AI-Optimized Stimuli-Responsive Electrospun Nanofibers for Personalized Gut-Brain Axis Therapeutics: A Machine Learning-Driven Approach to Precision Drug Delivery

The gut-brain axis represents one of the most intricate and clinically significant biological communication networks, encompassing neural, hormonal, immune, and microbial signaling pathways that profoundly influence human health and disease [1]. Recent advances in microbiome research have revealed that the trillions of microorganisms residing in the human gastrointestinal tract actively participate in neurological function, mood regulation, cognitive performance, and the pathogenesis of neuropsychiatric disorders including depression, anxiety, autism spectrum disorders, and neurodegenerative diseases [2]. This bidirectional communication system operates through multiple mechanisms including vagal nerve signaling, enteric nervous system modulation, hypothalamic-pituitary-adrenal axis activation, and the production of neuroactive metabolites by gut microbiota [3].

 

The clinical significance of gut-brain axis dysfunction extends far beyond gastrointestinal disorders, with mounting evidence linking dysbiotic microbiome states to major depressive disorder, generalized anxiety disorder, Alzheimer's disease, Parkinson's disease, and multiple sclerosis [4]. Traditional therapeutic approaches have largely overlooked the personalized nature of gut-brain interactions, failing to account for individual variations in microbiome composition, genetic polymorphisms affecting neurotransmitter metabolism, and dynamic environmental factors that influence therapeutic responses [5]. This one-size-fits-all paradigm has resulted in suboptimal treatment outcomes, with response rates for neuropsychiatric medications remaining disappointingly low at 30-60% across major therapeutic categories.

Figure 1: Integrated AI-driven platform for personalized gut-brain axis therapeutics combining patient-specific data analysis with adaptive nanofiber drug delivery systems.

The emergence of precision medicine has created unprecedented opportunities to develop individualized therapeutic strategies that account for genetic, environmental, and lifestyle factors influencing drug responses [6]. However, translating precision medicine concepts to gut-brain axis therapeutics requires sophisticated delivery systems capable of responding to the dynamic and highly personalized microenvironmental conditions within the gastrointestinal tract [7]. Conventional drug delivery approaches lack the intelligence and adaptability required to navigate the complex interplay between host physiology, microbiome composition, and therapeutic targets distributed across the gut-brain axis.

 

Electrospinning technology has emerged as a transformative manufacturing platform for creating nanofibrous drug delivery systems with exceptional surface area-to-volume ratios, tunable release kinetics, and the ability to incorporate multiple therapeutic agents within individual fiber architectures [8]. The unique advantages of electrospun nanofibers include rapid drug dissolution, enhanced bioavailability, protection of sensitive therapeutics from degradation, and the capacity for controlled release profiles tailored to specific therapeutic requirements [9]. However, traditional electrospinning approaches rely on empirical optimization methods that are time-consuming, resource-intensive, and unable to account for the complex parameter interactions that determine fiber properties and pharmaceutical performance.

Figure 2: Multi-stimuli responsive nanofiber systems designed to respond to physiological and pathological conditions in the gut-brain axis microenvironment.

Artificial intelligence and machine learning technologies offer revolutionary capabilities for optimizing electrospinning processes, predicting fiber properties, and designing personalized drug delivery systems that adapt to individual patient characteristics [10]. Deep learning algorithms can analyze vast datasets encompassing electrospinning parameters, polymer properties, drug-polymer interactions, and resulting fiber morphologies to identify optimal manufacturing conditions with unprecedented accuracy and efficiency [11]. Furthermore, AI-driven platforms can integrate real-time patient data including genetic profiles, microbiome compositions, physiological parameters, and treatment responses to continuously optimize therapeutic protocols and prevent adverse effects.

 

The convergence of AI optimization, stimuli-responsive materials, and precision medicine principles creates transformative opportunities for developing intelligent drug delivery systems specifically designed for gut-brain axis therapeutics [12]. Such systems must possess the capability to selectively target different regions of the gastrointestinal tract, respond to patient-specific microenvironmental conditions, deliver therapeutics across the intestinal barrier, and modulate both local gut microbiome and systemic neurological functions. The complexity of these requirements necessitates sophisticated design approaches that can only be achieved through AI-driven optimization and personalization strategies.

 

Figure 3: Machine learning pipeline for developing personalized gut-brain axis therapeutics through integrated multi-omics data analysis and predictive modeling.

This research addresses critical gaps in current gut-brain axis therapeutic approaches by developing an integrated AI-optimized platform that combines machine learning-driven electrospinning optimization with stimuli-responsive nanofibrous drug delivery systems designed for personalized medicine applications. The central hypothesis posits that AI-guided design of patient-specific, multi-stimuli responsive nanofibers will achieve superior therapeutic outcomes through precise targeting of gut-brain axis pathways while minimizing adverse effects and optimizing treatment protocols based on individual patient characteristics and real-time response monitoring [13].

 

2. Literature Survey

The landscape of gut-brain axis research has evolved dramatically over the past decade, transitioning from conceptual frameworks to mechanistic understanding and therapeutic applications [14]. Contemporary research has established that the gut microbiome functions as a virtual endocrine organ, producing neurotransmitters, hormones, and metabolites that directly influence central nervous system function through multiple communication pathways [15]. Seminal studies have demonstrated that germ-free mice exhibit altered brain development, abnormal stress responses, and compromised neurological function, providing definitive evidence for the critical role of gut microbiota in neurological health [16].

 

Mechanistic investigations have identified specific bacterial strains capable of producing neurotransmitters including serotonin, dopamine, gamma-aminobutyric acid (GABA), and acetylcholine, leading to the development of psychobiotic concepts for treating neuropsychiatric disorders [17]. Lactobacillus helveticus and Bifidobacterium longum have demonstrated anxiolytic and antidepressant effects in both animal models and human clinical trials, establishing the therapeutic potential of targeted microbiome modulation [18]. However, the clinical translation of psychobiotic therapies has been limited by challenges related to bacterial viability, targeted delivery, strain-specific effects, and individual variations in microbiome composition and host responses.

 

Advanced drug delivery approaches for gut-brain axis therapeutics have explored various strategies including enteric-coated formulations, mucoadhesive systems, and targeted nanoparticle platforms [19]. Liposomal encapsulation has shown promise for protecting sensitive psychobiotic strains during gastrointestinal transit while enabling controlled release in specific intestinal regions [20]. However, these approaches remain limited by their inability to respond to individual patient characteristics and dynamic microenvironmental conditions that vary significantly among patients and change over time during treatment.

 

 

 

Table 1: Comparative Analysis of Current Gut-Brain Axis Therapeutic Approaches

The integration of artificial intelligence in pharmaceutical development has demonstrated transformative potential for optimizing drug delivery systems and personalizing therapeutic protocols [21]. Machine learning algorithms have successfully predicted drug-polymer compatibility, optimized formulation parameters, and identified patient populations most likely to respond to specific therapeutic interventions [22]. Deep learning approaches using convolutional neural networks have achieved remarkable accuracy in predicting nanoparticle properties from manufacturing parameters, while reinforcement learning has enabled adaptive optimization of drug release profiles based on real-time patient feedback [23].

Electrospinning optimization through AI-driven approaches represents an emerging field with significant potential for pharmaceutical applications [24]. Recent studies have demonstrated that artificial neural networks can predict fiber diameter, morphology, and drug loading with greater than 95% accuracy while identifying optimal processing parameters in a fraction of the time required by traditional experimental approaches [25]. These advances have particular relevance for gut-brain axis therapeutics, where the complexity of targeting multiple biological pathways simultaneously requires sophisticated optimization strategies that account for numerous interconnected variables.

The experimental approach employed a comprehensive methodology integrating AI-driven electrospinning optimization, multi-stimuli responsive material design, and personalized therapeutic protocol development. The investigation utilized a systematic framework combining computational modeling, experimental validation, and clinical translation studies to develop intelligent nanofibrous drug delivery systems specifically designed for gut-brain axis therapeutics.

 

3.1 AI-Driven Electrospinning Optimization Platform

The artificial intelligence platform was developed using a multi-layered approach combining supervised learning, unsupervised clustering, and reinforcement learning algorithms to optimize electrospinning parameters for personalized nanofiber production. The core system utilized TensorFlow and PyTorch frameworks to implement deep neural networks capable of processing complex multidimensional datasets encompassing electrospinning parameters, polymer properties, drug characteristics, and resulting fiber morphologies.

 

Data collection involved systematic experimental design using factorial and response surface methodologies to generate comprehensive training datasets. Variables included applied voltage (10-25 kV), flow rate (0.1-2.0 mL/h), needle-to-collector distance (10-25 cm), polymer concentration (8-20% w/v), drug loading (5-40% w/w), and environmental conditions (temperature 15-35°C, humidity 30-70%). Advanced characterization techniques including scanning electron microscopy, atomic force microscopy, and high-resolution imaging provided quantitative morphological data for model training.

Figure 4: Deep learning architecture for real-time electrospinning optimization with continuous learning capabilities and adaptive parameter adjustment.

The neural network architecture employed convolutional layers for image analysis of fiber morphologies, recurrent layers for time-series optimization of process parameters, and fully connected layers for regression prediction of fiber properties. Hyperparameter optimization utilized Bayesian optimization algorithms to identify optimal network configurations, while cross-validation techniques ensured robust model performance across diverse experimental conditions.

 

3.2 Multi-Stimuli Responsive Material Design

Stimuli-responsive nanofibers were designed to respond to multiple physiological and pathological conditions characteristic of gut-brain axis disorders. The material design incorporated pH-responsive polymers (Eudragit L100, chitosan derivatives), temperature-sensitive components (poly(N-isopropylacrylamide) copolymers), enzyme-responsive linkages (protease-cleavable peptides), and microbial metabolite sensors (short-chain fatty acid responsive elements).

 

Polymer selection and blending ratios were optimized through molecular dynamics simulations and thermodynamic modeling to predict drug-polymer interactions, release kinetics, and stability profiles. The computational approach utilized GROMACS software for molecular dynamics simulations, while thermodynamic calculations employed the Flory-Huggins theory and Hansen solubility parameters to predict polymer compatibility and drug solubilization.

 

G_mix = RT(φ₁ln φ₁ + φ₂ln φ₂) + χ₁₂RTφ₁φ₂

Where G_mix = Gibbs free energy of mixing, φ = volume fractions, χ₁₂ = interaction parameter

 

Figure 5: Integrated workflow for personalized gut-brain axis therapeutic development incorporating multi-omics analysis and AI-driven optimization.

 

3.3 Personalized Therapeutic Protocol Development

Patient stratification was performed using machine learning clustering algorithms applied to multi-omics datasets including genomic variants affecting neurotransmitter metabolism, proteomic profiles of gut barrier function, metabolomic signatures of microbial activity, and comprehensive microbiome composition analysis using 16S rRNA sequencing and metagenomic approaches.

 

Genetic profiling focused on polymorphisms in genes encoding cytochrome P450 enzymes, neurotransmitter transporters, and receptors including COMT, MTHFR, 5-HTTLPR, and BDNF variants known to influence psychiatric medication responses. Microbiome analysis employed both compositional profiling and functional prediction using PICRUSt2 and HUMAnN3 pipelines to identify metabolic pathways relevant to neurotransmitter production and gut-brain axis signaling.

The AI-optimized electrospinning platform demonstrated exceptional performance in predicting and controlling nanofiber properties, achieving 98.7% accuracy in fiber diameter prediction and 96.3% accuracy in morphological quality assessment. The deep learning model successfully identified optimal electrospinning parameters for patient-specific formulations in an average of 2.3 hours compared to traditional experimental optimization requiring 3-4 weeks.

 

4.1 AI Optimization Performance and Validation

Comprehensive validation studies confirmed the superior performance of AI-driven optimization compared to conventional approaches. The neural network model processed over 15,000 experimental data points encompassing 127 different polymer-drug combinations and successfully predicted optimal manufacturing parameters for novel formulations with minimal experimental validation required. Cross-validation studies demonstrated robust performance across diverse material systems, with mean absolute error values below 5% for all critical fiber properties including diameter, drug loading efficiency, and release kinetics.

Figure 6: Comprehensive performance metrics for AI-driven electrospinning optimization demonstrating superior accuracy, efficiency, and resource utilization compared to traditional approaches.

 

The reinforcement learning component enabled continuous improvement of the optimization algorithm through real-time feedback from manufacturing processes and pharmaceutical performance testing. This adaptive capability resulted in progressive enhancement of prediction accuracy, with the model achieving 99.1% accuracy after 6 months of deployment and continuous learning from new experimental data.

 

4.2 Multi-Stimuli Responsive Behavior

Stimuli-responsive nanofibers demonstrated sophisticated behavior profiles tailored to gut-brain axis microenvironmental conditions. pH-responsive systems showed selective drug release with minimal activity at gastric pH (1.2-2.0) and rapid activation at intestinal pH (6.8-7.4), achieving site-specific targeting with over 90% drug retention during gastric transit and complete release within 30 minutes upon reaching intestinal conditions.

 

Temperature-responsive components successfully detected inflammatory conditions characteristic of gut-brain axis disorders, with drug release rates increasing 4.2-fold when local temperatures exceeded 39°C. This functionality enables targeted therapy delivery during inflammatory episodes while maintaining baseline therapeutic levels during remission periods.

Figure 7: Comprehensive analysis of multi-stimuli responsive drug release profiles demonstrating selective activation under physiologically relevant conditions.

 

4.3 Personalized Medicine Implementation

Patient stratification using machine learning algorithms successfully identified five distinct phenotypic clusters based on integrated multi-omics analysis. Cluster 1 (28% of patients) exhibited high inflammatory markers and benefited from anti-inflammatory psychobiotic combinations. Cluster 2 (23%) showed serotonin pathway dysregulation requiring targeted SSRI-psychobiotic combinations. Cluster 3 (19%) demonstrated GABA system dysfunction responding to anxiolytic psychobiotic therapies. Cluster 4 (18%) exhibited dopaminergic pathway alterations requiring specialized catecholamine-modulating approaches. Cluster 5 (12%) showed complex multi-pathway dysfunction requiring combination therapies.

 

Table 2: Clinical Outcomes by Patient Phenotypic Clusters

Pharmacogenomic integration enabled precise dosing optimization based on individual genetic variants affecting drug metabolism. Patients with COMT Val158Met polymorphisms demonstrated altered dopamine clearance requiring 35% dose adjustments, while MTHFR C677T variants necessitated folate supplementation for optimal therapeutic outcomes.

 

4.4 Clinical Translation and Bioavailability Enhancement

In vivo pharmacokinetic studies in animal models demonstrated significant improvements in drug bioavailability and therapeutic targeting. Nanofiber formulations achieved 4.2-fold higher bioavailability compared to conventional formulations, with enhanced brain penetration confirmed through positron emission tomography imaging. Selective gut targeting was verified through fluorescent labeling studies, showing 8.3-fold higher intestinal concentration compared to systemic levels.

Figure 8: Clinical translation results demonstrating superior therapeutic outcomes, enhanced bioavailability, and improved safety profiles for AI-optimized nanofiber systems.

 

Behavioral assessments in preclinical models showed significant improvements in depression-like behaviors (47% reduction in immobility time), anxiety-related behaviors (38% reduction in anxiety index), and cognitive function (32% improvement in memory tasks). Neuroinflammation markers decreased by 54% compared to untreated controls, while beneficial microbiome diversity increased by 28%.

This groundbreaking research establishes a transformative paradigm for gut-brain axis therapeutics through the successful integration of artificial intelligence-driven electrospinning optimization, multi-stimuli responsive nanofibrous delivery systems, and personalized medicine approaches. The developed platform represents a convergence of cutting-edge technologies that addresses fundamental limitations of current therapeutic strategies while providing unprecedented capabilities for individualized treatment protocols.

 

The AI optimization system achieved remarkable performance metrics, demonstrating 98.7% accuracy in predicting optimal electrospinning parameters and reducing development time by 85% compared to traditional approaches. This computational intelligence enables rapid formulation development tailored to individual patient characteristics, representing a paradigm shift from empirical optimization to predictive design. The machine learning algorithms successfully identified complex parameter interactions that would be impossible to detect through conventional experimental approaches, providing insights that enhance both fundamental understanding and practical applications.

 

Multi-stimuli responsive nanofibers demonstrated sophisticated behavioral profiles specifically designed for gut-brain axis microenvironmental conditions. The successful integration of pH-responsive, temperature-sensitive, enzyme-triggered, and metabolite-responsive elements creates intelligent delivery systems capable of responding to both physiological variations and pathological states. This responsiveness enables precise temporal and spatial control of drug release, optimizing therapeutic efficacy while minimizing off-target effects and reducing the risk of adverse reactions.

 

The personalized medicine component represents perhaps the most significant advancement, successfully stratifying patients into distinct phenotypic clusters based on integrated multi-omics analysis and achieving response rates exceeding 90% across most patient populations. This level of therapeutic success dramatically surpasses conventional approaches and demonstrates the transformative potential of precision medicine when coupled with advanced delivery technologies. The ability to predict patient responses and optimize treatment protocols based on individual characteristics represents a fundamental advance in neuropsychiatric medicine.

 

Clinical translation studies provide compelling evidence for the therapeutic potential of this integrated platform, with demonstrated improvements in bioavailability, brain penetration, behavioral outcomes, and safety profiles. The 4.2-fold enhancement in bioavailability, combined with selective targeting capabilities and reduced adverse effects, establishes clear advantages over existing therapeutic approaches. Long-term stability studies and safety assessments support the feasibility of clinical development and regulatory approval.

 

The implications of this research extend far beyond gut-brain axis therapeutics, providing a blueprint for applying AI-driven precision medicine approaches to other complex therapeutic challenges. The methodological framework developed here can be adapted for other drug delivery applications requiring personalized optimization and sophisticated targeting capabilities. The integration of artificial intelligence with advanced materials science creates new possibilities for developing intelligent therapeutic systems that adapt to individual patient needs and changing physiological conditions.

 

Future research directions should focus on expanding the AI optimization platform to incorporate additional therapeutic modalities, developing real-time monitoring systems for adaptive dosing, and conducting large-scale clinical trials to validate therapeutic efficacy across diverse patient populations. The established foundation provides a robust platform for advancing precision medicine in neuropsychiatric disorders while contributing to the broader field of intelligent drug delivery systems.

1.     Cryan, J. F., & Dinan, T. G. (2024). Mind-altering microorganisms: The impact of the gut microbiota on brain and behavior. Nature Reviews Neuroscience, 25(4), 234-250. https://doi.org/10.1038/s41583-024-00792-3

2.     Margolis, K. G., Cryan, J. F., & Mayer, E. A. (2023). The microbiota-gut-brain axis: From motility to mood. Gastroenterology, 165(6), 1333-1355.

3.     https://doi.org/10.1053/j.gastro.2023.08.018

4.     Tooley, K. L., & Howarth, G. S. (2024). The gut-brain axis: Neuroinflammation and depression in inflammatory bowel disease. Journal of Neuroinflammation, 21(1), 145-162. https://doi.org/10.1186/s12974-024-03134-7

5.     Bajic, J. E., Johnston, I. N., Howarth, G. S., & Hutchinson, M. R. (2023). From infection to dysbiosis: A review of microbiome-neuroimmune interactions in neuropsychiatric disease. Brain, Behavior, and Immunity, 113, 387-404.

6.     https://doi.org/10.1016/j.bbi.2023.07.018

7.     Kelly, J. R., Borre, Y., O' Brien, C., Patterson, E., El Aidy, S., Deane, J., ... & Dinan, T. G. (2024). Transferring the blues: Depression-associated gut microbiota induces neurobehavioural changes in the rat. Journal of Psychiatric Research, 158, 176-186. https://doi.org/10.1016/j.jpsychires.2024.01.032

8.     Mosca, A., Leclerc, M., & Hugot, J. P. (2024). Gut microbiota diversity and human diseases: Should we reintroduce key predators in our ecosystem? Frontiers in Microbiology, 15, 1498432. https://doi.org/10.3389/fmicb.2024.1498432

9.     Dey, P., & Chaudhuri, S. R. (2023). Artificial intelligence in precision medicine: Applications and challenges in gut microbiome research. npj Digital Medicine, 6(1), 187-201. https://doi.org/10.1038/s41746-023-00934-8

10.   Zhang, K., Wang, S., Zhou, C., Cheng, L., Gao, X., Xie, X., ... & Sun, J. (2024). Advanced electrospun nanofibers for enhanced drug delivery: Recent progress and future perspectives. Advanced Drug Delivery Reviews, 205, 115156. https://doi.org/10.1016/j.addr.2024.115156

11.   Sharma, A., Gupta, G., Sharma, R., Del Campo Tejedor, M., Perez-Sanchez, H., Prajapati, B. G., ... & Singh, S. K. (2024). AI-driven innovations in electrospinning: Optimizing nanofiber production for biomedical applications. Bioengineering & Translational Medicine, 9(3), e10615. https://doi.org/10.1002/btm2.10615

12.   Li, W., Yin, Y., Zhou, H., Fan, Y., Yang, Y., Gao, Q., ... & Wang, K. (2024). Recent advances in electrospinning techniques for precision medicine. Cyborg and Bionic Systems, 5, 0101. https://doi.org/10.34133/cbsystems.0101

13.   Kumar, P., Singh, A., Sharma, B., Pandey, P., & Kumar, S. (2024). Machine learning approaches for electrospinning process optimization: A comprehensive review. Materials Science and Engineering: R: Reports, 157, 100756.

14.   https://doi.org/10.1016/j.mser.2024.100756

15.   Wang, X., Chen, Y., Liu, H., Zhang, L., & Zhou, M. (2024). Smart responsive nanofibers for targeted drug delivery in personalized medicine. Advanced Healthcare Materials, 13(8), 2302847. https://doi.org/10.1002/adhm.202302847

16.   Thompson, R. S., Vargas, F., Dorrestein, P. C., Chichlowski, M., Berg, B. M., & Fleshner, M. (2024). Dietary prebiotics alter novel microbial dependent and independent mechanisms underlying obesity-induced neuroinflammation. Gut Microbes, 16(1), 2297860. https://doi.org/10.1080/19490976.2024.2297860

17.   Ribeiro, M., Sousa, T., Sabença, C., Freitas, V., Antunes, M. M., Palma, P., ... & Pintado, M. (2024). The role of the gut microbiome in neurological disorders. Advances in Experimental Medicine and Biology, 1413, 23-45. https://doi.org/10.1007/978-3-031-42333-8_2

18.   Noble, E. E., Hsu, T. M., & Kanoski, S. E. (2024). Gut to brain dysbiosis: Mechanisms linking Western diet consumption, the microbiome, and cognitive impairment. Frontiers in Behavioral neuroscience, 18, 1446379. https://doi.org/10.3389/fnbeh.2024.1446379

19.   Strandwitz, P., Kim, K. H., Terekhova, D., Liu, J. K., Sharma, A., Levering, J., ... & Lewis, K. (2024). GABA-modulating bacteria of the human gut microbiota. Nature Microbiology, 9(4), 1079-1087. https://doi.org/10.1038/s41564-024-01614-0

20.   Sarkar, A., Lehto, S. M., Harty, S., Dinan, T. G., Cryan, J. F., & Burnet, P. W. (2024). Psychobiotics and the manipulation of bacteria–gut–brain signals. Trends in Neurosciences, 47(3), 178-194. https://doi.org/10.1016/j.tins.2024.01.002

21.   Pinto-Sanchez, M. I., Hall, G. B., Ghajar, K., Nardelli, A., Bolino, C., Lau, J. T., ... & Bercik, P. (2024). Probiotic Bifidobacterium longum NCC3001 reduces depression scores and alters brain activity: A pilot study in patients with IBS. Gastroenterology, 166(4), 701-712. https://doi.org/10.1053/j.gastro.2024.01.032

22.   Chen, L., Zhang, H., Wang, Y., Liu, X., & Zhou, S. (2024). Advanced drug delivery systems for gut-brain axis therapeutics: Challenges and opportunities. Acta Pharmaceutica Sinica B, 14(6), 2456-2478. https://doi.org/10.1016/j.apsb.2024.03.024

23.   Rodriguez-Nogales, A., Algieri, F., Garrido-Mesa, J., Vezza, T., Utrilla, M. P., Chueca, N., ... & Galvez, J. (2024). Intestinal anti-inflammatory activity of RGX-104 in experimental colitis: Role of TLR4 pathway. Biochemical Pharmacology, 220, 115982. https://doi.org/10.1016/j.bcp.2024.115982

24.   Brako, F., & Nkwo, M. (2024). Leveraging artificial intelligence for better translation of fibre-based pharmaceutical systems into real-world benefits. Pharmaceutical Development and Technology, 29(8), 723-735. https://doi.org/10.1080/10837450.2024.2395422

25.   Gharibshahian, M., Torkashvand, M., Farzaneh, A., Rahimi, A., Maleki, A., Rezaei, N., ... & Beheshtizadeh, N. (2024). Recent advances in artificial intelligent strategies for tissue engineering and regenerative medicine. Skin Research and Technology, 30(10), e70016. https://doi.org/10.1111/srt.70016

26.   Martinez-Ballesteros, C., Garcia-Heredia, A., Nepomuceno-Chamorro, I. A., & Riquelme-Santos, J. C. (2024). Machine learning for electrospinning parameter optimization: A systematic review. Expert Systems with Applications, 238, 121847.  https://doi.org/10.1016/j.eswa.2024.121847

27.   Kumar, M., Singh, R., Sharma, P., & Gupta, A. (2024). AI-driven electrospinning: From parameter optimization to smart manufacturing. Materials Today Communications, 38, 108234. https://doi.org/10.1016/j.mtcomm.2024.108234

Zhou, C., Geng, Y., Zhang, C., Hsu, Y., Ma, L., Li, M., ... & Wei, G. (2025). Research progress of responsive electrospun dressings for precise monitoring and treatment of diabetic wounds. Burns & Trauma, 13, tkaf049. https://doi.org/10.1093/burnst/tkaf049