Wound healing (WH) is a complex and dynamic process that usually occurs in four basic phases: haemostasis, inflammation, proliferation and remodelling. Wounds are classified as acute and chronic according to their healing time. Acute wounds usually heal within a certain period of time, whereas chronic wounds are lesions that do not close for more than six weeks, often including examples such as diabetic foot ulcers, pressure sores and venous ulcers. Chronic wounds both reduce individual quality of life and create a significant economic burden on the health system². The main challenges encountered in wound treatment include the risk of infection, inadequate blood flow, nutritional disorders and chronic diseases such as diabetes. Modern treatment approaches developed in this context include wound dressings that provide a moist environment, negative pressure wound therapy, hyperbaric oxygen therapy, growth factors, electrical stimulation, ultrasound, laser and larval therapy. Especially wound dressings stand out thanks to their ease of application and usability in different wound types⁷.

Innovative research has focused on biomaterial-based development of these products. The literature reveals a shift in the focus on biomaterials between the pre-2008 and post-2008 periods. In the pre-2008 period, cell-based tissue engineering and fibrin-based materials were prominent; after 2008, interest in antibacterial effect, growth factors, controlled release systems, and natural biopolymer-based hydrogel wound dressings increased. Especially marine-derived materials, such as fish collagen, have potential due to their biocompatible properties⁴.

The primary aim of this study was to identify the temporal trends in the literature on wound healing and biomaterial-based approaches and to analyze the role of marine biomaterials in wound healing. As a result of LDA-based analyses, the limited emphasis on marine biomaterials points to an important research gap in this field. Accordingly, this study aims to contextualize marine biomaterials (e.g., fish collagen) within the evolving wound-healing biomaterials literature, highlighting their relative underrepresentation and outlining translational research needs.

Büyütmek İçin Tıklayın

Figure 1: The methodological framework of the study

The preprocessing, analyses, and visualization were carried out with the high programming language Python. The LDA model was implemented using the Gensim library in Python. Separate models were created for each time to observe thematic changes over time. To address sensitivity to random initialization, we fixed the random seed for all final models to ensure reproducibility. In addition, we performed repeated runs using multiple random seeds and observed that the resulting topic structures were qualitatively stable (i.e., dominant topic labels and high-weight terms remained consistent across runs).

The optimal number of topics (K) for each period was selected by evaluating topic coherence across candidate K values and then confirming interpretability via manual inspection of the top terms and representative documents. Specifically, we evaluated K = [2-8] (step = 1) for each period and computed coherence using the UMass metric (Gensim CoherenceModel). The final K was selected as the solution that provided the best balance between coherence and semantic interpretability, avoiding redundant topic splitting at higher K values. We report the coherence-based selection procedure to improve reproducibility, while retaining expert-driven labeling for semantic clarity. Dominant keywords per topic were extracted, and each topic was manually labeled based on semantic interpretation.

For transparency, topic coherence was also calculated for the final models (UMass coherence): -1.25 for the pre-2008 model (K=3) and -1.23 for the 2008-2024 model (K=4), supporting the internal consistency of the selected topic solutions.

Datasets: This study is based on a corpus of 4316 peer-reviewed scientific articles retrieved from the Web of Science (WoS) Core Collection database. The search strategy included the keywords ?biomaterial? and ?wound healing? in article titles, abstracts, and keywords. The time period of interest ranged from the early 1990s to 2024, and publications were limited to English-language journal articles. The dataset was divided into two periods for comparative analysis: pre-2008 and 2008-2024. In this framework, the abstracts of these articles in WoS were analyzed with the LDA algorithm.

To enable an interpretable temporal comparison, we selected a single split year rather than multiple arbitrary intervals. Using annual publication counts from the retrieved WoS records, we examined whether the growth trajectory exhibited a structural change and applied a segmented (piecewise) regression framework with an unknown breakpoint. The estimated breakpoint occurred around 2008, aligning with the onset of sustained acceleration in publication volume. Accordingly, the corpus was divided into pre-2008 and 2008?2024 subsets, and separate LDA models were trained for each period⁸.

Preprocessing: Article abstracts were extracted and subjected to a text preprocessing pipeline. This included:

i. Lowercasing all text

ii. Removing stopwords and punctuation

iii. Tokenization

iv. Lemmatization (using spaCy and NLTK libraries)

v. Filtering words with low frequency and overly common generic terms (e.g., "study," "result")

Only the abstract text was used for topic modeling to maintain consistency and reduce noise from full-text variability.

Natural Language Processing: LDA Algorithm: Computational methods are used in NLP to learn, understand, and produce knowledge in human languages. Automating the study of the linguistic structure of language and building basic technologies such as machine translation, speech recognition, and speech synthesis were the main goals of early computational approaches to language research. Today, researchers are developing and applying these techniques in practical ways, developing speech-to-speech translation engines and spoken dialogue systems, searching social media for financial or health-related information, and detecting attitudes and sentiments towards goods and services⁹.

Social media's growth has completely changed the kinds and quantity of information that NLP researchers can now access. Studying the connections between social contact, language use, and demography is made possible by data from websites like Facebook, YouTube, Twitter, blogs, and discussion forums. Using web scraping tools, researchers can acquire previously unthinkable volumes and types of data, frequently via website application program interfaces. Through the use of statistical and machine learning methods, they are able to predict the spread of disease from symptoms or food-related illnesses mentioned in tweets, detect deception in fake reviews, identify social networks of people interacting together online, track trending topics and popular sentiments, identify opinions and beliefs about politicians and products, and identify demographic information (like age and gender) from language¹⁰.

NLP provides a thorough examination of methods for extracting features from text. One of the NLP approaches, LDA, may pinpoint the primary subjects in the text and extract detailed information about each one. To determine latent subjects that are likely to generate such words, LDA treats text as a collection of words. It makes the assumption that each document covers various subjects and that each word in the text has a connection to a certain subject. Finding these hidden topics and estimating topic proportions for every text are the objectives of the LDA algorithm¹¹. Figure 2 illustrates the LDA model:

Büyütmek İçin Tıklayın

Figure 2: LDA graphical model

Formula for the LDA model is as follows:

p(θ, z, w | α, β) = p(θ | α) × ∏ (n=1 to N) [ p(z_n | θ) × p(w_n | z_n, β) ]

In the model given in Figure 2, the parameter m represents the number of documents to be produced; α represents the topic distribution; z represents a topic; θ represents the topic distribution for a given document; and represents words. The figure reveals three levels of LDA representation. In the process of building a corpus, the parameters α and β are assumed to be sampled once. The parameter θ is sampled separately for each document¹².

LDA assumes that the corpus contains topics and that each document may be represented as a distribution that includes these topics. Simultaneously, LDA represents each topic as a distribution across all the languages in the corpus. To efficiently simulate these distributions, LDA makes use of Dirichlet priors.

The following succinctly describes the generating process:

For each unique document:

1. Choose a probability distribution from a Dirichlet distribution for various themes.

2. For every word that appears in the text:

a. From the previously selected distribution among several themes, pick a specific topic.

b. Pick a word from the list of terms related to the selected subject.

Finding out which latent topic distributions are responsible for producing the recorded observations is the main goal of LDA. This is accomplished by probabilistic inference, which frequently uses techniques like Gibbs sampling and variational inference. Following training, LDA can be used to assign topics to new documents and identify the most likely topics found in a particular document¹³.

A review of the studies published before 2008 using the LDA algorithm showed that the articles mainly focused on three main themes. These were classified as Cellular Interactions?Tissue Formation (33.8%), Collagen?Matrix-Based Regeneration (31.3%), and Surface Properties-Controlled Drug Release (34.8%). The extracted keywords were clustered under these thematic headings, indicating that research in this period predominantly focused on structural characteristics and their interplay with biological responses (Table 1).

Büyütmek İçin Tıklayın

Table 1: Identification of themes and keywords with LDA before 2008

The word cloud visualization of keywords from publications prior to 2008 is presented in Figure 3.

Büyütmek İçin Tıklayın

Figure 3: Word cloud visualization of frequently occurring words in publications prior to 2008. The visualization includes keywords identified in themes. The size of each term reflects its relative frequency in the literature before 2008. Note: Word clouds are provided for descriptive and illustrative visualization only and should not be interpreted as statistical evidence

Between 2008 and 2024, analysis of the article corpus revealed four predominant thematic clusters: Infection Control-Wound Healing (22.8%), Functional Hydrogels-Tissue Engineering (25.1%), Therapeutic-Controlled Release Systems (24.8%), and Mechanical Properties?Structural Functionality (27.3%). The keywords associated with the publications during this period were systematically grouped under these thematic categories (Table 2).

Büyütmek İçin Tıklayın

Table 2: Identification of themes and keywords with LDA between 2008-2024

The visualization of the words belonging to the studies between 2008-2024 with word cloud analysis is given in Figure 4.

Büyütmek İçin Tıklayın

Figure 4: Word cloud visualization of frequently occurring words in publications between 2008 and 2024. The visualization includes keywords identified in the themes. The size of each term reflects the relative frequency of the literature between the years indicated. Note: Word clouds are provided for descriptive and illustrative visualization only and should not be interpreted as statistical evidence.

In addition, the keywords were analyzed and presented graphically for both the studies before 2008 and those between 2008 and 2024 (Figure 5).

Büyütmek İçin Tıklayın

Figure 5: Percentage change graph of the words that were common in the studies before 2008 and between 2008 and 2024

For transparency, the token ?marine? appeared in 34 abstracts overall (0.79% of the corpus). When stratified by period, it was absent pre-2008 (0/297; 0%) and occurred in 34/4019 abstracts post-2008 (0.85%). Therefore, it was interpreted as a secondary, qualitative signal rather than a dominant topic-forming term. These percentage distributions are reported for exploratory description only; we did not compute confidence intervals, effect sizes, or perform hypothesis testing, and therefore the results should not be interpreted as statistically significant differences.

In support of this, the LDA analysis revealed the co-occurrence of the term ?collagen? with concepts such as "therapeutic" and "regenerative," reinforcing the thematic transition observed in the field. In particular, the theme labeled as ?Collagen-Matrix Based Regeneration,? where collagen is prominently represented, highlights the therapeutic use of collagen-based biomaterials. Within this theme, collagen-based biomaterials are prominently represented; marine-derived collagens are therefore discussed here as a literature-supported sub-trend within the broader collagen landscape rather than as a distinct LDA-driven topic^16,20. Furthermore, these biomaterials exhibit therapeutic potential in advanced applications, including controlled release systems and the design of biocompatible scaffolds^16,21. Marine-sourced biomaterials may offer practical advantages (e.g., acceptability and supply sustainability) and are increasingly discussed in recent reviews; however, in our corpus they remained a secondary signal and were not a dominant driver of the LDA topics. Therefore, we frame marine-derived collagen as an emerging, literature-supported sub-trend that warrants targeted future work (e.g., standardized extraction/quality control and long-term in vivo safety studies) before stronger corpus-level claims can be made.

Expanding on the broader thematic trends, the findings also reveal distinct differences in biomaterials research before and after 2008. Studies prior to 2008 primarily focused on structural parameters, particularly surface properties and controlled drug release. The theme ?Surface Properties-Controlled Drug Release? was dominant, representing 34.8% of that period's content. This focus reflected efforts to understand how physicochemical properties of biomaterials influenced cellular behavior. However, after 2008, a clear transition toward functional biomaterials became evident. During this newer period, themes such as ?Mechanical Properties?Structural Functionality? (27.3%) and ?Functional Hydrogels-Tissue Engineering? (25.1%) emerged, suggesting that biomaterials were increasingly being designed to perform biological roles, not just structural support. Among these, nanofiber-based wound dressings (particularly those produced via electrospinning) have shown promise in enhancing wound healing and minimizing scar formation²². Functional hydrogels have similarly become central, given their high-water content, biocompatibility, and versatility in regenerative medicine. The expanding use of such hydrogels, especially in personalized disease modeling, reflects a broader trend toward more active and regenerative roles for biomaterials²³.

This progression underscores a paradigmatic shift-from passive, structural materials to dynamic, bioactive systems. Our findings are consistent with this evolution, as reflected by the post-2008 prominence of topics centered on functional hydrogels, infection control, and regenerative tissue engineering in the LDA output. These advances pave the way for disease-specific, patient-tailored therapies with enhanced clinical efficacy, particularly for complex wound types such as diabetic foot ulcers, burns, and postoperative lesions.

As this thematic transformation unfolds, it is bringing biomaterials research closer to real-world clinical application. Innovations are now targeting specific medical challenges, including diabetic wounds, antimicrobial resistance, and host biocompatibility. Nanofiber and hydrogel-based biomaterials, in particular, are well positioned to meet these demands²⁴. Moreover, while wound healing did not appear as a clearly defined thematic category prior to 2008, the emergence of the ?Infection Control?Wound Healing? theme post-2008 (though relatively lower in prevalence) indicates a growing clinical relevance. This evolution suggests that earlier studies were largely confined to laboratory environments, focusing on in vitro experiments. Supporting this, literature from the early 2000s often concentrated on cellular responses to material surfaces rather than holistic biological contexts^25,26.

Clinically, the post-2008 emergence of ?Infection Control?Wound Healing? is plausibly linked to the rising emphasis on antimicrobial resistance and on local bioburden management in both chronic and postoperative wounds. This has accelerated interest in antimicrobial dressings and multifunctional biomaterials that combine barrier function with active antimicrobial or antibiofilm mechanisms. Examples include silver- and nanoparticle-containing systems, honey-based dressings and hydrogels, and plant-derived bioactives incorporated into films/hydrogels for antioxidant and antimicrobial effects. In this context, the infection-control signal captured by topic modeling can be interpreted as reflecting a broader convergence of biomaterials engineering with antimicrobial stewardship needs^27-29.

Importantly, part of this translational trajectory is reflected in veterinary wound management, where biomaterial dressings are frequently evaluated in clinically realistic contexts (contaminated wounds, large tissue loss, and challenging anatomic sites). In companion animals, biomaterial-based dressings and biologic grafts have been reported to support healing after wide excisions and complex soft-tissue defects. In equine surgery, distal limb wounds represent a well-known clinical challenge characterized by delayed granulation and higher complication risk, making them a relevant model for testing hydrogel- and antimicrobial-dressing strategies. We therefore interpret the veterinary-oriented subset as complementary evidence that helps bridge laboratory innovation with real-world wound environments^30-32.

From a field-utility perspective, veterinary patients provide clinically realistic wound environments (contamination, motion, distal limb ischemia, large tissue loss) where biomaterial dressings are stress-tested under conditions similar to complex human wounds. Marine-origin dressings are particularly relevant here: acellular fish-skin xenografts and marine collagen matrices have been increasingly reported in companion animal soft-tissue reconstruction and in equine distal limb wounds, which are known for delayed granulation and higher complication rates. These applications strengthen the translational interpretation of the post-2008 ?infection control? and ?functional hydrogel? themes by illustrating where and why clinicians adopt bioactive scaffolds and antimicrobial dressings outside controlled laboratory settings^33,34.

In contrast, the post-2008 period has witnessed increasing attention to in vivo relevance and translational potential. For example, the emergence of functional wound healing themes corresponds with rising interest in direct biomaterial-tissue interactions. This shift has opened doors to translational research, leading to more clinically grounded strategies (35, 36). Notably, the increase in in vivo research reflects a demand for biomaterials that perform reliably in dynamic, real-world environments. Developments in dual-function biomaterials (those combining antimicrobial and regenerative capabilities) have played a role in improving success rates in vivo^37,38.

Beyond simple term-frequency shifts, the post-2008 vocabulary increasingly reflects engineering-driven formulation strategies. The rise of ?-based? is best interpreted not as an informative keyword per se, but as a linguistic marker of more specific material descriptors (e.g., collagen-based, hydrogel-based). In parallel, the field has moved toward composite and hybrid scaffolds (particularly fiber-hydrogel and nanofiber-enhanced hydrogel composites) that aim to combine ECM-mimicking architecture with tunable mechanics, moisture management, and controlled delivery of therapeutics. This composite design paradigm provides a more mechanistically meaningful interpretation of the lexical shift than the standalone token ?based?^39-41.

In parallel with these developments, the literature increasingly explores marine-derived collagens as alternatives to mammalian sources, motivated by biocompatibility, sustainability, and ethical considerations. Traditionally, collagen is derived from mammals like pigs, cattle, and rodents. However, concerns related to zoonotic transmission and religious acceptability have prompted increased exploration of marine organisms (including fish, jellyfish, and sponges) as safer alternatives^21,42. Marine-sourced collagens, gelatins, and chitosan offer compelling advantages, including lower immunogenicity, ethical acceptability, and potential for sustainable production^15,43-45.

A further driver of the shift toward non-mammalian collagen sources is biosafety. Mammalian-derived collagens (bovine/porcine) have historically raised concerns regarding zoonotic contamination and, although tightly regulated, a theoretical risk related to transmissible spongiform encephalopathies (e.g., prion/BSE) is frequently cited in the biomaterials safety literature. Marine-derived collagen avoids several mammalian-specific pathogen concerns and can offer advantages in acceptability and supply sustainability; nevertheless, marine sources introduce their own safety-control requirements (e.g., heavy metals, endotoxin burden, batch variability), making standardized extraction and quality-control pipelines essential for translational adoption^46,47.

These considerations are presented here as contextual evidence from the broader literature rather than as a direct output of our LDA results. While marine-derived materials are discussed in reviews in relation to sustainability, acceptability, and biosafety, our corpus-level signal remained low (i.e., a secondary, qualitative signal rather than a dominant topic driver). Accordingly, we interpret ethical/ecological arguments and quality-control challenges (batch variability, interspecies differences, potential contaminants such as heavy metals and endotoxin, and standardization needs) as literature-informed translational considerations that warrant targeted future work, rather than as findings emerging from the topic model itself.

A methodological limitation relates to the search query. By using ?biomaterial? AND ?wound healing? as required terms, the corpus may undercapture studies that primarily use specific material family labels (e.g., hydrogel, scaffold, nanofiber, electrospun mat) without explicitly using the umbrella term ?biomaterial?. While this choice improved precision and reproducibility, it may reduce recall. Future work could incorporate an expanded query set (e.g., adding major material-family keywords) and compare topic stability via sensitivity analyses.

Additional limitations include reliance on a single database (WoS Core Collection), English-language records, and abstract-only modeling, which may omit full-text nuance and non-indexed literature. In addition, the pre-2008 and post-2008 subsets were markedly unequal in size (297 vs 4,019), which may influence estimated topic prevalence; therefore, temporal contrasts are reported descriptively. As an unsupervised approach, LDA provides thematic structure rather than causal inference; therefore, findings should be interpreted as descriptive trends.

When examined comprehensively, marine-sourced biomaterials clearly present compelling advantages. Their inherent antibacterial properties, biodegradability, environmental sustainability, and compatibility with cellular systems make them particularly well-suited for next-generation wound care and regenerative solutions⁴⁸. These considerations are discussed here as literature-based translational context; however, consistent with the low corpus-level frequency of marinerelated terms, marine collagens should be interpreted as an emerging, literature-supported sub-trend rather than a dominant LDA-driven finding.

This study provides a comprehensive overview of the evolving research landscape in wound healing and highlights the growing importance of marine-derived collagen as a sustainable, biocompatible, and immunologically safe alternative to traditional biomaterials. Topic modeling of over two decades of literature revealed a clear thematic shift from structure- focused and in vitro studies toward functionality-driven applications such as controlled drug delivery, infection prevention, and tissue regeneration. Based on these trends, future research is expected to emphasize the development of biomaterials with enhanced mechanical and functional performance, integrated into smart systems that actively support the healing process. Marine collagen-based hydrogels and scaffolds, when combined with emerging technologies like 3D bioprinting and nanotechnology, may offer promising pathways toward personalized and translational wound therapies.

1) Chen X, Xie H, Tao X, et al. Artificial intelligence and multimodal data fusion for smart healthcare: Topic modeling and bibliometrics. Artif Intell Rev 2024; 57: 91.

2) Toffaha KM, Simsekler MCE, Omar MA. Leveraging artificial intelligence and decision support systems in hospital-acquired pressure injuries prediction: A comprehensive review. Artif Intell Med 2023; 141: 102560.

3) Tran BX, Latkin CA, Vu GT, et al. The current research landscape of the application of artificial intelligence in managing cerebrovascular and heart diseases: A bibliometric and content analysis. Int J Environ Res Public Health 2019; 16: 2699.

4) Alhashmi SM, Hashem IAT, Al-Qudah I. Artificial intelligence applications in healthcare: a bibliometric and topic model-based analysis. Intelligent Syst Appl 2024; 21: 200299.

5) Veredas F, Mesa H, Morente L. Binary tissue classification on wound images with neural networks and bayesian classifiers. IEEE Trans Med Imaging 2010; 29: 410-427.

6) Kolli A, Wei Q, Ramsey SA. Predicting time-to-healing from a digital wound image: A hybrid neural network and decision tree approach improves performance. Computation 2024; 12: 42.

7) D?Ascenzo F, Rocchi A, Iandolo F, Vito P. Evolutionary impacts of artificial intelligence in healthcare managerial literature. A ten-year bibliometric and topic modeling review. Sustain Futures 2024; 7: 100198.

8) Muggeo VM. Estimating regression models with unknown break?points. Stat Med 2003; 22: 3055-3071.

9) Hirschberg J, Manning CD. Advances in natural language processing. Science 2015; 349: 261-266.

10) Russell MA. Mining the Social Web. 2nd Edition, Sebastopol (CA): O?Reilly Media; 2013.

11) Jagwani V, Meghani S, Pai K, Dhage S. Resume evaluation through latent dirichlet allocation and natural language processing for effective candidate selection. ArXiv Preprint, 2023: arXiv:2307.15752.

12) Blei DM, Ng AY, Jordan MI. Latent dirichlet allocation. J Mach Learn Res 2003; 3: 993-1022.

13) Vayansky I, Kumar SA. A review of topic modeling methods. Inf Syst 2020; 94: 101582.

14) Lee CH, Singla A, Lee Y. Biomedical applications of collagen. Int J Pharm 2001; 221: 1-22.

15) Silva TH, Moreira-Silva J, Marques AL, et al. Marine origin collagens and its potential applications. Mar Drugs 2014; 12: 5881-5901.

16) Jafari H, Lista A, Siekapen MM, et al. Fish collagen: extraction, characterization, and applications for biomaterials engineering. Polymers (Basel) 2020; 12: 2230.

17) Geahchan S, Baharlouei P, Rahman A. Marine collagen: A promising biomaterial for wound healing, skin anti-aging, and bone regeneration. Marine Drugs 2022; 20: 61.

18) Xu N, Peng XL, Li HR, et al. Marine-derived collagen as biomaterials for human health. Front Nutr 2021; 8: 702108.

19) Hoyer B, Bernhardt A, Heinemann S, et al. Biomimetically mineralized salmon collagen scaffolds for application in bone tissue engineering. Biomacromolecules 2012; 13: 1059-1066.

20) Ahmed Z, Powell LC, Matin N, et al. Jellyfish collagen: a biocompatible collagen source for 3d scaffold fabrication and enhanced chondrogenicity. Mar Drugs 2021; 19: 405.

21) Lim YS, Ok YJ, Hwang SY, et al. Marine collagen as a promising biomaterial for biomedical applications. Mar Drugs 2019; 17: 467.

22) Liu X, Xu H, Zhang M, Yu DG. Electrospun medicated nanofibers for wound healing. Membranes 2021; 11: 770.

23) Chaudhary S, Chakraborty E. Hydrogel based tissue engineering and its future applications in personalized disease modeling and regenerative therapy. Beni-Suef Univ J Basic Appl Sci 2022; 11: 3.

24) Gao C, Zhang L, Wang J, et al. Electrospun nanofibers promote wound healing. J Mater Chem B 2021; 9: 3106-3130.

25) Anderson JM. Inflammation, wound healing, and the foreign body response. In: Ratner BD, Hoffman AS, Schoen FJ, Lemons JE (Editors). Biomaterials Science: An Introduction to Materials in Medicine. 2nd Edition, San Diego: Elsevier Academic Press, 2004: 165-172.

26) Place ES, George JH, Williams CK, Stevens MM. Synthetic polymer scaffolds for tissue engineering. Chem Soc Rev 2009; 38(4): 1139-1151.

27) Yousefian F, Hesari R, Jensen T, et al. Antimicrobial wound dressings: A concise review for clinicians. Antibiotics 2023; 12: 1434.

28) Tashkandi H. Honey in wound healing: An updated review. Open Life Sciences 2021; 16: 1091-1100.

29) Masimen MAA, Harun NA, Maulidiani M, Ismail WIW. Overcoming methicillin-resistance Staphylococcus aureus (MRSA) using antimicrobial peptides-silver nanoparticles. Antibiotics 2022; 11: 951.

30) Naghi R, Maxwell EA, Carrillo AJ, et al. Acellular fish skin may be used to facilitate wound healing following wide surgical tumor excision in dogs: A prospective case series. J Am Vet Med Assoc 2023; 261: 1547-1554.

31) Lan Z, Guo L, Fletcher A, et al. Antimicrobial hydrogel foam dressing with controlled release of gallium maltolate for infection control in chronic wounds. Bioact Mater 2024; 42: 433-448.

32) Helal IE, Al?Abbadi HA, El?Daharawy MH, Ahmed MF. Enhancement of chronic wound healing with maltodextrin/ascorbic acid gel: A clinical evaluation of distal limb wounds in horses. J Anim Sci Technol 2022; 64: 997.

33) Teshima R, Osawa S, Yoshikawa M, et al. Low-adhesion and low-swelling hydrogel based on alginate and carbonated water to prevent temporary dilation of wound sites. Int J Biol Macromol 2024; 254: 127928.

34) Anantama NA, Du Cheyne C, Martens A, et al. The granulation (t) issue: A narrative and scoping review of basic and clinical research of the equine distal limb exuberant wound healing disorder. Vet J 2022; 280: 105790.

35) Hoang Thi TT, Lee Y, Le Thi P, Park KD. Oxidized alginate supplemented gelatin hydrogels for the in situ formation of wound dressing with high antibacterial activity. Macromol Res 2019; 27: 811-820.

36) Slaughter BV, Khurshid SS, Fisher OZ, Khademhosseini A, Peppas NA. Hydrogels in regenerative medicine. Adv Mater 2009; 21: 3307-3329.

37) Paul S, Schrobback K, Tran PA, et al. Photo-cross-linkable, injectable, and highly adhesive gelma-glycol chitosan hydrogels for cartilage repair. Adv Healthc Mater 2023; 12: e2302078.

38) Guo Y, Jiang Y, Zhang Y, Huang L, Tang K. Highly stretchable and adhesive hydrogel dressing possessing rapid hemostasis and photothermal antibacterial properties for joint wound healing. Chem Eng J 2025; 519: 164834.

39) Teixeira MO, Antunes JC, Felgueiras HP. Recent advances in fiber?hydrogel composites for wound healing and drug delivery systems. Antibiotics 2021; 10: 248.

40) Meng Q, Li Y, Wang Q, et al. Recent advances of electrospun nanofiber-enhanced hydrogel composite scaffolds in tissue engineering. J Manuf Process 2024; 123: 112-127.

41) Devi VKA, Shyam R, Palaniappan A, et al. Self-healing hydrogels: preparation, mechanism and advancement in biomedical applications. Polymers (Basel) 2021; 13: 3782.

42) Subhan F, Ikram M, Shehzad A, et al. Marine collagen: An emerging player in biomedical applications. J Food Sci Technol 2015; 52: 4703-4707.

43) Gómez-Guillén MC, Giménez B, López-Caballero ME, Montero MP. Functional and bioactive properties of collagen and gelatin from alternative sources: A review. Food Hydrocoll 2011; 25: 1813-1827.

44) Azuma K, Izumi R, Osaki T, et al. Chitin, chitosan, and its derivatives for wound healing: Old and new materials. J Funct Biomater 2015; 6: 104-142.

45) Lopes AI, Pintado MM, Tavaria FK. Plant-based films and hydrogels for wound healing. Microorganisms 2024; 12: 438.

46) Davison-Kotler E, Marshall WS, García-Gareta E. Sources of collagen for biomaterials in skin wound healing. Bioengineering 2019; 6: 56.

47) Wosicka-Fr?ckowiak H, Poniedzia?ek K, Wo?ny S, et al. Collagen and its derivatives serving biomedical purposes: a review. Polymers (Basel) 2024; 16: 2668.

48) Zhao X, Lang Q, Yildirimer L, et al. Photocrosslinkable gelatin hydrogel for epidermal tissue engineering. Adv Healthc Mater 2016; 5: 108-118.