TY - JOUR
AU - Islam, Md. Shabiul
AB - 1. Introduction Cervical cancer is among the three primary malignant posing a significant threat to women’s survival, with its fatality rate ranking third among cancer-related deaths in developing nations [1–3]. Persistent infection with carcinogenic strains of human papillomavirus (HPV) is the leading cause of cervical cancer, resulting in the deaths of over 300,000 women annually, with nearly 90% of these fatalities in resource-limited countries. To address this disparity, there is a pressing need for targeted efforts in streamlined vaccination campaigns and the development of cost-effective, easily accessible screening methods. The World Health Organization advocates for the primary screening method to involve testing for high-risk HPV types, preferably using self-collected cervical-vaginal swab samples for efficient population screening. In situations with ample resources, an additional triage test is employed to refine the risk assessment for cervical precancer or cancer among HPV-positive individuals, constituting a screen-triage-treat strategy to mitigate overtreatment. Even in resource-limited settings, practical HPV screening is recommended, and a similar screen-triage-treat strategy is preferred over universal treatment, especially when balancing test sensitivity with minimal unnecessary interventions [4–6]. Cancer symptoms typically manifest when patients typically find themselves in the middle or advanced stages because precancerous lesions are often symptom-free and difficult to identify. Traditional testing approaches prove effective only when symptoms are evident, leading to increased patient suffering during treatment at these later stages. Various methods, including thinprep-cytologic-testing [7], uterine scraping [8], human papillomavirus screening [9] and vaginoscopy examination [10] can be employed for cervical cancer detection. To achieve a precise diagnosis report, a composite of these methods is commonly utilized, requiring the skills of an experienced medical professional. This indicates that the testing procedure is intricate, inconvenient, expensive, and time-intensive. Thus, the creation of a quick, uncomplicated, accurate, and efficient testing approach is highly important for the early-stage detection and diagnosis of cancers. The terahertz (THz) wave falls within the spectrum between microwaves and infrared region. Due to its low energy, it lacks ionization potential to harm living tissues or the human body. Additionally, as numerous biological macromolecules and polar molecules exhibit vibrational and rotational energy levels within the THz range, it serves as a valuable tool for characterizing biological tissues [11, 12]. Consequently, it has practical applications in medical diagnosis [13–16]. Typically, cancerous tissue exhibits a higher water content compared to normal tissue, and THz waves are readily absorbed by water. The absorption spectrum of THz waves can serve as a reliable method to differentiate between cancerous and normal tissues. Over the past two decades, there has been a growing interest surrounding THz time-domain spectroscopy (THz-TDS) technology. Numerous previous studies have highlighted the potential of THz technology in identifying various types of cancerous tissues, including colon cancer [17], liver cancer [18], blood cancer [19], breast cancer [20], and skin cancer [21–24]. This increased interest can be attributed to the consistent observation that cancerous tissues typically manifest higher absorption coefficients and refractive indices compared to their normal tissues [25–27]. Furthermore, THz imaging technology presents distinct advantages when contrasted with traditional detection methods. Notably, it possesses the ability to precisely sketch the boundaries of cancerous areas. This represents a notable advancement, as traditional techniques often fall short in providing such precise identification. Metamaterials (MTMs), are periodic sub-wavelength structures with a broad range of applications in imaging, antenna engineering and biosensing [28–30], and optoelectronics [31]. These materials have the capability to modify the fundamental principles of optics and electromagnetics [32, 33]. Specifically, metamaterial absorbers (MTMAs) are pivotal in the process of transforming incoming waves into thermal energy [34–36]. This characteristic allows for the detection and analysis of absorbed waves, making MTMAs essential in various applications [37, 38]. The investigation of metamaterials (MTMs) becomes imperative in addressing challenges arising from the lack of conventional electronic and quantum photonic responses in the terahertz (THz) frequency range. The development of a customized and controlled resonant response in metamaterials (MTMs) holds the potential to amplify interactions between terahertz (THz) radiation and materials, leading to the emergence of novel capabilities and outstanding performance [39, 40]. The Metamaterials proposed in this article were utilized as perfect absorbers to design a biosensor with exceptional sensitivity to low refractive index variations. This innovative approach overtakes other absorber types, presenting the unique advantages of metamaterials in bio sensing applications. The proposed metamaterial perfect absorber enables precise detection of subtle changes in refractive index, making it an ideal surface for sensitive labeling. Unlike traditional absorbers, this metamaterial-based biosensor offers enhanced performance, creating the way for more efficient and accurate detection methods in various fields, particularly in bioscience and medical diagnostics where slight variations play a crucial role in accurate analyses. Over the years, metamaterials have found application in the development of various sensors across different fields: Yang et al. presented a THz sensor with tunable characteristics, employing two concentric split-ring resonators. The sensor demonstrated a tuning range spanning from 0.958 to 1.390 THz [41]. Saadeldin et al. introduced a novel MTM (metamaterial) absorber design for THz (terahertz) applications related to sensing in the biomedical field. The sensor demonstrated an impressive 99% absorptivity specifically at 2.249 THz, accompanied by a Q-factor of 22.05. Additionally, it exhibited a sensitivity of 23.7 GHz/μm when employed for measuring analyte thickness [42]. Wang investigated a prototype designed for sensing applications, consisting of two identical square patches made up of unit cells. The suggested configuration achieved close to 100% absorption at frequencies of 1.7780 and 2.4591 THz, accompanied by quality factors of 6.9156 and 296.2771, respectively [43]. Azab et al. offered a thorough examination of optical biosensors designed specifically for the early detection of cancer, incorporating metamaterial sensors within their comprehensive overview [44]. Geng et al. introduced a micro-ring metamaterial resonator that incorporates microfluidics for the detection of biomarkers related to liver cancer. The resonator exhibited a significant maximum frequency shift of 14 GHz when exposed to a concentration of 0.025 μg/ml [45]. In this letter Banerjee et al. introduced a biosensor using two circular ring resonators, demonstrating a high sensitivity of 1500 GHz/RIU and a FOM of 25 RIU-1. This emphasizes its effectiveness in cancerous cell detection with a favorable trade-off between sensitivity and overall performance [46]. Vafapour et al. [47] presented a bio sensor designed for colon cancer detection using a water-based system. Moreover, another sensor utilizing a graphene-based metamaterial demonstrated sensitivity towards various cancer cell types, achieving a sensitivity of 207 GHz/RIU, a FOM of 3.86 RIU-1 and Q of 13 [48]. Moreover, Askari et al. [49] presented an extremely responsive sensor utilizing metamaterial, characterized by its sensitivity of 4077.2 nm/RIU operating in the frequency range between 150 and 500 THz. In [50] Bhati et al. introduced a specialized metamaterial sensor designed for the identification of cancerous cells, displaying a sensitivity of 1462 GHz/RIU. Ma et al. [51] also contributed significantly with their proposal of a refractive index metamaterial sensor characterized by high sensitivity, measuring at 2372 GHz/RIU. However, triple-band sensors both, offering broader coverage, heightened versatility, superior performance, and increased sensitivity in detecting the THz spectrum. This makes triple-band sensors more advanced and effective across various applications compared to their single and dual-band. Recent studies have demonstrated ultra-thin hepta-band metamaterial absorbers with high absorption efficiency across multiple THz frequencies, highlighting the potential of such designs for sensing applications [52]. Additionally, the development of ultra-compact polarization-sensitive triple-band microwave absorbers supports the multi-band absorption characteristics of our proposed biosensor [53]. Graphene-based tunable metamaterial absorbers have shown high absorption efficiency and tunability in the THz range, which aligns with our sensor’s ability to discriminate between healthy and malignant tissues based on dielectric properties [54]. Furthermore, research on ultra-thin metamaterial perfect absorbers for single, dual, and multi-band applications reinforces the versatility and effectiveness of these designs, paralleling our sensor’s performance in the THz spectrum [55]. Lastly, studies on quad-band polarization-sensitive THz metamaterial absorbers provide insights into achieving high sensitivity and accuracy, which are crucial for our non-invasive cervical cancer diagnosis sensor [56]. In this manuscript presented an advanced diagnostic tool, the design of an exceptionally compact Triple-band Biosensor based on Metamaterials (MTMs) operating in the Terahertz (THz) region represents a novel innovation for Cervical Cancer diagnostics. Utilizing the unique properties of MTMs, this biosensor serves as a perfect absorber, exploiting resonant modes in the THz spectrum to achieve exceptional sensitivity. The discussion extensively explored various models, investigating factors such as polarization and the distribution of electric and magnetic fields. Of particular significance was the application of terahertz waves as a compelling approach for early cancer imaging. This promising method holds the potential to revolutionize non-invasive diagnostics, helping in transformative advancements in early cancer detection and subsequent treatment strategies. The triple-band configuration ensures precise and simultaneous detection of specific disease sign associated with cervical cancer, enhancing the diagnostic accuracy. The miniature size of the biosensor, facilitating non-invasive and portable applications, makes it an ideal candidate for point-of-care diagnostics, holding great potential for early detection. 2. Model’s unit cell layout The input signal of the THz structure carries information and energy. The incident wave interacts with the structure, which causes absorption, transmission, and reflection. The metamaterial-inspired structure helps for the absorption of EM waves. The minimum reflection indicates a maximum absorption of inputs. The transmitted waves pass through the structure absorb maximum energy and reduce transmission. The absorption using the proposed design helps to control transmission characteristics. The response of sensitivity, specificity and selectivity were taken into consideration during the design of the biosensor. The effectiveness of the sensor depends upon these variables. The challenging aspect for the sensor is to enhance specificity and sensitivity. To enhance clinical data and meaningful value for cancer detection optimisation of structure is required. The proposed optimised design structure is represented in Fig 1(C). The optimized structure was designed based on structures shown in Fig 1(A) and 1(B). Fig 1(A) shows a model 5 structure. Fig 1(B) shows the Model 2 structure. The Aluminum (Al) metal material is chosen as the conducting material and polyimide material is chosen as the substrate material. In the proposed structure the Aluminum metal is employed as the top and bottom layers of the model’s three-layered construction, with a polyimide dielectric spacer serving as the intermediate layer. The Aluminum has a conductivity of 3.56×107 S/m, and the dielectric spacer was 125 μm thick. The Al layers were 0.2 μm thick on top and bottom. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 1. Recommended structural design for a perfect absorber; (a) model 5, (b) model 2, and (c) model 10 (proposed biosensor design). https://doi.org/10.1371/journal.pone.0311431.g001 The combination of two structures helps reduce interfacing and overlapping of magnetic and electric forces. The dimensions of different parameters are represented in Table 1. The suggested absorber was constructed using a commercial full-wave finite integration approach (FIT) based on EM solver computer simulation technology (CST) in the microwave studio. Metamaterials (MTMs) may be analyzed using numerical computations to identify their properties and how they function across various frequency ranges and boundaries. Scientists have used numerical analysis to analyze various structures such as free space, unit cells, perfect electric and magnetic conductors, as well periodic arrays. To simplify simulation methods, a unit cell was allocated in the x and y-axis, with an open add-space in the z-axis. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 1. A complete list of the variables that have been adjusted for the recommended sensor. https://doi.org/10.1371/journal.pone.0311431.t001 To maximize power penetration and propagation in polyethene terephthalate (PET), the top metal layer was designed to match its impedance with the incident medium. To understand the transmission-line theory (TLT), the bottom AI layer is designed to block all incident electromagnetic (EM) waves and give zero impedance. The metallic layer at the bottom prevents transmission, and absorption devices with large electrical and/or magnetic losses collect moving waves. A top-plane electromagnetic (EM) wave was used to attain the absorption values. 3. Results and discussion This section aims in the comparative analysis of a wide range of compact absorber-based sensors targeting the operation in the Sub THz regime. To adequately cover the rich amount of possible architectures, in this work, ten different component designs are incrementally presented, simulated and evaluated with respect to its absorption spectrum, power flow and field density. The first design addressed is derived from a modified square shaped split ring resonator. One possible implementation is shown in Fig 2(A), where it was also simulated and its absorption curve was plotted in respect to the frequency. The absorption characteristics of the model presented two resonances, one at 0.7 THz and other near 0.95 THz, but the absorption peaks were far from ideal and a new set of absorber strategies were developed. A modified design was developed using a reduced symmetric split ring, targeting to achieve stronger absorption. This design was labeled Model 2 and is presented in Fig 2(B) together with its simulated absorption curve. Model 2 absorption curve also presented 2 peaks, but in this version, they were very narrower and the simulation resulted in much improved characteristics in the absorption peak at 0.7 THz. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 2. Comparison between the absorption properties of the two initial split ring designs: (a) shows the Model 1; and (b) shows the reduced symmetric ring version labeled as Model 2. https://doi.org/10.1371/journal.pone.0311431.g002 Targeting to obtain a better absorber at near THz frequency, two additional alternatives were designed. Model 3 consists of a D shaped split ring dipole and Model 4 is based on a dual ring approach mixing the characteristics of the previously studied Models. Fig 3 presents the design layout and the absorption curves for Model 3 and Model 4. The results for Model 3 presented an almost ideal absorption around 0.92 THz, being this design a very important candidate to compose the final sensor. Model 4 showed dual absorption peak characteristics in well behavior frequency locations, but with not near perfect absorbance. The promising results of Model 3 and the mixed output peaks found in Model 4 using its bi-component design triggered the design of two more versions of the absorber, basing them in the combination of the constituents of the Model 1 and Model 3 generating the Model 5 and with Model 2 and Model 3 generating the basis for Model 6. The combined designs Model 5 and Model 6 and its absorbance curves are presented in Fig 4. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 3. Comparison between the absorption properties of the Model 3 and Model 4 designs. https://doi.org/10.1371/journal.pone.0311431.g003 Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 4. Absorption characteristics of mixed models: (a) Model 5; and (b) Model 6. https://doi.org/10.1371/journal.pone.0311431.g004 The absorption curve obtained from Model 5 presented slightly better results in 0.88 THz in comparison to Model 3, but now it also has wide untuned side absorption sections. The results for Model 6 also presented the dual peaks seen before in Models 1 and two, but with improved absorbance. In order to provide a good coverage of the design options, mixed structures with circular rings were also investigated. Model 7 proposes an alternative using double circular rings in together with a split ring external section, as shown in Fig 5(A). On the other hand, Model 8 tests the opposite case where the D shaped ring was maintained and used along with an external circular ring, as explain in Fig 5(B). Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 5. Absorption characteristics for the designs with circular shaped structures: (a) Model 7 and (b) Model 8. https://doi.org/10.1371/journal.pone.0311431.g005 Model 7 with its triple element approach generated strong absorption in three main frequencies around 0.75 THz, 0.85 and 0.91 THz. Unfortunately, the results from the Model 8 were not adequate, and this directed the efforts to keep with the external rectangular split ring design. Two final design options were developed trying to achieve near three band THz absorption uniting the qualities of the triple element design, but without the fabrication difficulties of the circular structure and using the previously explored external split ring. Fig 6(A) presents an alternative design for the absorber labeled as Model 9 which uses losanges in place of the circular rings. Finally, Fig 6(B) shows Model 10, which comprises the integration of the D shaped internal structure (initially tested in Model 3) with the dual external rectangular split ring resonators. Fig 6(A) shows the simulation results of Model 9 which presented multiple absorption peaks, but just the band near 0.71 THz was near ideal. On the other hand, the absorption curve of Model 10 presented near perfect absorption characteristics in a triple band fashion including peaks at 0.684 THz, 0.918 THz and 0.971 THz. The structural design of Model 10, in conjunction with its mixed approach reusing several previously optimized subdesigns was very successful in the near THz range and was selected as a proposed model to be evaluated as the main biosensor of this work. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 6. These are the two designs’ distinct absorption characteristics: (a) Model 9; and (b) Model 10 (purpose biosensor). https://doi.org/10.1371/journal.pone.0311431.g006 To complete the analysis of the proposed resonator-based sensor, a set of simulations were made targeting to cover some possible available materials for fabrication of the device, both the substrate material and the conductor metalization options were tested and the results are presented in Fig 7. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 7. Absorption spectra for the proposed design under various: (a) substrate material and (b) resonator material conditions. https://doi.org/10.1371/journal.pone.0311431.g007 The simulated substrate materials were FR4, Polyimide, Arlon AD 410/430 and the famous Rogers Duroid RT 5870. Fig 7(A) shows very mixed results for this case, favoring Arlon and Polyimide. Deeper optimizations could be made for each specific case, but in order to direct the resonator to the biological application, an easily available polymer with good coupling was selected, the Polyimide. In relation to choice between the different available metals, most of the tested materials would be acceptable to be used, except for Iron. The simplicity of production and low cost of material, together with the good simulation results guided the current choice to Aluminum, as explain in Fig 7(B). This investigation delves into the performance nuances of a metamaterial biosensor concerning varying incident and polarization angles, specifically tailored for terahertz imaging in early cancer detection. Despite minor fluctuations, a robust, angle-independent feature consistently manifests across both incident and polarization analyses (Fig 8(A) and 8(B)), showcasing remarkable angular insensitivity. This intrinsic characteristic ensures uniform biosensor responses irrespective of incident wave angles or polarizations, underscoring its resilience and applicability in real-world scenarios. Despite slight variations in incident angles, a consistent and angle-independent feature, denoted by a dense red line in Fig 8(A), underscores the biosensor’s sensitivity to diverse angles of incidence. Polarization angle analysis (Fig 8(B)) reveals minimal deviations from incident angle results. The incident angle, representing the impact angle of the incident electromagnetic wave on the biosensor surface, holds significance in influencing interaction dynamics and signal specificity in terahertz imaging. Simultaneously, the polarization angle, indicating the orientation of the incident wave’s electric field vector, plays a crucial role in shaping the biosensor’s response to terahertz waves. The interplay between incident/polarization angles and the signal-to-noise ratio (SNR) emerges as crucial, where optimal angles and polarizations maximize signal strength and minimize noise, leading to heightened SNR. The observed angle-insensitivity, coupled with meticulous SNR optimization, attests to the exceptional efficacy of this metamaterial biosensor for robust terahertz imaging in early cancer detection. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 8. The impact of angle modification on absorption rate is investigated for: (a) incidence angle and (b) polarization angle. https://doi.org/10.1371/journal.pone.0311431.g008 In Fig 9, the presented images illustrate a significant electric field intensity ranging from 20,000 V/m to 300,000 V/m within a metamaterial biosensor, with notable resonances at 0.68 THz, 0.918 THz, and 0.971 THz. The spatial distribution of the electric field likely corresponds to or proximate to these resonance frequencies, unveiling distinct high (red) and low (blue) intensity regions. These variations suggest localized interactions between terahertz waves and the metamaterial, featuring intricate patterns indicative of intentional design for optimized biosensing. Each spectrum peak correlates with a resonance, demonstrating efficient coupling between the metamaterial and terahertz waves, concentrating the electric field. Strategically positioned high-intensity regions hold potential for interactions with cancer-specific biomolecules crucial for early detection. The triple resonance peaks indicate multi-band functionality, enabling detection of various cancer-related biomolecules at distinct frequencies within the 0.68–0.97 THz range. Analyzing electric field distributions for each resonance provides insights into field localization and resonant behavior across bands, offering opportunities for targeted biomarker detection. Fig 9(A)–9(C) likely depict electric field distributions within the biosensor at resonant frequencies (0.68 THz, 0.918 THz, and 0.971 THz) designed for early cancer detection. Spatial distributions reveal high-intensity (red) regions concentrated due to resonance and low-intensity (blue) regions, with complex patterns signifying interactions with surrounding tissues and potential biomarkers. Resonant behavior at each frequency corresponds to optimal coupling, resulting in enhanced electric field, particularly in high-intensity regions. Analyzing the location and intensity distribution across figures provides crucial insights into the biosensor’s detection capabilities at different frequencies, crucial for early cancer diagnosis. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 9. Distributions of the metamaterial structure field are shown on a colour map: (a) E-field at 0.684 THz, (b) E-field at 0.918 THz, and (c) E-field at 0.971 THz. https://doi.org/10.1371/journal.pone.0311431.g009 Observed magnetic field intensity variations between 100 A/m and 1000 A/m in Fig 10(A)–10(C) suggest significant enhancements within the metamaterial structure at resonances around 0.68 THz, 0.918 THz, and 0.971 THz. The spatial distribution of the magnetic field at these frequencies shows both high and low intensity regions, suggesting localized interactions between terahertz waves and the metamaterial. These patterns indicate intentional design features for optimized biosensing. High-intensity regions may facilitate interactions with cancer-specific biomolecules for early detection. The complex patterns in the H-field distribution are likely influenced by the metamaterial’s design, including materials, structure, and dimensions. Understanding these design aspects is crucial for interpreting field localization and overall biosensing performance. The triple-band functionality of the biosensor suggests potential detection of various cancer-related biomolecules at different frequencies within the 0.68–0.97 THz range. Analyzing H-field distributions for all three peaks could reveal differences in field localization and resonant behavior across bands, providing insights into targeting specific biomarkers. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 10. An illustrated colour map shows the field distributions of the proposed metamaterial structure: (a) H-field at 0.684 THz, (b) H-field at 0.918 THz, and (c) H-field at 0.971 THz. https://doi.org/10.1371/journal.pone.0311431.g010 The electric and magnetic field distributions within our metamaterial-based biosensor are pivotal for detecting cancer-specific biomolecules in the terahertz regime. Leveraging a triple-band design, the biosensor generates localized regions of heightened electric fields within its cavity, facilitating enhanced interactions with polar biomolecules prevalent in cancer cells, such as proteins and nucleic acids. This intensified interaction results in elevated terahertz absorption by cancerous tissues compared to healthy counterparts. Additionally, magnetic field analysis provides essential spatial sensitivity, enabling precise localization and assessment of cancerous lesions. While establishing definitive electric field intensity thresholds for clinical applications presents challenges due to biological variability, ongoing research aims to further elucidate these relationships through comprehensive studies with biological samples, aiming to enhance diagnostic accuracy and clinical efficacy. To gain a deeper understanding of the physical absorption mechanisms inherent in the proposed metamaterial absorbers, an examination of the surface current distribution within the upper layers of the recommended design was undertaken. The presence of circular antiparallel current flow was identified, indicative of a robust magnetic response. Fig 11(A) elucidates a magnetic dipole resonance mode, portraying parallel and antiparallel surface current patterns at 0.68 THz. Furthermore, Fig 11(B) and 11(C) delineate the current distributions corresponding to the second resonance modes at 0.918 THz and the third resonance modes at 0.971 THz, respectively, both exhibiting pronounced magnetic responses. The interplay between parallel and antiparallel current flows within these distributions governs the resultant electric and magnetic field distributions, contributing to the opposition between internally generated and externally incident magnetic fields (H-field). A comprehensive examination of power flow was conducted at the surface area of the recommended sensor. Meanwhile, observed variations in surface current intensity ranging from 10 A/m to 100 A/m in Fig 11(A)–11(C) signify substantial enhancements within the metamaterial structure, particularly at resonances around 0.68 THz, 0.918 THz, and 0.971 THz. These figures are interpreted as spatial representations of surface current distributions, revealing regions of both high (red) and low (blue) intensity, indicative of localized interactions with terahertz waves. The intricate patterns observed are a result of intentional design features aimed at optimizing biosensing performance. Each resonance peak within the spectrum corresponds to efficient coupling, strategically concentrating surface current density for potential interactions with cancer-specific biomolecules. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 11. The surface current distribution of the recommended metamaterial design: (a) at 0.684 THz, (b) at 0.918 THz, and (c) at 0.971 THz. https://doi.org/10.1371/journal.pone.0311431.g011 The triple-band metamaterial biosensor exhibits a compelling power flow profile, particularly at resonances of 0.68 THz, 0.918 THz, and 0.971 THz, emphasizing its efficacy in early cancer detection. The power flow analysis, depicted in Fig 12(A)–12(C) for each resonance point, underscores the biosensor’s robust capability. These results affirm intense power flows, especially at the resonant frequencies, indicative of the biosensor’s proficiency in detecting cancer cells. The intentional design of the biosensor strategically amplifies power flow, ranging from 10,000,000 V.A/m2 to 100,000,000 V.A/m2, as observed in the metamaterial absorbers. The intricate patterns in the power flow distribution highlight deliberate design features for biosensing applications. The high-power flow of the biosensor holds crucial advantages for terahertz imaging. By concentrating power flow at specific resonances, the biosensor enhances sensitivity, allowing for targeted interactions with biomolecules. This focused power flow within the metamaterial structure aligns with the shared principle of resonant enhancement in terahertz imaging, amplifying the potential for biomolecule detection. The triple-band functionality further broadens the biosensor’s utility, promising the detection of cancer-related biomolecules across the 0.68–0.97 THz range. The correlation with terahertz imaging underscores the biosensor’s emphasis on metamaterial power flow enhancements, showcasing its unique approach centered on directing and intensifying power flow for optimal interactions with terahertz waves and subsequent indirect cancer biomolecule detection. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 12. An examination of the power flow at: (a) at 0.684 THz, (b) at 0.918 THz, and (c) at 0.971 THz. https://doi.org/10.1371/journal.pone.0311431.g012 The elucidation of surface power loss patterns in Fig 13(A)–13(C) at resonant frequencies (0.68 THz, 0.918 THz, and 0.971 THz) within the metamaterial structure offers indispensable insights into the operational characteristics of the triple-band biosensor. The discernible high (red) and low (blue) intensity regions denote variations in surface power loss density, with the red regions indicating substantial surface power loss attributed to resonance effects. This amplified surface power loss is especially advantageous for terahertz imaging, enhancing the biosensor’s utility in early cancer detection. The resonance-induced heightened surface power loss assumes a pivotal role in the interaction with cancer-specific biomolecules, thereby augmenting the biosensor’s sensitivity and effectiveness in detecting subtle cellular changes associated with early-stage cancer. The intricacies of the observed patterns within the metamaterial structure are ascribed to the nuanced design, encompassing materials, structure, and dimensions. Discrepancies between Figs 13–15 emanate from distinctive resonance behaviors at each frequency, exerting influence on the location and distribution of high-intensity regions. Furthermore, elucidation of the color scale, ranging from 100,000 W/m2 to 1,000,000 W/m2, is imperative for the comprehensive contextualization and interpretation of intensity variations within the figures. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 13. An examination of the surface power loss at: (a) at 0.684 THz, (b) at 0.918 THz, and (c) at 0.971 THz. https://doi.org/10.1371/journal.pone.0311431.g013 Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 14. An examination of the electric energy density at: (a) at 0.684 THz, (b) at 0.918 THz, and (c) at 0.971 THz. https://doi.org/10.1371/journal.pone.0311431.g014 Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 15. An examination of the magnetic energy density at: (a) at 0.684 THz, (b) at 0.918 THz, and (c) at 0.971 THz. https://doi.org/10.1371/journal.pone.0311431.g015 4. Diagnosis of cervical cancer The cervical cancer diagnostic biosensor integrates advanced technology to detect and diagnose cervical cancer in its early stages, overcoming present limitations. Utilizing microwave imaging (MWI), it enhances signal transmission and absorption analysis, improving the detection of cancer cells. Fig 16 illustrates the suggested detection of cervical cancer, employing two coverslips for error-free findings. A HeLa cell carcinoma sample, with a refractive index of 1.392 for malignant tissue and 1.368 for healthy cervical tissue, is positioned between the coverslips [57–61]. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 16. The suggested biosensor investigates the absorption coefficient in both healthy cervical and cervical-cancer. https://doi.org/10.1371/journal.pone.0311431.g016 In Figs 17(A) and 18(A), the results of cervical cancer identification against healthy cervical tissue are presented. The examination focused on distinct peaks within the terahertz (THz) range. Fig 17(B) highlights a substantial difference between clean cervical tissue and cervical cancer, corresponding to 0.00119 THz (1190 MHz). Similarly, Fig 18(B) illustrates a significant distinction of 0.00163 THz (1630 MHz) between healthy cervical tissue and cervical cancer, observed in the third peak between 0.930 THz and 0.942 THz. These findings underscore the efficacy of terahertz imaging in discriminating between healthy and malignant cervical tissues, providing valuable insights for early cancer detection applications. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 17. The proposed biosensor detects the absorption coefficients for normal Cervical and Cervical cancer; the frequency range is between: (a) 0–1 THz, (b) 0.663–0.670 THz. https://doi.org/10.1371/journal.pone.0311431.g017 Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 18. The proposed biosensor detects the absorption coefficients for normal Cervical and Cervical cancer; the frequency range is between: (a) 0–1 THz, (b) 0.930–0.942 THz. https://doi.org/10.1371/journal.pone.0311431.g018 Fig 18 illustrates the absorption coefficients for normal and cancerous cervical tissues across the 0–1 THz frequency range, with a focused view from 0.930 to 0.942 THz. The biosensor’s functionality in the THz range is demonstrated by detecting and distinguishing between healthy and malignant tissues based on their distinct absorption characteristics. While absorption coefficients can sometimes be inconsistent due to tissue heterogeneity and measurement conditions, the frequency shift observed between healthy and cancerous tissues provides a more reliable diagnostic indicator. Specifically, the cancerous tissue shows a notable frequency shift at the third resonant peak compared to the normal tissue. This frequency shift, centered at approximately 0.93536 THz for cancerous tissue and 0.93699 THz for normal tissue, is a robust marker for identifying malignant tissue. The metamaterial-based design of the biosensor enhances its sensitivity, allowing it to operate as a perfect absorber at specific resonant frequencies, thereby providing a dependable method for early-stage cervical cancer diagnosis. In the assessment of triple-band perfect absorber metamaterials serving as sensors, critical parameters include the sensitivity (S), Figure of Merit (FOM) and the quality factor (Q factor). The FOM quantifies sensor efficacy by dividing a key performance parameter, such as sensitivity, by another relevant metric like bandwidth or noise level. Simultaneously, the dimensionless Q factor evaluates the resonance quality of the system. Elevated values of both FOM and Q coefficients in sensing applications signify heightened sensor performance. Sensitivity of perfect metamaterial absorbers, gauged by their response to variations in the surrounding medium’s refractive index, is evaluated through the absorber’s resonance frequency or absorption peak shift. The observed heightened sensitivity suggests its potential for discerning subtle refractive index changes, making it suitable for applications such as early cancer detection. Sensitivity plays a pivotal role in securing accurate results, especially when confronted with minimal refractive index fluctuations. A comprehensive evaluation of a sensor’s efficiency and applicability in diverse sensing applications is derived from the full output measurement FOM, embodying its selectivity calculated as sensitivity normalized to the full width at half maximum (FWHM) of the resonant dip. (2) The sharpness of the resonance is measured using the formula-based Qfactor. (3) Resonant wavelength, denoted by λ. Terahertz (THz) biosensors significantly increase the sensitivity of terahertz (THz) detection when used in a metamaterial-oriented manner. In scientific literature, sensitivity is commonly interpreted in two ways. The first is frequency sensitivity. (4) where Δf is the resonance peak’s frequency shift and Δn are the change in refractive index (RI), which is frequently expressed in terms of refractive index units (RIU). An alternative approach is to compute S = ΔI⁄Δn, the intensity frequency, in which ΔI represents the resonance intensity variation [62, 63]. A penicillin detection biosensor, leveraging metamaterial technology, exhibits notable performance metrics, including a figure of merit (FOM) of 0.1216 RIU-1, a quality factor of 5.58, and sensitivity (S) of 0.02432 THz/RIU [64]. A distinct biosensor tailored for breast cancer detection displays a quality factor of 2.43, a figure of merit (FOM) of 2.75 RIU-1, and a sensitivity (S) of 1.21 THz/RIU [65]. Additionally, a biosensor designed for avian influenza virus detection showcases a sensitivity (S) of 1.06 THz/RIU and a figure of merit (FOM) of 0.166 RIU-1, although the quality factor is not explicitly provided [66]. Furthermore, our prior research conducted in 2023 introduced a dual-band biosensor tailored for the early detection of nonmelanoma skin cancer. The biosensor exhibited notable performance metrics, including a figure of merit (FOM) of 0.86 and 1.15 RIU-1, a quality factor of 12.8 and 13.5, and sensitivity (S) of 0.0515 and 0.076 THz/RIU, correspondingly for the first and second peaks. We have proposed a biosensor in this paper, compared with the results obtained in the above-mentioned papers, we believe will be a good alternative as a biosensor for detecting early detection of cervical cancer. The proposed biosensor features three resonance bands, each characterized by specific quality factors, figures of merit, and sensitivities. The first resonance band exhibits a Q-factor of 85.77, FOM of 6.3 RIU-1, and S of 0.049 THz/RIU. Since second resonance has not reached perfect absorber (80%) hence we neglect it. Also, the third resonance band demonstrates a Q-factor of 41.46, FOM of 3 RIU-1, and S of 0.068 THz/RIU, particularly applied for cervical cancer diagnosis. The results indicate promising avenues for biosensing applications in cervical cancer research, showcasing the potential of terahertz imaging to enable accurate and early diagnosis, guiding timely interventions and treatments in healthcare settings, this biosensor represents a substantial development, as detailed in Table 2. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 2. Comparisons of bio-sensing performance of various sensor applications based on THz metamaterial. https://doi.org/10.1371/journal.pone.0311431.t002 The study employed microwave imaging techniques in the terahertz band to validate the findings. A HeLa cell sample on a coverslip was tested using the proposed biosensor, as depicted in Fig 19. Terahertz E-field imaging with a triple-band, micron-sized metamaterial perfect absorber biosensor shows promise for detecting cervical cancer. The notable contrast in E-field intensity, as seen in Fig 20(A), suggests potential early-stage diagnostic capabilities, warranting further research for robust clinical translation. High E-field intensity hotspots corresponding to cervical cancer likely arise from distinct tissue dielectric properties and enhanced resonant interactions at specific terahertz frequencies, as shown in Fig 20(B). These hotspots indicate the potential presence of cervical cancer, with disparate dielectric properties affecting terahertz wave interactions. Molecules’ vibrational resonances within tissue, like water and proteins, further contribute to differential dielectric properties, influencing terahertz wave propagation and scattering. This integrated understanding provides a comprehensive framework for interpreting hotspots, advancing non-invasive cervical cancer diagnostic methodologies. Furthermore, analogous investigations were undertaken for the subsequent selections of E-field. The E-field microwave imaging (MWI) technique results depicted in Fig 21 demonstrate the biosensor’s effectiveness in detecting cervical cancer by analyzing electric field intensity variations at 0.918 THz. The images reveal distinct differences in electric field distribution between healthy and cancerous cervical tissues, with cancerous tissues exhibiting significant variations due to their higher permittivity and conductivity. These variations create detectable anomalies, which the biosensor, enhanced by its metamaterial-based design, can accurately identify. This sensitivity to electric field changes allows for non-invasive, early-stage cervical cancer detection, providing a rapid and patient-friendly alternative to traditional diagnostic methods. The findings highlight the biosensor’s capability to not only detect cancer but also estimate the spatial extent of malignant regions, essential for effective diagnosis and treatment planning. Figs 21(A) and 22(A) delineate a zone characterized by diminished electric field intensity, indicative of a healthy cervical. Conversely, Figs 21(B) and 22(B) depict an area exhibiting heightened electric field intensity, suggestive of cervical cancer. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 19. The diagnosis of cervical cancer using the MWI approach. https://doi.org/10.1371/journal.pone.0311431.g019 Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 20. The E-field MWI technique results at 0.684 THz: (a) healthy Cervical, (b) Cervical cancer. https://doi.org/10.1371/journal.pone.0311431.g020 Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 21. The E-field MWI technique results at 0.918 THz: (a) healthy Cervical, (b) Cervical cancer. https://doi.org/10.1371/journal.pone.0311431.g021 Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 22. The E-field MWI technique results at 0.971 THz: (a) healthy Cervical, (b) Cervical cancer. https://doi.org/10.1371/journal.pone.0311431.g022 The terahertz biosensor in our study employs advanced magnetic field (H- field) analysis to provide precise spatial insights into cervical tissue. By detecting nuanced deviations in magnetic field patterns that correlate with tissue dielectric properties, the biosensor effectively localizes and maps malignant lesions. This capability enhances diagnostic accuracy by distinguishing between healthy and cancerous tissues early in disease progression. These spatial insights not only facilitate timely intervention but also optimize treatment strategies tailored to individual patient needs, thereby advancing the efficacy of cervical cancer diagnostics and therapeutic outcomes. Exploration into magnetic fields for terahertz imaging shows promise in cervical cancer detection, notably with terahertz H-field imaging via microwave biosensors. An intense magnetic field envelops the specimen, yet within this field, a region of diminished magnetic field density is discernible, denoting a healthy cervical, as elucidated in Fig 23(A). Conversely, Fig 23(B) illustrates an area characterized by elevated magnetic field intensity at the center of the image, indicative of cervical carcinoma. This contrast aids early diagnosis, influenced by differing magnetic properties stemming from tissue composition and molecular distributions. Biosensor resonance at specific frequencies enhances local magnetic fields, potentially amplifying signals within cancerous tissue. Such findings underscore magnetic fields’ potential in terahertz imaging for cervical cancer detection, warranting further research and clinical validation. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 23. The H-field MWI technique results at 0.684 THz: (a) healthy Cervical, (b) Cervical cancer. https://doi.org/10.1371/journal.pone.0311431.g023 To further validate the biosensor’s applicability for early cancer detection, supplementary investigations were conducted concerning electrical and magnetic energy density. Fig 24(A) depicts an area with low electrical energy density, suggesting healthy cells. In contrast, Fig 24(B) reveals a region with markedly high electrical energy density, indicative of cancerous cells. Similarly, Fig 25 presents the magnetic energy density distribution at 0.684 THz for both healthy and cancerous cervical tissues, revealing insights into the biosensor’s interaction with differing tissue types based on their distinct dielectric properties. In healthy tissue (Fig 25(A)), magnetic energy density shows a uniform distribution, predominantly centralized, indicating minimal magnetic field disturbance and consistent dielectric properties conducive to stable absorption patterns. Conversely, cancerous tissue (Fig 25(B)) displays heterogeneous magnetic energy density with irregular, intense regions, reflecting altered dielectric properties and significant disruption of the magnetic field due to the presence of cancerous cells. This disparity underscores the biosensor’s efficacy in distinguishing between healthy and cancerous tissues, utilizing the irregular magnetic energy density patterns and heightened absorption in cancerous regions as robust indicators of malignancy. Leveraging magnetic energy density distributions enhances the biosensor’s diagnostic precision, offering a reliable method for early-stage cervical cancer detection in the terahertz spectrum, even amidst varying absorption coefficients and tissue heterogeneity. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 24. The electric energy density MWI technique results at 0.971 THz: (a) healthy Cervical, (b) Cervical cancer. https://doi.org/10.1371/journal.pone.0311431.g024 Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 25. The magnetic energy density MWI technique results at 0.684 THz: (a) healthy Cervical, (b) Cervical cancer. https://doi.org/10.1371/journal.pone.0311431.g025 Our proposed terahertz biosensor, outlined in Table 3, differs from conventional microwave methods by utilizing frequency-domain data to detect changes in refractive index, providing spatial insights into the target. This suggests that biosensors could revolutionize Cervical cancer diagnostics by enabling more accurate detection of cancer indicators, facilitating early identification and treatment. Early-stage intervention is critical for improving prognosis and survival rates, making biosensors valuable tools for healthcare practitioners in detecting biomarkers associated with Cervical cancer. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 3. A comparison between the terahertz band study on perfect metamaterials and the recommended biosensor. https://doi.org/10.1371/journal.pone.0311431.t003 5. Future perspective Developing more sensitive biosensors for early cancer detection, such as terahertz (THz) electromagnetic (EM) wave imaging biosensors for the early identification of blood, colon, breast, adrenal gland (PC-12), and Non-Melanoma Skin Cancer (NMSC), among other cancers. 6. Conclusions In this manuscript, we introduced an innovative structure of a biosensor intended for dependable early-stage diagnosis of cervical cancer. The proposed device operates within the terahertz (THz) spectrum. Its geometrical details are carefully developed to ensure triple-band operation as a perfect absorber, thereby enhancing sensitivity. An essential component of the sensor is a metamaterial layer, which facilitates realization of the required absorption characteristics. The operating principles of the device involve discriminating between healthy and cancerous tissue based on their different dielectric properties. Suitability of the suggested sensor for cervical cancer detection has been demonstrated through comprehensive numerical studies. In particular, a number of specific case studies were carried out involving healthy and cancer-affected cervical. The analysis of magnetic field strength allows for detecting the latter as well as for determination of its spatial extent. The proposed sensor was also compared to a number of state-of-the-art designs reported in the literature showing its competitive performance, also in terms of utilizing unconventional approach (here, the employment of frequency-domain data for detecting changes in refractive index).
TI - Design and validation of ultra-compact metamaterial-based biosensor for non-invasive cervical cancer diagnosis in terahertz regime
JF - PLoS ONE
DO - 10.1371/journal.pone.0311431
DA - 2025-02-03
UR - https://www.deepdyve.com/lp/public-library-of-science-plos-journal/design-and-validation-of-ultra-compact-metamaterial-based-biosensor-bACu4hGGuS
SP - e0311431
VL - 20
IS - 2
DP - DeepDyve
ER -