We collected literature from the Web of Science Core Collection (WOSCC) database and conducted a literature search on June 6, 2024. The data retrieval strategy and collection process are shown in Figure 1. Briefly, the search strategy was set as “TS = (Electrochemical Biosensor) AND TS = (Tear OR Eye Diseases) AND publication date = (01/01/2000 to 31/12/2023),” then 117 papers were retrieved. The type of publication was restricted to original articles and reviews, and the language was limited to English. Finally, a total of 114 documents were included. This process was executed independently by two researchers, and when disagreements arose, the decision was made by a third researcher.

VOSviewer (version 1.6.18), CiteSpace (version 6.1.R6), Microsoft Office Excel (version 2021) and Scimago Graphica were used for literature data extraction and visualization analysis. VOSviewer is a software developed by van Eck and Waltman for bibliometric analysis and mapping (20). In this study, VOSviewer was used for the co-authorship analyses of countries/regions, institutions and authors, the co-citation analyses of journals and authors, and the co-occurrence analysis of keywords. Furthermore, combined with the Scimago Graphica, VOSviewer was utilized for the countries/regions relationship analysis and visualization. In the network map developed by VOSviewer, different nodes represent different elements, such as countries, institutions, authors, journals and keywords. Node size represents the quantity or frequency of elements, while links between nodes reflect the level of collaboration or co-citation. In network visualization, the same color signifies the same cluster. The overlay visualization illustrates the temporal relationship among the nodes. The purple nodes indicate earlier publication times, while the yellow nodes represent more recent ones.

CiteSpace, another widely used bibliometric analysis and visualization tool developed by Professor Chaomei Chen (21), was used to conduct a co-citation network analysis of reference. The citation burst intensities of the references and keywords were also examined using CiteSpace to identify rapidly expanding topic keywords and emerging areas. Moreover, based on the data exported from CiteSpace and the cited report of WOSCC, we used Excel to create an annual publication and citation curve.

A total of 114 articles were determined in this study, including 67 original articles (86.4%) and 47 reviews (41.23%), the total number of citations was 7,168 times and the average number of citations reached 62.88. This research came from 640 authors in 33 countries and 243 institutions and was published in 74 journals. 8,131 references from 2037 journals were cited.

Although the inclusion criteria for literature in this study spanned from January 1^st, 2000 to December 31^st, 2023, it is noteworthy that the emergence of initial research in this field only occurred after 2008. Furthermore, there has been a consistent upward trajectory since then, with a particularly significant acceleration in growth observed since 2019. Eighty-seven articles were published from 2019 to 2023, accounting for 76.32% of the total (Figure 2). Moreover, there has been a substantial increase in the number of citations to the literature since 2019, particularly within the past 3 years when annual citations surpassed 1,000. This trend signifies an escalating level of attention and significance accorded to this field by researchers.

Scholars from 33 countries/regions contributed to this field. Figure 3A illustrates the number of publications, total citation count, and average citation count for the top 10 most prolific countries/regions. China is the most prolific country in terms of the number of publications (n = 41), followed by the United States (n = 29), and South Korea ranks third with 11 publications. It is noteworthy that the United States ranked 2nd in terms of the number of articles published, but it had the highest average number of citations (n = 145.8). Conversely, China has the highest number of articles but a lower average number of citations (n = 32.6). In addition, Figure 3B shows the annual publication trend of the top 3 countries in terms of publication volume, and it can be seen that all 3 countries will have a significant increase in the number of publications by 2020. Figure 3C depicts the cooperative network relationship among the 25 countries with close cooperation. The United States has the broadest international collaborative network with 10 countries and the highest intensity of collaboration (TLS = 16), which suggests that the USA is more influential in this area of research.

Two hundred and forty-three institutions were involved in this field, Figure 4A presents the number of publications, total citation count, and average citation count for the top 10 most prolific institutions, and Figure 4B is a density visualization of the number of publications for each institution. The University of California San Diego has the most publications (n = 9) as well as the highest total citations (n = 246.8), while the California Institute of Technology tops the list with 109.5 total citations and the highest average citations (n = 365.0) even though it only published 3 articles. Among the top 10 most prolific research institutions, 4 are located in the United States, 3 in China, and the rest are located in Iran, Sweden, and South Korea, respectively. The overlay visualization of the collaboration network among 243 institutions is presented in Figure 4C. Notably, the University of California San Diego exhibits a high level of collaboration intensity (TLS = 11) with 10 prominent partner institutions.

In this field, 640 authors contributed to various studies, with a core group of 45 authors engaging in close collaboration. An overlay visualization was used to illustrate the collaborative interactions among 45 authors, categorized by their average publication year, as depicted in Figure 5A. The yellow nodes represent a more recent average publication year, while the purple nodes indicate an earlier one. As shown in the figure, Chen et al. are the most recent active researchers among the 45 authors, with an average publication year of 2023. Figure 5B and Table 1 show the five most prolific authors, with Joseph Wang leading both categories with seven published papers and 2,443 citations. Joseph Wang also formed collaborations with 34 authors, showing the highest collaboration intensity (TLS = 43). The study involved 6,408 co-cited authors, with 79 authors cited at least 10 times, as shown in Figure 5C. Figure 5D and Table 1 show the five most co-cited authors, with Jayoung Kim from the United States emerging as the most cited author, totaling 178 citations. Despite having only three relevant publications, Jayoung Kim’s work received 1,938 citations, averaging 646 citations per paper, which highlights his significant impact on the field.

A total of 74 journals published articles in this research area and Figure 6A shows the overlay visualization of 74 journals for citation analysis. Table 2 provides a summary of the top 10 journals based on publications and co-citations. After sorting by the number of publications, Biosensors & Bioelectronics with 7 published articles in the field at the top of the list, and ACS Applied Materials & Interfaces is second with 5. In terms of citations, Biosensors & Bioelectronics also ranked first with 614 citations, followed by Analytical Chemistry (377 citations) and Sensors (279 citations). Among them, Biosensors and Bioelectronics has the highest impact factor of 12.6 in 2024. Notably, ACS Sensors ranked first with an average of 105.5 citations, despite having published only 2 articles in this field. As for the co-cited aspect, there are 2037 co-cited Journals involved in this study, among which 76 journals have 20 co-citations (Figure 6B). Biosensors and Bioelectronics leads significantly with a count of 888 co-citations, which is far ahead of the second-ranked Analytical Chemistry (454 co-citations), and Sensors and Actuators b-Chemical ranks third with 419 co-citations.

Out of the 114 articles included, Table 3 lists the top 10 most cited articles. Among them, a review entitled “wearable biosensors for healthcare monitoring” was published in Nature Biotechnology by Wang et al. in 2019, which has been mostly up to 1759 times by 20th July 2024 (2). This review provides a comprehensive summary of the recent advances in wearable biosensor technology and identifies key scientific and technical issues that need to be addressed in the future, providing directional guidance for the further development of the field.

Two or more papers that are simultaneously cited by one or more papers are said to constitute a co-citation relationship (22). Based on CiteSpace, we created a co-occurrence analysis network (Figure 7A), cluster analysis network (Figure 7B), and burst analysis of co-cited references (Figure 7C). In Figure 7A, the outer rings of Bandodkar AJ (2014) (23) and Chen YH (2017) (24) nodes are in purple, indicating high betweenness centrality and playing a pivotal bridging role (25), emphasizing their instrumental role in facilitating collaboration in the field of electrochemical biosensors and tears/eye disease research. In our co-cited reference clustering analysis, the Modularity Q value reached 0.6912 (>0.3), which implied the robustness of the clustering structure. Moreover, the weighted mean silhouette score reaches 0.8266 (>0.7), which implies a high degree of consistency among the members of each cluster (26). Six clustering results are presented in Figure 7B, “interstitial fluid,” “wearable devices,” “glucose monitoring “, “electrochemical,” “electrochemical sensors,” and “hydrogel,” these clusters reflect the research frontiers in the field of ocular biosensors.

A co-citation burst refers to a citation that appears with high frequency in an article published within a condensed timeframe, indicating increased interest and relevance within the research community. The red line indicates the duration of the burst and the blue line indicates the time from the appearance of the term to its end. Figure 7C shows the details of the top 10 bursts of co-cited literature. Yao HF (2011) (27) and Bandodkar AJ (2015) (28) show the longest duration of bursts, while Kim J (2015) (29) and Bandodkar AJ (2014) (23) have the highest burst intensity.

A co-occurrence analysis of all keywords was performed using VOSviewer to identify impacts and trends within the field, which includes a total of 792 keywords, Figure 8A shows an overlay visualization of 43 keywords that appeared at least five times. Interconnections between the keywords denote co-occurrence, with purple denoting earlier appearing keywords, and yellow representing more recent ones. This network diagram illustrated that the keyword “biosensor” has the highest frequency of occurrence (n = 50), with an average appearance year (AAY) of 2019. “Tear glucose” (AAY = 2020) follows as the second most frequent (n = 24). Notably, “diagnosis” (AAY = 2022) and “in-vivo” (AAY = 2021) are recent keywords with at least five occurrences, indicating emerging interests in this research domain. To further understand the evolution of research topics, we utilized CiteSpace to examine keyword bursts. Figure 8B illustrates the top 10 keywords with the highest burst intensity, where “sensors” and “contact lens” have the longest burst duration, spanning over 5 years, and “uric acid” exhibits the highest burst intensity.

In the early screening and diagnosis of diseases, highly sensitive and convenient tests can help us predict the future course of a condition and accurately assess a patient’s health. With the continuous development of electrochemical biosensors, the medical field is witnessing a new type of detection method. This technology enables the detection of samples by collecting and filtering multiple biological samples and integrating them into a chip (33). With the advantages of high sensitivity, low cost, and non-invasiveness, this technology brings great convenience to patients and healthcare professionals. The application of electrochemical biosensors in medicine has achieved remarkable results from the initial detection of glucose content (1, 34, 35), to applications in a variety of fields such as the detection of pathogenic microorganisms (36), diagnosis of respiratory diseases (37–39) and screening of cancer (40–42).

In this field, research using tears as the biomarkers of electrochemical sensors is of particular interest. Tears are a fluid naturally secreted by the human body, which is not only easily accessible but also contains a great deal of biological information. Tear is a meaningful biomarker source and specific diseases showed distinguished proteomic patterns in tears. Tears can be considered an ideal biomarker source because the body is exposed to different pathophysiological conditions during disease periods and the tear content changes accordingly (9). According to proteome analysis of tears, this fluid has approximately 500 proteins; on the other hand, plasma may contain up to 10,000 proteins (43). The composition of its proteins also varies with the disease (44–47). Despite the differences in the components between blood and tears, a portion of biomolecules in the blood can still diffuse and enter into the tear, which probably shows a close correlation between blood and tears. Thus, the biomarkers in tears can be also used to a certain extent to replace those in the whole body, which provides a promising potential for the diagnosis of ocular diseases (8). In addition, the components of tears are not as complicated as blood and contain a variety of trace elements (e.g., potassium, manganese, chloride and sodium ions), biomolecules (e.g., glucose, urea), gamma globulin (e.g., IgA, IgG and IgE) and enzymes (e.g., lysozyme), which can act as vital biomarkers for the diagnosis of diseases and make tears an attractive diagnostic biofluid (43, 48).

By detecting specific substances in tears, we can achieve rapid, non-invasive diagnosis of a wide range of ocular and systemic diseases. Many novel protein-based biomarkers have been correlated to specific ocular diseases, namely dry eye syndrome, trachoma, glaucoma, keratoconus, allergic conjunctivitis, allergy with corneal lesions, blepharitis, conjunctivochalasis, mycotic keratitis, pterygia, thyroid-associated orbitopathy, etc., and systemic disorders, namely diabetes mellitus, cancer, systemic or multiple sclerosis, cystic fibrosis, Parkinson’s disease, sclerosis, etc. (9, 23, 49). Many methods have been used to identify quantitative changes in protein expression in a biological system. Relatively speaking compared to other body fluids, tears are a unique source of protein. Many researchers have therefore focused on changes in the proteomic, lipidomic and metabolomic composition of the tear film during disease.

For the different substances in tears, the methods used to detect them vary. For certain metabolites in tears, such as glucose, lactate and uric acid, an enzymatic amperometric assay is commonly used to detect (50, 51). For proteins and small molecules in tears, voltammetry using bio-affinity recognition is generally used to detect the target substances (52–55). For the detection of ionic components such as the electrolytes Na⁺, K⁺ and Cl-in tear, the presence of the target substance is detected by measuring the potential signal associated with a change in surface charge at the sensing electrode, using a potentiometric biosensor with a high degree of particle selective penetration (51, 56–59). Among the most commonly used methods of tear sensors for contact lens detection are organic electrochemical transistors (OECT) biosensors and electrochemiluminescence (ECL) sensors. OECT sensors not only have the advantages of low cost, high sensitivity, and low operating voltage (<1 V), but they can also detect a wide range of electroactive substances (e.g., dopamine, glucose, and epinephrine) (60, 61), as well as non-electroactive substances (e.g., cortisol, deoxyribonucleic acid (DNA), proteins, etc.) (62–64). ECL sensors combine the advantages of optical and electrochemical analyses to quantify target molecules by electrochemically emitting different light to emit signals that are associated with target organisms (e.g., glutathione, lactate, DNA, microRNA) (65–67)to identify the luminescence-inducing active substances, and electrochemiluminescence bioimaging to perform cellular Analysis. Electrochemiluminescent cellular bioimaging has been carried out for different small molecules released in cells and for membrane protein targets on the cell surface (68–70). The application of these techniques has led to a deeper study of the tear.

In addition to water and electrolytes, tears carry low molecular weight organic compounds, proteins, enzymes, and other biomolecules that can be used for disease screening (7). In the application of biosensors, electrochemical sensors have shown unique advantages in detecting biological components in tears. Generally speaking, most biomarkers should be integrated into the biosensors by functional modification such as oxidoreductases, antibodies/antigens, or enzymes, which aims to improve the selectivity and sensitivity of the biosensors for a specific biomarker (27, 71–73). Concerning the detection of proteins and organic small molecules, electro-active materials (e.g., enzymes, nanomaterials and electro-active molecules) need to be additionally modified to the biosensor interface since those proteins and organic molecules are usually electrochemically inactive and cannot directly produce electrochemical signals (74–77). Several typical electrode materials widely used by researchers include precious metals such as silver, gold, and platinum, as well as diverse carbon-based materials such as non-metallic carbon materials including graphene, graphite, carbon nanotubes, and glassy carbon (78). Different carbon substitutes are used for different tear biomarkers to achieve sensitive modulation of electrode performance, thus achieving precision and accuracy of detection (79). In addition, to further enhance the performance of electrochemical sensors, researchers often employ polymers as modifying materials for the electrodes and finely modify the electrode surfaces by introducing nanofibres or nanoparticles, etc., to effectively improve the sensitivity and responsiveness of the sensors (78, 79).

A glucose oxidase-modified biosensor, pioneered by Leland Clark Jr. in 1960, enabled electrochemical detection of a diabetes biomarker (i.e., glucose), opening up a new avenue for early diagnosis of analytes in the biosensor field (1). The use of an electrical biosensor device inserted into the lacrimal gland to detect glucose and health-related biomarkers in tears was first proposed in 2008 by Joseph Wang of the University of California, San Diego, as an alternative to traditional blood sampling (80). Yan et al. developed an electrochemical sensor with low sampling volume, good selectivity, and high reproducibility, which enabled the first noninvasive detection of glucose in rabbit tears (81). In subsequent studies, more and more researchers have tried to change the electrode materials and compositions by, for example, using graphene electrodes and silicon oxide electrodes (82), nanoporous gold electrodes, etc., to improve the sensitivity and specificity of the assay and enable more accurate identification of components in tea (83–85). To achieve a more convenient and non-invasive monitoring method, Yao et al. from the University of Washington developed a contact lens that is non-invasive and allows continuous monitoring of tear glucose levels over a long period (86). With the advent of these electrochemical sensing contact lenses, we have transitioned from traditional invasive and invasive testing to a non-invasive, portable and wearable era. This breakthrough undoubtedly opens up unlimited possibilities for the application of biosensors in medicine (87–89).

Researchers and academics are convinced that smart wearables will inevitably play a central role in the future of healthcare (2, 90–92). This trend has been widely recognized and a large number of researchers have begun to focus on contact lenses. Contact lenses are more than just a fashion accessory; they can monitor a variety of biomarkers through tears, providing important information for disease diagnosis (8). Since early studies, scientists have explored techniques for detecting glucose in tears (1), and to this day we have been able to accurately detect amyloid Aβin tears using small, non-invasive biosensors. Remarkably, the concentration of Aβ in tears is approximately 10 times higher than the concentration of Aβ in blood, making it an important indicator of early diagnosis of Alzheimer’s disease (93, 94). Song et al. explored the relationship between tears and blood cholesterol levels using a rabbit model of hyperlipidemia and successfully used contact lenses with electrochemical sensors to detect the feasibility of cholesterol-associated disorders in tears, a technique that not only provides a non-invasive diagnostic method but also has the potential to bring about a much more convenient healthcare experience for patients (95). In addition, the researchers found that the amount of urea in tears was comparable to that in blood, which provides a more effective way to test kidney function (8, 96). Also, the concentration of TNF-α in tears can be monitored by a rapid and simple assay, which is important for early screening and assessment of cancer (97, 98). The detection of cortisol has also been a hot topic in studies exploring tears as a biomarker; most cortisol is released from the adrenal glands where it binds to corticosteroid-binding globulin, but there is also an active form of free cortisol that occurs in biological fluids, including tears, a finding that provides a new way to monitor and assess the body’s stress response (99). Sempionatto et al. achieved the detection of vitamin C in tears by immobilizing ascorbic acid oxidase on a flexible printable tattoo electrode (100). This innovative technology not only provides a new means of health monitoring but also offers a deeper understanding of tears as a biomarker. Overall, smart wearable devices have great potential for future applications in medicine. By detecting various biomarkers in tears, these devices will be able to provide important information for the early diagnosis and treatment of diseases, thus protecting people’s health.

With the continuous development of science and technology, more and more researchers are focusing their attention on the combination of electrochemical biosensors with wireless smart devices for real-time visual monitoring (10). Kim et al. utilized enzyme-functionalized graphene-silver nanowire hybrid nanostructures to monitor glucose concentration in tears and intraocular pressure (IOP) through contact lenses with integrated glucose sensors, wireless power transmission circuits, and display pixels inside the lens (101). Later, researchers at Yonsei University in South Korea developed contact lenses capable of real-time cholesterol monitoring (95). In addition, significant progress has been made in research on contact lens-based biosensors for optical monitoring, Elsherif et al. developed an electrochemical sensor that produces a change in the volume of the hydrogel when the glucose concentration changes thereby changing the intensity of the diffracted light and synchronizes the information to a smartphone for monitoring (102). Ye et al., development of a contact lens for detection of matrix metalloproteinase-9 (MMP-9) and intraocular pressure in eye-related diseases by surface-enhanced Raman scattering and color change, respectively (13). Sempionatto et al. developed an electrochemical sensor that can simultaneously monitor the levels of three biomarkers, glucose, alcohol and vitamins, in tear (103). More and more researchers are committed to expanding the application scenarios of electrochemical sensors to the home healthcare industry, and these sensors will be closely connected with smartphones to provide patients with a full range of medical information and protection, which is undoubtedly research of great significance and is expected to change the way of life of people’s healthcare and make healthcare services more efficient and convenient.