces. Following the PRISMA methodology, this review explores the current status of using AI to support the training and assessment of professionals. We examined 83 research papers, analyzing (1) the targeted professionals, (2) the skills assessed, (3) the AI algorithms utilized, (4) the data and devices employed, (5) data fusion techniques utilized, (6) the architecture of the proposed platforms, (7) the management of ethics and privacy, and (8) validations of the proposals. The review highlights a trend in evaluating healthcare professionals (especially surgeons) motivated by the critical role of hands-on training in these professions. Besides, the review reveals that data fusion techniques and certain technologies, like transfer learning and explainable AI, are not widely utilized despite their huge potential. Finally, the review underscores that most proposals remain within the research domain, lacking the integration and maturity needed for sustained use in real-world environments. Therefore, most of the proposals are not currently available to support the training of professionals. The insights of this review can guide researchers aiming to improve the training of professionals and, consequently, their education.

Unlocking agro-ecosystem sustainability: exploring the bottom-up effects of microbes, plants, and insect herbivores.

Agricultural ecosystem formation and evolution depend on interactions and communication between multiple organisms. Within this context, communication occurs between microbes, plants, and insects, often involving the release and perception of a wide range of chemical cues. Unraveling how this information is coded and interpreted is critical to expanding our understanding of how agricultural ecosystems function in terms of competition and cooperation. Investigations examining dual interactions (e.g. plant-microbe, insect-microbe, and insect-plant) have resolved some basic components of this communication. However, there is a need for systematically examining multitrophic interactions that occur simultaneously between microorganisms, insects, and plants. A more thorough understanding of these multitrophic interactions has been made possible by recent advancements in the study of such ecological interactions, which are based on a variety of contemporary technologies such as artificial intelligence sensors, multi-omics, metabarcoding, and others. Frequently, these developments have led to the discovery of startling examples of each member manipulating the other. Here, we review recent advances in the understanding of bottom-up chemical communication between microorganisms, plants, and insects, and their consequences. We discuss the components of these "chemo-languages" and how they modify outcomes of multi-species interactions across trophic levels. Further, we suggest prospects for translating the current basic understanding of multitrophic interactions into strategies that could be applied in agricultural ecosystems to increase food safety and security.

Data-Driven Sparse Sensor Placement Optimization on Wings for Flight-By-Feel: Bioinspired Approach and Application.

Flight-by-feel (FBF) is an approach to flight control that uses dispersed sensors on the wings of aircraft to detect flight state. While biological FBF systems, such as the wings of insects, often contain hundreds of strain and flow sensors, artificial systems are highly constrained by size, weight, and power (SWaP) considerations, especially for small aircraft. An optimization approach is needed to determine how many sensors are required and where they should be placed on the wing. Airflow fields can be highly nonlinear, and many local minima exist for sensor placement, meaning conventional optimization techniques are unreliable for this application. The Sparse Sensor Placement Optimization for Prediction (SSPOP) algorithm extracts information from a dense array of flow data using singular value decomposition and linear discriminant analysis, thereby identifying the most information-rich sparse subset of sensor locations. In this research, the SSPOP algorithm is evaluated for the placement of artificial hair sensors on a 3D delta wing model with a 45° sweep angle and a blunt leading edge. The sensor placement solution, or design point (DP), is shown to rank within the top one percent of all possible solutions by root mean square error in angle of attack prediction. This research is the first to evaluate SSPOP on a 3D model and the first to include variable length hairs for variable velocity sensitivity. A comparison of SSPOP against conventional greedy search and gradient-based optimization shows that SSPOP DP ranks nearest to optimal in over 90 percent of models and is far more robust to model variation. The successful application of SSPOP in complex 3D flows paves the way for experimental sensor placement optimization for artificial hair-cell airflow sensors and is a major step toward biomimetic flight-by-feel.

Enhancing the Heat Dissipation Efficiency of Computing Units Within Autonomous Driving Systems and Electric Vehicles

With the rapid development of autonomous driving technology and electric vehicles, the demand for computing units in vehicles has surged, particularly for high-density computing related to artificial intelligence, sensor fusion, and real-time data processing. This increase has posed significant heat dissipation challenges for automotive electronic systems. Effective heat dissipation is essential for ensuring system stability, safety, and extending the lifespan of these systems. This study explores the thermal design of computing units in current autonomous driving systems and electric vehicles. The electronic components within the computing units generate significant heat, but due to environmental constraints, fans can not be used for internal cooling, as their operation can draw in dust from the environment, contaminating the internal electronic components. Therefore, computing units are often designed with a sealed structure, utilizing the thermal conductivity of metal casings to quickly dissipate internal heat sources. The results of this study show that comparing embedded cooling modules within the casing to commonly used thermal pads reveals that changing the materials of the cooling module can significantly enhance cooling performance, noticeably reducing the temperature of the internal electronic components and minimizing the risk of overheating. Additionally, this can improve the overall reliability and performance of the system, providing new ideas and application prospects for future automotive thermal design.

Self‐Powered Iontronic Capacitive Sensing Unit with High Sensitivity in Charge‐Output Mode

AbstractThe operation of iontronic capacitive sensors typically requires an external alternating current (AC) power source, resulting in additional energy consumption and AC‐frequency‐related sensing performance. Here, a class of self‐powered iontronic capacitive sensing units (SICSUs) is proposed based on a dynamic electric double layer (EDL), with a significant charge sensitivity of up to 24270 pC N−1, surpassing most piezoelectric materials by nearly 10 times. The effects of various design parameters and loading conditions on the sensing performance of the SICSUs are systematically investigated. The EDL at the hydrogel‐electrode interface is characterized in situ, revealing the underlying mechanism for high sensitivity and linearity. The capability of SICSUs in detecting diverse human‐related mechanical loads is demonstrated. Furthermore, a robotic hand equipped with a SICSU‐based artificial algesia sensor is fabricated to mimic the withdrawal reflex behavior of a human hand when its skin detects noxious stimuli caused by sharp objects.

Infants Sucking Pattern Identification Using Machine-Learned Computational Modeling

Abstract Breastfeeding involves a complex coordination of swallowing, breathing, and sucking, with the infant's sucking proficiency being crucial for adequate nutrient intake. However, real-time assessment of milk intake is difficult, and issues with sucking often become apparent after the infant shows signs of nutrient deficiency. Traditional assessments by clinicians rely on the expertise and subjective judgment of healthcare professionals, enabling personalized evaluations. In this research, we introduce a novel approach to identifying sucking patterns by leveraging data collected from infants during breastfeeding sessions. This method utilizes artificial nipple-based sensors to capture the tongue forces exerted by infants, generating valuable clinical data. In the analysis of the collected time-series data, we applied machine-learned computational modeling (MLCM) algorithms to extract pertinent features and identify distinctive sucking patterns. The best-performing model demonstrated an accuracy of 90%, an 80% recall score, a perfect 100% precision score, a 0.90 f1-score, and an area under the curve (AUC) of 0.80. The proposed classification system has the potential to serve as a reliable decision-support tool for clinicians, offering valuable insights into infants' sucking behaviors. By integrating machine learning (ML)-based computational modeling into clinical practice, we aim to enhance the early identification of unhealthy sucking patterns, allowing for timely interventions and pro-active healthcare management.

Machine Learning-Based Nanozyme Sensor Array as an Electronic Tongue for the Discrimination of Endogenous Phenolic Compounds in Food.

The detection of endogenous phenolic compounds (EPs) in food is of great significance in elucidating their bioactivity and health effects. Here, a novel bifunctional vanillic acid-Cu (VA-Cu) nanozyme with peroxidase-like and laccase-like activities was successfully prepared. The peroxidase mimic behavior of VA-Cu nanozyme can catalyze 3,3',5,5'-tetramethylbenzidine (TMB) to generate oxidized TMB (oxTMB). Owing to the high reducing power of EPs, this process can be inhibited, and the degree of inhibition increases with the increase of reaction time. Additionally, owing to the outstanding laccase mimic behavior of the VA-Cu, it can facilitate the oxidation of various EPs, resulting in the formation of colored quinone imines, and the degree of catalysis increases with the increase of reaction time. Based on the interesting experimental phenomena mentioned above, a six-channel nanozyme sensor array (2 enzyme-mimic activities × 3 time points = 6 sensing channels) was constructed, successfully achieving discriminant analysis of nine EPs. In addition, the combination of artificial neural network (ANN) algorithms and sensor arrays has successfully achieved accurate identification and prediction of nine EPs in black tea, honey, and grape juice. Finally, a portable method for identifying EPs in food has been proposed by combining it with a smartphone.

Biomechanics on Ultra-Sensitivity of Venus Flytrap's Micronewton Trigger Hairs.

Numerous plants evolve ingeniously microcantilever-based hairs to ultra-sensitively detect out-of-plane quasi-static tactile loads, providing a natural blueprint for upgrading the industrial static mode microcantilever sensors, but how do the biological sensory hairs work mechanically? Here, the action potential-producing trigger hairs of carnivorous Venus flytraps (Dionaea muscipula) are investigated in detail from biomechanical perspective. Under tiny mechanical stimulation, the deformable trigger hair, composed of distal stiff lever and proximal flexible podium, will lead to rapid trap closure and prey capture. The multiple features determining the sensitivity such as conical morphology, multi-scale functional structures, kidney-shaped sensory cells, and combined deformation under tiny mechanical stimulation are comprehensively researched. Based on materials mechanics, finite element simulation, and bio-inspired original artificial sensors, it is verified that the omnidirectional ultra-sensitivity of trigger hair is attributed to the stiff-flexible coupling of material, the double stress concentration, the circular distribution of sensory cells, and the positive local buckling. Also, the balance strategy of slender hair between sensitivity and structural stability (i.e., avoiding disastrous collapse) is detailed revealed. The unique basic biomechanical mechanism underlying trigger hairs is essential for significantly enhancing the performance of the traditional industrial static mode microcantilever sensors, and ensure the stability of arbitrary load perception.

An artificial intelligence handheld sensor for direct reading of nickel ion and ethylenediaminetetraacetic acid in food samples using ratiometric fluorescence cellulose paper microfluidic chip

User-friendly in-field sensing protocol is crucial for the effective tracing of intended analytes under less-developed countries or resources-limited environments. Nevertheless, existing sensing strategies require professional technicians and expensive laboratory-based instrumentations, which are not capable for point-of-care on-site analyses. To address this issue, artificial intelligence handheld sensor has been designed for direct reading of Ni2+ and EDTA in food samples. The sensing platform incorporates smartphone with machine learning-driven application, 3D-printed handheld device, and cellulose paper microfluidic chip stained with ratiometric red-green-emission carbon dots (CDs). Intriguingly, Ni2+ introduction makes green fluorescent (FL) of CDs glow but red FL fade because of the coordination of Ni2+ with CDs verified by density functional theory (DFT), concurrently manifesting continuous FL colour transition from red to green. Subsequent addition of EDTA renders FL of CDs-Ni2+ recover owing to the capture of Ni2+ from CDs by EDTA based on strong chelation effect of EDTA on Ni2+ confirmed via DFT, accompanying with a noticeable colour returning from green to red. Inspired by above FL phenomena, CDs-based cellulose paper microfluidic chips are first fabricated to facilitate point-of-care testing of Ni2+ and EDTA. Designed fully-automatic handheld sensor is utilized to directly output Ni2+ and EDTA concentration in water, milk, spinach, bread, and shampoo based on wide linear ranges of 0–48 μM and 0–96 μM, and low limits of detection of 0.274 μM and 0.624 μM, respectively. The proposed protocol allows for speedy straightforward on-site determination of target analytes, which will trigger the development of automated and intelligent sensors in near future.

Application of artificial intelligence sensor and visual image technology in the analysis of hydrophilic space landscape characteristics

With the continuous development of intelligent sensor technology and artificial intelligence, the application of artificial intelligence feature recognition at the level of intelligent sensor information acquisition to urban landscape analysis has become a current research trend and hot spot, but also brings us new research directions and challenges. In this paper, the multi-sensor collaborative filtering algorithm is optimized, and a method based on evidence theory is proposed to deal with the fuzzy and unclear information generated in the process of multi-sensor collaboration. Then, the analysis and recognition of multi-sensor fusion information using AI technology are analyzed in detail. This paper discusses the problems in multi-sensor information fusion and related solutions. Combined with the key nodes of 3D object AI feature recognition, multi-sensor collaborative Dempster Shafer evidence theory and 3D convolutional neural network waterfront space landscape feature recognition sub-model are constructed, and the waterfront space landscape recognition analysis model is tested and analyzed. The results show that the multi-sensor information collection Kalman filtering fusion algorithm effectively realizes the recognition of landscape feature information and the filtering of irrelevant interference information combined with an artificial intelligence algorithm to build a landscape feature recognition model. Among all kinds of landscape recognition, the recognition effect of beach inflow is the best, and the recognition accuracy, recall, and F1 value are all above 95. However, the identification effect of hydrophilic plank roads and underwater breakwater is relatively poor. The recognition accuracy of the hydrophilic platform is the lowest at 83.16 % among the eight landscape types, and the recall rate of underwater breakwater is the lowest at 79.96 % among the eight landscape types. In general, the multi-sensor Kalman filtering algorithm information fusion model designed in this paper has higher recognition accuracy and better classification performance for the collection and classification of the entire landscape dataset. The work of this paper provides a new idea and direction for the design and application of waterfront landscape identification and analysis systems based on intelligent sensors and artificial intelligence technology and lays a certain foundation for the intelligent design of urban waterfront landscapes.

Smart Nursing Wheelchairs: A New Trend in Assisted Care and the Future of Multifunctional Integration.

As a significant technological innovation in the fields of medicine and geriatric care, smart care wheelchairs offer a novel approach to providing high-quality care services and improving the quality of care. The aim of this review article is to examine the development, applications and prospects of smart nursing wheelchairs, with particular emphasis on their assistive nursing functions, multiple-sensor fusion technology, and human-machine interaction interfaces. First, we describe the assistive functions of nursing wheelchairs, including position changing, transferring, bathing, and toileting, which significantly reduce the workload of nursing staff and improve the quality of care. Second, we summarized the existing multiple-sensor fusion technology for smart nursing wheelchairs, including LiDAR, RGB-D, ultrasonic sensors, etc. These technologies give wheelchairs autonomy and safety, better meeting patients' needs. We also discussed the human-machine interaction interfaces of intelligent care wheelchairs, such as voice recognition, touch screens, and remote controls. These interfaces allow users to operate and control the wheelchair more easily, improving usability and maneuverability. Finally, we emphasized the importance of multifunctional-integrated care wheelchairs that integrate assistive care, navigation, and human-machine interaction functions into a comprehensive care solution for users. We are looking forward to the future and assume that smart nursing wheelchairs will play an increasingly important role in medicine and geriatric care. By integrating advanced technologies such as enhanced artificial intelligence, intelligent sensors, and remote monitoring, we expect to further improve patients' quality of care and quality of life.

Resilience Informatics: Role of Informatics in Enabling and Promoting Public Health Resilience to Pandemics, Climate Change, and Other Stressors.

Climate change, local epidemics, future pandemics, and forced displacements pose significant public health threats worldwide. To cope successfully, people and communities are faced with the challenging task of developing resilience to these stressors. Our viewpoint is that the powerful capabilities of modern informatics technologies including artificial intelligence, biomedical and environmental sensors, augmented or virtual reality, data science, and other digital hardware or software, have great potential to promote, sustain, and support resilience in people and communities. However, there is no "one size fits all" solution for resilience. Solutions must match the specific effects of the stressor, cultural dimensions, social determinants of health, technology infrastructure, and many other factors.

Embedded Hybrid-Dimensional Heterointerface for Filament Modulation in 2D Material-Based Artificial Nociceptor.

Nociceptors are key sensory receptors that transmit warning signals to the central nervous system in response to painful stimuli. This fundamental process is emulated in an electronic device by developing a novel artificial nociceptor with an ultrathin, nonstoichiometric gallium oxide (GaOx)-silicon oxide heterostructure. A large-area 2D-GaOx film is printed on a substrate through liquid metal printing to facilitate the production of conductive filaments. This nociceptive structure exhibits a unique short-term temporal response following stimulation, enabling a facile demonstration of threshold-switching physics. The developed heterointerface 2D-GaOx film enables the fabrication of fast-switching, low-energy, and compliance-free 2D-GaOx nociceptors, as confirmed through experiments. The accumulation and extrusion of Ag in the oxide matrix are significant for inducing plastic changes in artificial biological sensors. High-resolution transmission electron microscopy and electron energy loss spectroscopy demonstrate that Ag clusters in the material dispersed under electrical bias and regrouped spontaneously when the bias is removed owing to interfacial energy minimization. Moreover, 2D nociceptors are stable; thus, heterointerface engineering can enable effective control of charge transfer in 2D heterostructural devices. Furthermore, the diffusive 2D-GaOx device and its Ag dynamics enable the direct emulation of biological nociceptors, marking an advancement in the hardware implementation of artificial human sensory systems.

Robotic technology in Autism intervention: Design and application of the CARS cat-like robot

This study introduces a novel robotic system designed to aid individuals with Autism Spectrum Disorder (ASD) in enhancing their social communication and interaction skills. Incorporating advanced artificial intelligence, sensors, and image recognition technologies, the robot simulates human social interactions and provides tailored feedback, aiming to improve the quality of life for individuals with ASD. Through a detailed examination of hardware design, software functionalities, and data processing mechanisms, this paper evaluates the system's practical applications and identifies ongoing challenges and potential improvements. The findings suggest that while the robotic system significantly enhances social interaction and learning for autistic individuals, challenges such as technological adaptability, cost, and physical design require further innovation. This research contributes to the field of assistive technologies by offering insights into how robotic systems can be optimized to meet the unique needs of the autism community, thus paving the way for more effective therapeutic tools.

Asymmetric flexible graphene oxide papers for moisture-driven actuators and water level indicators

Smart actuators have shown great potential for a range of industries, including artificial muscles, sensors, and smart devices. However, traditional actuators often suffer from small deformation and complex materials and manufacturing processes. Graphene oxide (GO) has emerged as an ideal candidate for moisture-responsive actuators due to its excellent moisture sensitivity. In this study, we present the development of a moisture-responsive actuator using a simple and straightforward approach based on a monolayer graphene oxide (GO) paper. By utilizing the different affinity of its two sides for water, the graphene oxide actuator demonstrates excellent performance in response to humidity gradients, achieving a large bending amplitude with a curvature of 3 cm−1. Furthermore, the graphene oxide actuator was used as a humidity-sensitive switch to perform on and off switch movements, demonstrating its capability for autonomous detection for real-time humidity monitoring. This work provides a new avenue for the development of moisture-responsive materials and their potential applications in the design of intelligent control systems, actuators, soft robots, artificial muscles, and switches.

Optimization of GMAW Processes in Robotic Welding: A Case Study from Automotive Manufacturing

The optimization of Gas Metal Arc Welding (GMAW) processes in robotic welding has become crucial for enhancing efficiency and quality in automotive manufacturing. This paper explores the integration of advanced robotic welding technologies, focusing on the use of the KUKA KR8 L8 Arc HW robot in GMAW applications. By leveraging artificial intelligence and sensor integration, this study demonstrates significant improvements in welding precision and process stability. Key advancements include the use of geometric and process sensors for real-time adaptive control, and the implementation of Fronius CMT for superior arc control and penetration. Detailed case studies from automotive manufacturing illustrate the practical benefits of these technologies, showcasing enhanced productivity, reduced cycle times, and improved weld quality. The findings underscore the importance of continuous innovation and optimization in robotic welding processes to meet the evolving demands of the automotive industry.

Artificial olfactory sensor based on single metal atom oxide for intelligent nanoelectronics

Gas sensors for flammable and explosive H2 and CO based on artificial olfactory receptor to operate at room temperature (RT) is an imperious pre-requisite for realizing artificial intelligent (AI) olfaction. To tailor the H2 and CO sensing abilities, herein, artificial olfactory receptors based on single metal (Fe, Ni) atom oxide (SMAO) anchored on DNA coordinated graphene is introduced. In this design, DNA strands facilitate homogeneous decoration of single atoms bio-mimicking the natural olfactory receptor, meanwhile prohibit the stacking of graphene layers. The designed olfactory sensors demonstrate outstanding sensing performances (complete recovery within 50 s to 200–1000 ppm H2 and CO) at RT with durability for at least half a year. Furthermore, H2 and CO were proactively discriminated and recognized via principal component analysis with the designed DG-Fe and DG-Ni devices. Innovatively, in-situ XPS analysis and the corresponding DFT simulation suggest that the great H2 and CO sensing properties emerges from the oxidized surface of SMAO and the oxygen vacancies formed in SMAO during H2/CO adsorption. The proposed sensing mechanism provides insights for the development of robust RT gas sensors. The novel biomimetic design of DNA-graphene anchoring SMAO is very promising for intelligent nanoelectronics in the AI olfaction field.

Self-Powered and Self-Recovered Mechanoluminescent Electronic Skins for Detecting and Distinguishing Diverse External Stimuli.

The concept of simulating external mechanical stimuli to generate luminescence has been a long-standing aspiration in real-time dynamic visualization. However, creating self-power and self-restoring mechanoluminescent electronic skins for artificial sensors poses significant challenges. In this study, we introduce a cutting-edge triboelectric-mechanoluminescent electronic skin (TMES) that exhibits a remarkable response to multiple external stimuli. This advancement is achieved by integrating a mechanoluminescent intermediate layer within a triboelectric nanogenerator (TENG). When pressure is applied to TMES, the maximum detection voltage can reach hundreds of volts and the maximum correlation sensitivity is 11.76 V/N. Moreover, we incorporate luminescence materials into mechanoluminescence layer, and the maximum absolute sensitivity SR can reach 1.41%. The device can not only distinguish between external stimuli such as pressing and bending but also continuously track external mechanical stimuli. A 4 × 4 matrix and motion prediction of 8 different postures were established to further demonstrate the significant advantages of the developed device in spatial detection. The versatility and performance of the TMES hint at its vast potential in areas such as human-computer interaction and wearable electronics, paving the way for more intuitive and dynamic technological interfaces.

Frequency-Selective, Multi-Channel, Self-Powered Artificial Basilar Membrane Sensor with a Spiral Shape and 24 Critical Bands Inspired by the Human Cochlea.

A spiral-artificial basilar membrane (S-ABM) sensor is reported that mimics the basilar membrane (BM) of the human cochlea and can detect sound by separating it into 24 sensing channels based on the frequency band. For this, an analytical function is proposed to design the width of the BM so that the frequency bands are linearly located along the length of the BM. To fabricate the S-ABM sensor, a spiral-shaped polyimide film is used as a vibrating membrane, with maximum displacement at locations corresponding to specific frequency bands of sound, and attach piezoelectric sensor modules made of poly(vinylidene fluoride-trifluoroethylene) film on top of the polyimide film to measure the vibration amplitude at each channel location. As the result, the S-ABM sensor implements a characteristic frequency band of 96-12,821Hz and 24-independent critical bands. Using real-time signals from discriminate channels, it is demonstrated that the sensor can rapidly identify the operational noises from equipment processes as well as vehicle sounds from environmental noises on the road. The sensor can be used in a variety of applications, including speech recognition, dangerous situation recognition, hearing aids, and cochlear implants, and more.

An artificial intelligence-assisted microfluidic colorimetric wearable sensor system for monitoring of key tear biomarkers

The precise, simultaneous, and rapid detection of essential biomarkers in human tears is imperative for monitoring both ocular and systemic health. The utilization of a wearable colorimetric biochemical sensor exhibits potential in achieving swift and concurrent detection of pivotal biomarkers in tears. Nevertheless, challenges arise in the collection, interpretation, and sharing of data from the colorimetric sensor, thereby restricting the practical implementation of this technology. To overcome these challenges, this research introduces an artificial intelligence-assisted wearable microfluidic colorimetric sensor system (AI-WMCS) for rapid, non-invasive, and simultaneous detection of key biomarkers in human tears, including vitamin C, H+(pH), Ca2+, and proteins. The sensor consists of a flexible microfluidic epidermal patch that collects tears and facilitates the colorimetric reaction, and a deep-learning neural network-based cloud server data analysis system (CSDAS) embedded in a smartphone enabling color data acquisition, interpretation, auto-correction, and display. To enhance accuracy, a well-trained multichannel convolutional recurrent neural network (CNN-GRU) corrects errors in the interpreted concentration data caused by varying pH and color temperature in different measurements. The test set determination coefficients (R2) of 1D-CNN-GRU for predicting pH and 3D-CNN-GRU for predicting the other three biomarkers were as high as 0.998 and 0.994, respectively. This correction significantly improves the accuracy of the predicted concentration, enabling accurate, simultaneous, and quick detection of four critical tear biomarkers using only minute amounts of tears ( ~ 20 μL). This research demonstrates the powerful integration of a flexible microfluidic colorimetric biosensor and deep-learning algorithm, which holds immense potential to revolutionize the fields of health monitoring.