Flexible ACEK-Enhanced Capacitive Aptasensor for Rapid Cortisol Detection in Sweat

1. Introduction

Mental health has become a critical determinant of overall well-being in modern society, where increasing life pressures contribute to stress-related disorder[1]. Cortisol, the primary stress hormone regulated by the hypothalamic-pituitary-adrenal (HPA) axis, plays a pivotal role in stress response, metabolic regulation, and circadian rhythm maintenance[2]. Clinical studies indicate that abnormal fluctuations in cortisol levels are closely associated with various psychological disorders (such as depression and anxiety)[3,4]. Consequently, cortisol holds significant potential as a vital biomarker for monitoring psychological stress and managing individual health[5].

Traditional cortisol detection techniques and methods include enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), and liquid chromatography-tandem mass spectrometry (LC-MS/MS)[6,7]. While these methods offer high sensitivity and accuracy, they generally suffer from drawbacks such as complex operation, lengthy detection cycles, high costs, and strong equipment dependency, making them ill-suited for real-time monitoring and point-of-care testing under cortisol's dynamic fluctuations.

In recent years, biosensors have garnered widespread attention due to their advantages of high sensitivity, excellent biocompatibility, strong specificity, and potential for miniaturization. Atul Sharma et al. [8]immobilized antibodies on a vertically layered graphene electrode surface via non-covalent bonds and developed a highly sensitive detection method with a range of 1 fg/mL to 10 ng/mL based on the Fe³⁺/Fe²⁺ redox reaction. Although antibodies are widely used as traditional biomarkers, they still exhibit certain limitations: complex preparation processes, prolonged development cycles, high costs, and susceptibility to cross-reactions during molecular recognition, resulting in insufficient specificity-all of which restrict their application in biosensing. In contrast, aptamers offer significant advantages such as simple synthesis, high stability, and low cost. Ma et al.[9] developed a ratio-based cortisol electrochemical biosensor utilizing a gold electrode modified with multi-walled carbon nanotubes, where aptamers serve as recognition elements combined with methylene blue-labeled zirconium metal-organic frameworks for signal amplification, achieving a detection limit of 0.0046 nM in sweat analysis.

Furthermore, with advancements in microfluidic technology, microfluidic devices have been successfully applied to cortisol detection[10]. Luca et al. [11]reported the first paper-based microfluidic device utilizing filter paper to control fluid flow and load reagents for competitive magnetic bead immunoassay analysis of sweat cortisol, achieving a linear detection range of 10-140 ng/mL. Weng et al. [12]developed a portable 3D microfluidic origami biosensor based on a smartphone platform, which successfully detected cortisol concentrations ranging from 10 to 1000 ng/mL in human sweat within 25 minutes. However, paper-based microfluidic chips fall short in meeting the demands for high-precision quantification, complex sample analysis, and long-term stability, and exhibit prolonged detection times.

To overcome these limitations, rapid detection technologies based on alternating current electrokinetics have emerged as an effective strategy for accelerating molecular enrichment at electrode surfaces, thereby significantly enhancing sensor response speed and sensitivity. Integrating the ACEK-driven enrichment mechanism with the high specificity of aptamer-based recognition offers a promising approach for the development of electrochemical sensing platforms for cortisol that are rapid, require minimal sample volumes, and exhibit high sensitivity. Based on this, this study proposes a rapid cortisol detection method based on Alternating current electrokinetics, develops an aptamer electrochemical sensor based on ACEK and microfluidic technology utilizing a flexible fork-finger electrode structure, as shown in Figure 1(a), and establishes a rapid cortisol detection system integrating microfluidics, ACEK enrichment, and aptasensor recognition, as illustrated in Figure 1(b). By optimizing ACEK parameters (AC excitation voltage, frequency) and aptamer modification concentration, efficient enrichment and highly specific recognition of cortisol were achieved. The sensing interface was characterized by scanning electron microscopy (SEM) and X-ray photoelectron spectroscopy (XPS). Experimental results demonstrated that the sensor exhibited excellent linearity within the concentration range of 1-1000 ng/mL, with a detection limit of 0.337 ng/mL in artificial sweat and a short response time and only 10 μL of sample required for detection. This study provides technical support for rapid cortisol detection and offers novel design insights for the development of non-invasive stress monitoring sensors.

2. Theory of sensing and enrichment mechanisms

The ACEK-based capacitive biosensor detects target molecules by monitoring variations in the electrode–electrolyte interfacial capacitance. When cortisol molecules specifically bind to the surface-immobilized aptamers, the structure of the electrical double layer (EDL) is altered, leading to measurable capacitance changes that serve as the basis of detection.

The interfacial capacitance of an electrode immersed in an electrolyte can be expressed as:

$$C={\displaystyle \frac{\epsilon A}{d}}$$

(1)

where $\epsilon$ is the dielectric constant, A is the effective electrode area, and d is the effective thickness of the dielectric layer[13].

With surface modification and subsequent target binding, the dielectric layers increase, resulting in a decrease in total capacitance. To reduce system errors and improve comparability, the relative capacitance change is commonly used as the detection signal:

$${\displaystyle \frac{∆C}{C_0}}={\displaystyle \frac{C-C_0}{C_0}}$$

(2)

Herein, C₀ is the initial capacitance, and C is the capacitance after target molecule binding. The schematic diagram of the interfacial capacitance change during electrode modification and target binding is shown on the right side of Figure 2, which clearly presents the evolution law of the electrical double layer structure during the modification and binding processes.

However, relying solely on free molecular diffusion results in low binding efficiency. The ACEK effect induced by a non-uniform alternating electric field can drive fluid flow and molecular migration, thereby achieving rapid enrichment of target molecules[14]. Among various mechanisms, dielectrophoresis (DEP) has a weak effect on small molecules[15], alternating current electroosmosis (ACEO) is more prominent under low conductivity conditions, while alternating current electrothermal (ACET) can still function effectively in high-conductivity electrolytes such as phosphate-buffered saline (PBS), thus becoming the main driving force in this study[16]. The left side of Figure 2 intuitively shows the ACEK-assisted interfacial capacitance change mechanism: under the action of an alternating electric field, cortisol molecules are rapidly enriched on the electrode surface, accelerating the specific binding with aptamers, thereby amplifying the capacitance signal change and improving detection efficiency and sensitivity.

According to Equation (2), the binding level of cortisol can be quantitatively characterized by detecting the relative change $∆C/C_0$ in interfacial capacitance. However, due to variations in the bottom-up functionalization process on electrode surfaces and fluctuations in sample electrical properties, inconsistent responses may occur among sensors. To address this, this study employs the rate of interfacial capacitance change $dC/dt$ as a quantitative indicator-a method that not only enhances detection accuracy but also effectively reduces measurement uncertainty. The impedance analyzer was configured in time-scanning mode with a continuous measurement duration of t min and a scanning interval of 1s. Under these parameters, the interfacial capacitance change rate $dC/dt$ at t min was obtained as follows:

$$dC/dt=\left[\left(C_{60t}-C_0\right)/C_0\right]$$

(3)

In the formula,$C_{60t}$ and $C_0$ represent the interface capacitance at time 60t and the initial interface capacitance, respectively.

3. Materials and Methods

3.1. Reagents and Equipment

The main reagents used in this study include: 1×PBS buffer (Solarbio, analytical grade), ethanol(Aladdin, analytical grade), isopropanol(McLean, with a purity of ≥99.5%)), acetone(Chengdu Kelon, purity: AR), nitrogen gas(Chongqing Shengma Gas, with a purity of 99.99%), TCEP solution(Source Leaf, with a purity of 99% ), 5% bovine serum albumin (BSA) blocking solution(Solubon, with a purity of 99% ), and 5′-thiol-modified cortisol aptamers (HPLC purified, Sangon Biotech, Shanghai). The experimental instruments used in this study included an ultrasonic cleaner (KQ-300DE), a UV-ozone cleaner (SDP-UVT), a vortex mixer (VORTEX 2), an analytical balance (ML204T/02), a temperature-humidity chamber (FYL-YS-100L), micropipettes (Discovery-H series), and an impedance analyzer (TH2839, Tonghui).

The aptamer was synthesized with a -SH group modified at the 5' end, and its sequence is: SH-5'-ATG GGC AAT GCG GGG TGG AGA ATG GTT GCC GCA CTT CGG C-3', which was referenced from previously published literature[17,18]. Cortisol powder was first dissolved in ethanol to prepare a 1 mg/mL stock solution, then diluted with PBS to different concentrations for standard solutions. Structural analogues including β-estradiol, progesterone, and corticosterone were prepared following the same protocol. All stock solutions were aliquoted and stored at -20 °C until use.

3.2. Construction of Aptamer Sensor

Flexible interdigitated gold electrodes (Polyethylene terephthalate (PET) substrate, 10mm×10mm, line width/spacing =100μm, 10pairs of fingers) were used as the sensing platform. Electrodes were cleaned sequentially with acetone, isopropanol, and deionized water under ultrasonication, followed by UV-ozone treatment to improve surface cleanliness and hydrophilicity. A silicone chamber (2 mm diameter, 0.9 mm depth) was attached to define the reaction area.

Mix TCEP with the thiol-modified aptamer in a molar ratio of 50:1 and react at room temperature for 1 hour to restore the -SH groups, followed by immobilization on the electrode surface via Au-S bonding after an incubation period of 24 hours. Unbound probes were removed by rinsing with ultrapure water. To minimize nonspecific adsorption, the electrode was first sealed with a 5% bovine serum albumin solution for 1 hour, then washed. The resulting functionalized electrode was prepared for use as a cortisol aptasensor.

During measurement, electrodes were exposed to cortisol solutions of varying concentrations, while PBS served as the negative control. The impedance analyzer was operated in time-scan mode (1 min, 1 s intervals) to record interface capacitance changes. The normalized capacitance variation ΔC/C₀ was calculated as the quantitative signal, directly reflecting target binding events. All measurements were repeated at least five times to ensure reliability. The Figure 3 provides a schematic diagram illustrating the entire process of the sensor, from preparation to testing.

3.3. Simulation and Parameter Optimization

To enhance the detection performance of aptamer sensors based on the ACEK effect, systematic optimizations were conducted on the alternating current signal voltage, frequency, and aptamer probe concentration. A three-dimensional model of the interdigital electrode was established using COMSOL Multiphysics to systematically investigate the electric field distribution under varying excitation conditions. In this model, PET was selected as the substrate material due to its favorable mechanical flexibility and chemical stability, while gold was employed as the electrode material to ensure excellent electrical conductivity and biocompatibility. To realistically simulate the sensing environment within the detection chamber, the conductivity of the solution was set to 1.6 S/m and the relative permittivity was defined as 78, corresponding to typical aqueous conditions.

At a fixed frequency of 10 kHz, sinusoidal AC voltages with amplitudes of 1 mV, 10 mV, 100 mV, 400 mV, 700 mV, and 1000 mV were applied. The electric field modulus was evaluated at one-quarter of the signal cycle to capture the representative field distribution. As shown in Table 1, the simulation results demonstrated a clear positive correlation between the applied voltage and the electric field intensity. However, the rate of increase became markedly more pronounced when the voltage exceeded 400 mV. While higher voltages can enhance the electric field strength, they may also induce excessive electrokinetic flow within the microfluidic channel, potentially disrupting the interaction between target molecules and surface-immobilized probes. Therefore, 400 mV was determined to be the optimal voltage, balancing field enhancement and reaction stability.

Subsequently, under the fixed voltage of 400 mV, frequency-dependent simulations were performed at 1 kHz, 5 kHz, 10 kHz, 15 kHz, 20 kHz, 25 kHz, and 35 kHz. As shown in the Table 2, the results indicated that the electric field modulus reached its maximum at 15 kHz, with the value at 10 kHz being comparable. Considering that lower frequencies generally provide improved resistance to external interference and reduce the influence of interfacial capacitance, 10 kHz was ultimately selected as the optimal operating frequency for the system.

The probe concentration was optimized using a gradient concentration approach. The aptamer probe concentrations selected were 10 μg/mL, 30 μg/mL, 60 μg/mL, 90 μg/mL, and 120 μg/mL. Sensors prepared with each probe concentration were used to detect cortisol solutions at concentrations of 1 ng/mL, 10 ng/mL, 100 ng/mL, and 1000 ng/mL, respectively. The experiment used 1×PBS buffer as the background reference. Each concentration combination was measured in parallel five times, and the interface capacitance change rate over 1 minute was recorded. The test results under each concentration condition are summarized in the Table S1. To visually assess the impact of probe concentration on detection performance, the Figure 4 shows error bars and line plots for the detection performance of probes at different concentrations across various cortisol concentration gradients.

The results showed that the response was insufficient at low concentrations of 10 μg/mL and 30 μg/mL; the response decreased at a high concentration of 120 μg/mL; both 60 μg/mL and 90 μg/mL effectively detected all concentrations, with 90 μg/mL exhibiting the maximum signal response at the high concentration (1000 ng/mL). After comprehensive consideration of sensitivity and binding efficiency, 90 μg/mL was determined as the optimal probe concentration.

Through systematic optimization of experimental parameters, the optimal operating conditions for the sensor were determined as follows: aptamer probe concentration of 90 μg/mL, and AC electrical signal parameters generating the ACEK effect of 400 mV at 10 kHz.

3.4. Design and Fabrication of Microfluidic Chip

To meet the demand for cortisol detection in sweat samples, a three-layer microfluidic chip was designed and fabricated, as shown in Figure 5(a). The layers from top to bottom are: the microchannel layer, the collection/detection layer, and the electrode layer, as illustrated in the schematic diagram. The microchannel layer features an injection port with a diameter of 1 mm and an injection channel measuring 20 mm in length, 300 μm in width, and 170 μm in depth. The collection/detection layer is primarily responsible for solution collection and detection; its detection zone adopts a cylindrical structure with a diameter of 2 mm and a depth of 1 mm, whose edge is tangent to the tip of the underlying finger-like electrode. The sample outlet channel measures 200 μm in width, 170 μm in depth, and 6.5 mm in length. The electrode layer houses a gold-based finger-like electrode cortisol aptamer sensor fabricated on a PET substrate (10.2 mm in length and 6 mm in width), with the groove depth optimized to 170 μm. Detailed dimensions of the chip are shown in the Figure 5(b).

As shown in the Figure 6(a), during the mold and chip fabrication process, SU-8 2100 photoresist is first selected to create a microchannel positive mold on the silicon wafer using conventional soft lithography technology. Following steps including resin homogenization, pre-baking, contact exposure, post-baking, development, and 5 minutes of 150°C film curing, the developed mold surface is treated with trimethylchlorosilane(TMCS) for vapor-phase deposition to prevent adhesion. Subsequently, a high-mechanical-strength epoxy resin mold is fabricated via the reverse molding process to extend its service life. The successfully fabricated silicon wafer mold is illustrated in Figure 6(b) and has been stored at 4-21 °C under light protection for subsequent use in manufacturing microfluidic chips.

Subsequently, using the prepared SU-8 silicon wafer mold described above, dimethylsiloxane was employed to fabricate the microfluidic chip, as illustrated in the Figure 7(a). The polydimethylsiloxane (PDMS) prepolymer and curing agent were thoroughly mixed in a mass ratio of 10:1, degassed under vacuum, poured onto the resin mold surface, covered with a PET film and flattened, then cured in a 75°C oven for 1.5 hours. After removal and drilling, the microchannel layer and collection/detection layer were treated under oxygen plasma for 40 seconds; the introduction of hydroxyl groups on the surface formed stable Si-O-Si covalent bonds, achieving irreversible bonding between the two layers. A schematic diagram of the bonding process is shown in the Figure 7(b).

Finally, due to the raised structure of the electrode's gold layer and the PET substrate material, directly bonding the collection/detection layer to the electrode layer requires stringent process specifications and is prone to bonding failure, which compromises chip sealing integrity and may lead to liquid leakage. Therefore, after positioning the aptamer sensor on the electrode layer, a detection aperture is created using Optically clear adhesive (OCA) before bonding the collection/detection layer to the electrode layer. OCA offers advantages such as high transparency, uniform thin-layer adhesion, ease of application, and excellent biocompatibility, making it the ideal choice for bonding these layers. To facilitate detection procedures, the electrode layer is also bonded to the glass slide.

The Figure 8 shows both a schematic diagram and an actual image of the microfluidic sensing chip. During sample analysis, the system employs a syringe to provide fluid drive force and delivers samples via a medical-grade hose. To ensure connection reliability, a steel needle connects the syringe to the hose/chip interface, effectively preventing liquid leakage while maintaining the system's sealing performance. Experimental results confirm that the microfluidic chip features a rational structural design, reliable sealing performance, and precise fluid control; detailed information is provided in the Figure S3-Figure S6.

4. Results and Discussion

4.1. Sensor Interface Characterization

To verify the functionalization process of the sensor surface, this study systematically characterized the electrodes at different modification stages using scanning electron microscopy and X-ray photoelectron spectroscopy. To evaluate the cleaning efficacy of electrode pretreatment, scanning imaging analysis was first performed on the cleaned bare electrodes. Under an acceleration voltage of 20 kV, the electrode surface morphology was observed at magnifications of 150× (Figure 9(a)) and 2000× (Figure 9(b)). Characterization results demonstrated that the pretreated electrode surfaces were smooth and free from contamination, indicating effective cleaning and providing an optimal environment for sensor fabrication. Under the testing conditions of 20 kV acceleration voltage and 2000× magnification, comparative analyses were conducted between probe-modified electrodes and blocked electrodes. Figure 9(c) shows granular deposits on the electrode surfaces after aptamer probe incubation, confirming successful fixation of the probes via Au-S bonds over the 24-hour incubation period. Figure 9(d) reveals additional granular deposits of varying sizes on blocked electrodes, verifying the effective deposition of the BSA blocking agent and confirming successful completion of the site-blocking step.

XPS analysis results further confirmed the modification effect: compared with the bare gold electrode (Figure 9(e)), the aptamer-modified electrode(Figure 9(f)) surface showed a significant increase in carbon content, the appearance of a characteristic nitrogen signal, and a distinct decrease in the gold signal intensity. This change is highly consistent with the compositional characteristics of aptamer molecules-the pentose sugar structure in nucleotides contributes a large amount of carbon, the nitrogenous bases introduce characteristic nitrogen signals, and the weakened gold signal indicates that the electrode surface has been effectively covered by the aptamer layer. Combined with the elemental analysis results of energy dispersive spectroscopy (EDS), the successful realization of the aptamer modification process on the sensor surface is fully confirmed.

Furthermore, for capacitive sensors, interfacial capacitance serves as an effective characterization parameter. This method was employed to investigate variations in the initial surface capacitance of electrodes during different modification stages. As the EDL thickness and dielectric constant at the electrode-solution interface change during surface modification, corresponding shifts in interfacial capacitance occur. Experiments conducted using the same electrode for consecutive modification and measurement demonstrated characteristic changes in the initial capacitance values after each step, confirming successful modification of both the aptamer probe and the blocking solution on the electrode surface.

4.2. Capacitance Response Characteristics and Concentration Linearity Analysis

To comprehensively evaluate sensor performance, cortisol solutions at concentrations of 1 ng/mL, 10 ng/mL, 100 ng/mL, 200 ng/mL, 400 ng/mL, 600 ng/mL, 800 ng/mL, and 1000 ng/mL were selected for testing, covering both normal physiological fluctuations and abnormal levels in sweat. 1×PBS was used as the background solution, with each concentration measured in quintuplicate and a sample volume of 10 μL.

Based on real-time detection data collected within 1 minute, Figure 10(a) displays the dynamic curves of interfacial capacitance versus time at different concentrations. According to the data in Table S2, using the average dC/dt values obtained from parallel measurements of each concentration sample as the ordinate and the cortisol concentration as the abscissa, weighted linear regression analysis was performed using Origin 2025 software. The weight ${\omega }_i=1/S{D}_i^2$ was taken as the reciprocal of the square of the standard deviation of the response at each concentration point. Weighted linear fitting of the data within the range of 1–1000 ng/mL yielded the standard curve equation for the system in the PBS system as follows: $y=-0.0028x-1.2086$, where x represents the cortisol concentration (ng/mL), and dC/dt represents the interfacial capacitance change rate (%/min). The coefficient of determination (R²) was 0.9776, and the sum of squared errors (SSE) was 0.1494. The detection limit of the sensor was 0.412 ng/mL. Figure 10(b) shows the linear fitting results between the interfacial capacitance change rate and the sample concentration. The results demonstrate that, within the investigated concentration range, a strong linear correlation exists between cortisol concentrations and system response values. This indicates that the developed cortisol detection system not only exhibits rapid response to cortisol in PBS solutions but also demonstrates robust quantitative detection capabilities, providing a valuable reference for subsequent simulations of real-sample testing and system performance validation.

4.3. Specificity and Repeatability

This experiment selected progesterone, β-estradiol, and corticosterone as interferents for comparative analysis. The cortisol concentration was set at 100 ng/mL, while interferent concentrations were uniformly 1000 ng/mL. Figure 11(a) presents bar charts and error bars showing the specific detection results for the four substances. At the high concentration of 1000 ng/mL, the response values for β-estradiol, progesterone, and corticosterone were all below 1%, comparable to the background signal and significantly lower than the response to 100 ng/mL cortisol. Among these, corticosterone exhibited a slightly higher response than the other two interferents due to a structural difference involving only one hydroxyl group compared to cortisol. However, the weak cross-reactivity did not compromise the sensor's specificity, demonstrating that the aptamer sensor possesses excellent specific recognition capability for cortisol.

To evaluate sensor repeatability and reproducibility, five independent sensors were prepared for each of the 1, 10, 100, and 1000 ng/mL concentration points and tested in parallel. The fabrication process and data processing of all sensors adhered to unified standards to ensure consistency in experimental conditions. Table S3 summarizes the response rates of interfacial capacitance changes for each sensor across four orders of magnitude of cortisol concentrations.

The relative standard deviation (RSD%) of the sensors across all four concentration gradients was below 5.5%, indicating low dispersion in the test results. This finding confirms excellent consistency among the different sensors under identical detection conditions, thereby enhancing the reliability of the experimental data. To visually demonstrate the reproducibility characteristics, Figure 11(b) presents the detection results from five sensors at each concentration level.

4.4. Preliminary System Integration Testing and Sweat Sample Response

To more realistically simulate physiological conditions, this study employed artificial sweat as the background medium and prepared cortisol samples at varying concentration gradients for testing. A microfluidic chip detection system based on the ACEK effect was utilized to achieve sealed transport and uniform contact of sweat samples, thereby enhancing detection stability and sensitivity. The rapid cortisol detection system based on the ACEK effect is illustrated in Figure 12.

The system comprises three core components: a cortisol aptamer sensor utilizing a microfluidic chip, an AC electrical signal detection module, and data analysis software, designed for evaluating the response characteristics of cortisol samples in human sweat. For data analysis, two key functions of the software-surface capacitance mapping and interface capacitance change rate calculation-are employed to provide foundational data support for developing new mathematical models.

Figure 13(a) illustrates the dynamic response curves of cortisol samples at different concentrations. Analysis revealed that, except for brief signal overlap at 100 ng/mL and 200 ng/mL during the initial 10 seconds, all concentration gradients maintained distinct signal discrimination characteristics throughout the detection cycle, with particularly enhanced resolution between the low-concentration ranges of 1 ng/mL and 10 ng/mL. Furthermore, compared to the conventional open-system detection method shown in Figure 10(a), the microfluidic chip system demonstrated more stable response characteristics, primarily attributed to its sealed structure that effectively isolates environmental interference factors.

As shown in Figure 13(b), Linear regression yielded the fitting equation $y=-0.0031x-1.5293$，with a coefficient of determination (R²) of 0.9865 and a sum of squared errors (SSE) of 0.1303. The detection limit in artificial sweat was determined to be 0.337 ng/mL. These metrics confirm the system's robust quantitative detection capability even in complex physiological matrices.

5. Conclusions

Cortisol is an important biomarker reflecting human stress levels, and its rapid detection is of significant value for health monitoring. In this study, based on the ACEK effect, a low-cost, label-free sensor was constructed using flexible PET interdigitated gold electrodes functionalized with aptamers, integrated with a microfluidic chip and data analysis software to form a rapid detection system. Through simulation optimization, the excitation voltage and frequency were set to 400 mV and 10 kHz, respectively, with an aptamer concentration of 90 μg/mL. Detection could be completed within one minute, covering a range of 1-1000 ng/mL, with a limit of detection of 0.412 ng/mL. The sensor’s interfacial capacitance response exhibited good linearity with cortisol concentration: $y=-0.0028x-1.2086$, $R^2=0.9776$. In artificial sweat samples, the microfluidic chip enhanced contact between the sample and the electrode, improving signal stability. The system showed clear responses for different cortisol concentrations, with a linear fitting relationship of: $y=-0.0031x-1.5293$, $R^2=0.9865$, and a detection limit of 0.337 ng/mL. These results confirm the accuracy and reliability of the system in complex matrices, indicating its potential for development into a wearable cortisol detection device.