Characterization and Biomedical Applications of Green-Synthesized Selenium Nanoparticles Using Tridax procumbens Stem Extract.

Background Selenium nanoparticles (SeNPs) are one of the metal nanoparticles that have been widely utilized for their anti-microbial, anti-oxidant, anti-inflammatory activities, and other biomedical applications. Tridax procumbens (TP) stem extract is a promising herb species rich in flavonoids, tannins, alkaloids, phytosterols, and hydroxycinnamates, which play a major role in wound healing applications.  Aim The study aims to synthesize SeNPs using TP stem extract, characterizations, and its biomedical applications. Materials and methods SeNPs were synthesized using TP stem extract. The green synthesis of SeNPs was confirmed by ultraviolet-visible (UV-vis) spectra analysis. The synthesized SeNPs were characterized using Fourier-transform infrared spectroscopy (FTIR) and scanning electron microscopy (SEM). The agar well diffusion method was utilized to evaluate the anti-bacterial properties of the green synthesized SeNPs using TP stem extract. The anti-oxidant effect of SeNPs was tested using the 2,2-diphenyl-1-picrylhydrazyl (DPPH) assay, ferric-reducing anti-oxidant power assay (FRAP), and hydroxyl radical scavenging assay (H2O2). The anti-inflammatory effect was investigated using the bovine serum albumin assay and egg albumin denaturation method, and the cytotoxic effect of the green synthesized SeNPs was tested using the brine shrimp lethality (BSL) assay. Results The green synthesis of SeNPs was confirmed using different types of analysis techniques. The characterizations were done by UV-visible spectroscopy analysis, exhibiting a maximum peak at the range of 330 nm. SEM analysis revealed the shape of the nanoparticle to be hexagonal. The agar well diffusion method exhibited the anti-bacterial efficacy of SeNPs against wound microorganisms with a zone of inhibition of 14.6 mm for Escherichia coli (E. coli), 15.8 mm for Staphylococcus aureus (S. aureus), and 15.4 mm for Pseudomonas aeruginosa (P. aeruginosa). The TP stem-mediated SeNPs showed potential effects in anti-oxidant, anti-inflammatory, and cytotoxic activity, which shows very little toxicity. Conclusion Overall, the green synthesis of TP-stem-mediated SeNPs has great potential in biomedical applications. Thus, the synthesized SeNPs exhibit significant anti-bacterial efficacy against wound pathogens. The TP stem-mediated SeNPs showed potential effects in anti-oxidant, anti-inflammatory, and cytotoxic activity, which shows low toxicity. Furthermore, the green-synthesized SeNPs can be utilized in therapeutic management.

Imperceptible augmentation of living systems with organic bioelectronic fibres.

The functional and sensory augmentation of living structures, such as human skin and plant epidermis, with electronics can be used to create platforms for health management and environmental monitoring. Ideally, such bioelectronic interfaces should not obstruct the inherent sensations and physiological changes of their hosts. The full life cycle of the interfaces should also be designed to minimize their environmental footprint. Here we report imperceptible augmentation of living systems through in situ tethering of organic bioelectronic fibres. Using an orbital spinning technique, substrate-free and open fibre networks-which are based on poly (3,4-ethylenedioxythiophene):polystyrene sulfonate-can be tethered to biological surfaces, including fingertips, chick embryos and plants. We use customizable fibre networks to create on-skin electrodes that can record electrocardiogram and electromyography signals, skin-gated organic electrochemical transistors and augmented touch and plant interfaces. We also show that the fibres can be used to couple prefabricated microelectronics and electronic textiles, and that the fibres can be repaired, upgraded and recycled.

Enhancing cellular affinity for skin disorders: Electrospun polyurethane/collagen nanofiber mats coated with phytoceramides.

Although the use of thermoplastic polyurethane (Tpu) nanofiber mats as wound dressings is of great interest due to their mechanical properties, they are hindered by their poor wettability and bioavailability. In this study, we aimed to improve the cellular affinity of Tpu nanofiber mats for skin disorders by incorporating extracted collagen (Col) from tendons and physically mixed with a layer of phytoceramides (Phyto) to produce TpuCol@X-Phyto mats in which the weight % of Phyto relatively to the weight of the solution was X = 0.5, 1, or 1.5 wt% via facile electrospinning approach. The collective observations strongly indicate the successful incorporation and retention of Phyto within the TpuCol architecture. An increase in the Phyto concentration decreased the water contact angle from 69.4°â€¯±â€¯3.47° to 57.9°â€¯±â€¯2.89°, demonstrating improvement in the hydrophilicity of Tpu and binary blend TpuCol nanofiber mats. The mechanical property of 1.0 wt% Phyto aligns with practical requirements owing to the presence of two hydroxyl groups and the amide linkage likely contributing to various hydrogen bonds, providing mechanical strength to the channel structure and a degree of rigidity essential for transmitting mechanical stress. The proliferation of human skin fibroblast (HSF) peaked significantly 100 % with TpuCol@X-Phyto mats coated for X =1.0 and 1.5 wt% of Phyto. Electrospun scaffolds with the highest Phyto content have shown the lowest degree of hemolysis, demonstrating the high level of compatibility between them and blood. The TpuCol@1.5Phyto mat also demonstrated higher efficacy in antibacterial and antioxidant activities, achieving a rate of DPPH radical scavenging of 83.3 % for this latter property. The most notable wound closure among all tested formulations was attributed to higher Phyto. Thus, the developed TpuCol@1.5Phyto nanofiber formula exhibited enhanced healing in an in vitro epidermal model.

Artificial Polymerizations in Living Organisms for Biomedical Applications.

Within living organisms, numerous nanomachines are constantly involved in complex polymerization processes, generating a diverse array of biomacromolecules for maintaining biological activities. Transporting artificial polymerizations from lab settings into biological contexts has expanded opportunities for understanding and managing biological events, creating novel cellular compartments, and introducing new functionalities. This review summarizes the recent advancements in artificial polymerizations, including those responding to external stimuli, internal environmental factors, and those that polymerize spontaneously. More importantly, the cutting-edge biomedical application scenarios of artificial polymerization, notably in safeguarding cells, modulating biological events, improving diagnostic performance, and facilitating therapeutic efficacy are highlighted. Finally, this review outlines the key challenges and technological obstacles that remain for polymerizations in biological organisms, as well as offers insights into potential directions for advancing their practical applications and clinical trials.

Emerging Roles of Graphitic Carbon Nitride-based Materials in Biomedical Applications.

Graphite carbon nitride (g-C3N4) is a two-dimensional conjugated polymer with a unique energy band structure similar to graphene. Due to its outstanding analytical advantages, such as relatively small band gap (2.7 eV), low-cost synthesis, high thermal stability, excellent photocatalytic ability, and good biocompatibility, g-C3N4 has attracted the interest of researchers and industry, especially in the medical field. This paper summarizes the latest research on g-C3N4-based composites in various biomedical applications, including therapy, diagnostic imaging, biosensors, antibacterial, and wearable devices. In addition, the application prospects and possible challenges of g-C3N4 in nanomedicine are also discussed in detail. This review is expected to inspire emerging biomedical applications based on g-C3N4.

Designing intelligent bioorthogonal nanozymes: Recent advances of stimuli-responsive catalytic systems for biomedical applications.

Bioorthogonal nanozymes have emerged as a potent tool in biomedicine due to their unique ability to perform enzymatic reactions that do not interfere with native biochemical processes. The integration of stimuli-responsive mechanisms into these nanozymes has further expanded their potential, allowing for controlled activation and targeted delivery. As such, intelligent bioorthogonal nanozymes have received more and more attention in developing therapeutic approaches. This review provides a comprehensive overview of the recent advances in the development and application of stimuli-responsive bioorthogonal nanozymes. By summarizing the design outlines for anchoring bioorthogonal nanozymes with stimuli-responsive capability, this review seeks to offer valuable insights and guidance for the rational design of these remarkable materials. This review highlights the significant progress made in this exciting field with different types of stimuli and the various applications. Additionally, it also examines the current challenges and limitations in the design, synthesis, and application of these systems, and proposes potential solutions and research directions. This review aims to stimulate further research toward the development of more efficient and versatile stimuli-responsive bioorthogonal nanozymes for biomedical applications.

A framework for decolonising and diversifying biomedical sciences curricula: rediscovery, representation and readiness.

To date, most efforts to decolonise curricula have focussed on the arts and humanities, with many believing that science subjects are objective, unbiased, and unaffected by colonial legacies. However, science is shaped by both contemporary and historical culture. Science has been used to support imperialism, to extract and exploit knowledge and natural resources, and to justify racist and ableist ideologies. Colonial legacies continue to affect scientific knowledge generation and shape contemporary research priorities. In the biomedical sciences, research biases can feed into wider health inequalities. Reflection of these biases in our taught curricula risks perpetuating long-standing inequities to future generations of scientists. We examined attitudes and understanding towards decolonising and diversifying the curriculum among students and teaching staff in the biomedical sciences at the University of Bristol, UK, to discover whether our current teaching practice is perceived as inclusive. We used a mixed methods study including surveys of staff (N = 71) and students (N = 121) and focus groups. Quantitative data showed that staff and students think decolonising the curriculum is important, but this is more important to female respondents (P < 0.001). Students are less aware than staff of current efforts to decolonise the curriculum, while students from minority ethnic groups feel less represented by the curriculum than white students. Thematic analysis of qualitative data revealed three themes that are important for a decolonised curriculum in our context: rediscovery, representation and readiness. We propose that this '3Rs framework' could guide future efforts to decolonise and diversify the curriculum in the biomedical sciences and beyond.

Biomedical potentials of alginate via physical, chemical, and biological modifications.

Alginate is a linear polysaccharide with a modifiable structure and abundant functional groups, offers immense potential for tailoring diverse alginate-based materials to meet the demands of biomedical applications. Given the advancements in modification techniques, it is significant to analyze and summarize the modification of alginate by physical, chemical and biological methods. These approaches provide plentiful information on the preparation, characterization and application of alginate-based materials. Physical modification generally involves blending and physical crosslinking, while chemical modification relies on chemical reactions, mainly including acylation, sulfation, phosphorylation, carbodiimide coupling, nucleophilic substitution, graft copolymerization, terminal modification, and degradation. Chemical modified alginate contains chemically crosslinked alginate, grafted alginate and oligo-alginate. Biological modification associated with various enzymes to realize the hydrolysis or grafting. These diverse modifications hold great promise in fully harnessing the potential of alginate for its burgeoning biomedical applications in the future. In summary, this review provides a comprehensive discussion and summary of different modification methods applied to improve the properties of alginate while expanding its biomedical potentials.

Design of Collagen and Gelatin-based Electrospun Fibers for Biomedical Purposes: An Overview.

Collagen and gelatin are essential natural biopolymers commonly utilized in biomaterials and tissue engineering because of their excellent physicochemical and biocompatibility properties. They can be used either in combination with other biomacromolecules or particles or even exclusively for the enhancement of bone regeneration or for the development of biomimetic scaffolds. Collagen or gelatin derivatives can be transformed into nanofibrous materials with porous micro- or nanostructures and superior mechanical properties and biocompatibility using electrospinning technology. Specific attention was recently paid to electrospun mats of such biopolymers, due to their high ratio of surface area to volume, as well as their biocompatibility, biodegradability, and low immunogenicity. The fiber mats with submicro- and nanometer scale can replicate the extracellular matrix structure of human tissues and organs, making them highly suitable for use in tissue engineering due to their exceptional bioaffinity. The drawbacks may include rapid degradation and complete dissolution in aqueous media. The use of gelatin/collagen electrospun nanofibers in this form is thus greatly restricted for biomedicine. Therefore, the cross-linking of these fibers is necessary for controlling their aqueous solubility. This led to enhanced biological characteristics of the fibers, rendering them excellent options for various biomedical uses. The objective of this review is to highlight the key research related to the electrospinning of collagen and gelatin, as well as their applications in the biomedical field. The review features a detailed examination of the electrospinning fiber mats, showcasing their varying structures and performances resulting from diverse solvents, electrospinning processes, and cross-linking methods. Judiciously selected examples from literature will be presented to demonstrate major advantages of such biofibers. The current developments and difficulties in this area of research are also being addressed.

A real-time interactive restoration system for intraoral digital videos using segment anything model.

Objective: Poor conditions in the intraoral environment often lead to low-quality photos and videos, hindering further clinical diagnosis. To restore these digital records, this study proposes a real-time interactive restoration system using segment anything model. Methods: Intraoral digital videos, obtained from the vident-lab dataset through an intraoral camera, serve as the input for interactive restoration system. The initial phase employs an interactive segmentation module leveraging segment anything model. Subsequently, a real-time intraframe restoration module and a video enhancement module were designed. A series of ablation studies were systematically conducted to illustrate the superior design of interactive restoration system. Our quantitative evaluation criteria contain restoration quality, segmentation accuracy, and processing speed. Furthermore, the clinical applicability of the processed videos was evaluated by experts. Results: Extensive experiments demonstrated its performance on segmentation with a mean intersection-over-union of 0.977. On video restoration, it leads to reliable performances with peak signal-to-noise ratio of 37.09 and structural similarity index measure of 0.961, respectively. More visualization results are shown on the https://yogurtsam.github.io/iveproject page. Conclusion: Interactive restoration system demonstrates its potential to serve patients and dentists with reliable and controllable intraoral video restoration.

Ecofriendly biosynthesis of copper nanoparticles from novel marine S. rhizophila species for enhanced antibiofilm, antimicrobial and antioxidant potential.

Marine microorganisms offer a promising avenue for the eco-friendly synthesis of nanoparticles due to their unique biochemical capabilities and adaptability to various environments. This study focuses on exploring the potential of a marine bacterial species, Stenotrophomonas rhizophila BGNAK1, for the synthesis of biocompatible copper nanoparticles and their application for hindering biofilms formed by monomicrobial species. The study begins with the isolation of the novel marine S. rhizophila species from marine soil samples collected from the West coast region of Kerala, India. The isolated strain is identified through 16S rRNA gene sequencing and confirmed to be S. rhizophila species. Biosynthesis of copper nanoparticles using S. rhizophila results in the formation of nanoparticles with size of range 10-50 nm. The nanoparticles exhibit a face-centered cubic crystal structure of copper, as confirmed by X-Ray Diffraction analysis. Furthermore, the synthesized nanoparticles display significant antimicrobial activity against various pathogenic bacteria and yeast. The highest inhibitory activity was against Staphylococcus aureus with a zone of 27 ± 1.00 mm and the least activity was against Pseudomonas aeruginosa with a zone of 22 ± 0.50 mm. The zone of inhibition against Candida albicans was 16 ± 0.60 mm. The antibiofilm activity against biofilm-forming clinical pathogens was evidenced by the antibiofilm assay and SEM images. Additionally, the copper nanoparticles exhibit antioxidant activity, as evidenced by their scavenging ability against DPPH, hydroxyl, nitric oxide, and superoxide radicals, as well as their reducing power in the FRAP assay. The study highlights the potential of the marine bacterium S. rhizophila BGNAK1 for the eco-friendly biosynthesis of copper nanoparticles with diverse applications. Synthesized nanoparticles exhibit promising antibiofilm, antimicrobial, and antioxidant properties, suggesting their potential utility in various fields such as medicine, wastewater treatment, and environmental remediation.

What public health challenges and unmet medical needs would benefit from interdisciplinary collaboration in the EU? A survey and multi-stakeholder debate.

In the past decade, significant European calls for research proposals have supported translational collaborative research on non-communicable and infectious diseases within the biomedical life sciences by bringing together interdisciplinary and multinational consortia. This research has advanced our understanding of disease pathophysiology, marking considerable scientific progress. Yet, it is crucial to retrospectively evaluate these efforts' societal impact. Research proposals should be thoughtfully designed to ensure that the research findings can be effectively translated into actionable policies. In addition, the choice of scientific methods plays a pivotal role in shaping the societal impact of research discoveries. Understanding the factors responsible for current unmet public health issues and medical needs is crucial for crafting innovative strategies for research policy interventions. A multistakeholder survey and a roundtable helped identify potential needs for consideration in the EU research and policy agenda. Based on survey findings, mental health disorders, metabolic syndrome, cancer, antimicrobial resistance, environmental pollution, and cardiovascular diseases were considered the public health challenges deserving prioritisation. In addition, early diagnosis, primary prevention, the impact of environmental pollution on disease onset and personalised medicine approaches were the most selected unmet medical needs. Survey findings enabled the formulation of some research-policies interventions (RPIs), which were further discussed during a multistakeholder online roundtable. The discussion underscored recent EU-level activities aligned with the survey-derived RPIs and facilitated an exchange of perspectives on public health and biomedical research topics ripe for interdisciplinary collaboration and warranting attention within the EU's research and policy agenda. Actionable recommendations aimed at facilitating the translation of knowledge into transformative, science-based policies are also provided.

[Childhood overweight and obesity: A biomedical challenge in Mexico]. / Sobrepeso y obesidad infantil: un reto biomédico en México.

In Mexico, 1 out of 3 schoolchildren aged 5 to 11 years is overweight or obese, which represents one of the main public health concerns, due to the fact that this condition in the child population is highly associated with the development of metabolic complications in adults. To date, dietary and physical activity interventions to prevent this problem have shown modest results worldwide. Biomedical studies in Mexico have shown that the pathophysiology of childhood overweight and obesity presents different molecular patterns, inflammation and oxidative stress, possibly associated with specific variants in the genome. However, the challenge is to achieve a secure characterization of this evidence so that it can be used in intervention studies aimed to improve the ability to predict and treat childhood overweight and obesity in Mexico. The biomedical challenge is to make knowledge a prevention strategy in families, in society and in the country, in order to fight the serious problem of obesity and its consequences.

En México 1 de cada 3 escolares de 5 a 11 años presenta sobrepeso u obesidad, lo cual representa una de las principales preocupaciones de salud pública, debido a que en la población infantil este padecimiento se asocia altamente con el desarrollo de complicaciones metabólicas en el adulto. Hasta el momento las intervenciones dietéticas y de actividad física para prevenir este problema han mostrado resultados modestos a nivel mundial. Los estudios biomédicos en México han demostrado que la fisiopatología del sobrepeso y la obesidad infantil presenta diferentes patrones moleculares, de inflamación y de estrés oxidativo, posiblemente asociados a variantes específicas en el genoma. Sin embargo, el reto es lograr la caracterización segura de estas evidencias para que sea posible emplearlas en los estudios de intervención encaminados a mejorar la capacidad de predicción y tratamiento del sobrepeso y la obesidad infantil en México. El reto biomédico es hacer del conocimiento una estrategia de prevención en las familias, en la sociedad y en el país, a fin de combatir el grave problema de la obesidad y sus consecuencias.

A Quarter-Century of Online Informatics Education: Learners Served and Lessons Learned.

The value and methods of online learning have changed tremendously over the last 25 years. The goal of this paper is to review a quarter-century of experience with online learning by the author in the field of biomedical and health informatics, describing the learners served and the lessons learned. The author details the history of the decision to pursue online education in informatics, describing the approaches taken as educational technology evolved over time. A large number of learners have been served, and the online learning approach has been well-received, with many lessons learned to optimize the educational experience. Online education in biomedical and health informatics has provided a scalable and exemplary approach to learning in this field.

The sample locator: A federated search tool for biosamples and associated data in Europe using HL7 FHIR.

BACKGROUND: This study outlines the development of a highly interoperable federated IT infrastructure for academic biobanks located at the major university hospital sites across Germany. High-quality biosamples linked to clinical data, stored in biobanks are essential for biomedical research. We aimed to facilitate the findability of these biosamples and their associated data. Networks of biobanks provide access to even larger pools of samples and data even from rare diseases and small disease subgroups. The German Biobank Alliance (GBA) established in 2017 under the umbrella of the German Biobank Node (GBN), has taken on the mission of a federated data discovery service to make biosamples and associated data available to researchers across Germany and Europe. METHODS: In this context, we identified the requirements of researchers seeking human biosamples from biobanks and the needs of biobanks for data sovereignty over their samples and data in conjunction with the sample donor's consent. Based on this, we developed a highly interoperable federated IT infrastructure using standards such as Fast Healthcare Interoperability Resources (HL7 FHIR) and Clinical Quality Language (CQL). RESULTS: The infrastructure comprises two major components enabling federated real-time access to biosample metadata, allowing privacy-compliant queries and subsequent project requests. It has been in use since 2019, connecting 16 German academic biobanks, with additional European biobanks joining. In production since 2019 it has run 4941 queries over the span of one year on more than 900,000 biosamples collected from more than 170,000 donors. CONCLUSION: This infrastructure enhances the visibility and accessibility of biosamples for research, addressing the growing demand for human biosamples and associated data in research. It also underscores the need for improvements in processes beyond IT infrastructure, aiming to advance biomedical research and similar infrastructure development in other fields.

Synthesis of Hydrofluoric (HF) Acid Free, Two-Dimensional (2D) Tantalum Carbide MXene Nanosheets and Quantum Dots.

MXenes are two-dimensional (2D) transition metal-based carbides, nitrides, and carbonitrides that are synthesized from its precursor MAX phase. The selective etching of the "A" from the MAX phase yields multi-functional MXenes that hold promise in a wide range of energy-based applications and biomedical applications. Based on its intended application, MXenes are prepared as multilayered sheets, monolayer flakes, and quantum dots. Conventionally, MXenes are prepared using hydrofluoric (HF) acid etching; however, the use of HF impedes its effective use in biomedical applications. This calls for the use of nontoxic HF-free synthesis protocols to prepare MXenes safe for biological use. Therefore, we have discussed a facile process to synthesize biocompatible, HF-free MXene nanosheets and quantum dots.

Hybrid brain tumor classification of histopathology hyperspectral images by linear unmixing and an ensemble of deep neural networks.

Hyperspectral imaging has demonstrated its potential to provide correlated spatial and spectral information of a sample by a non-contact and non-invasive technology. In the medical field, especially in histopathology, HSI has been applied for the classification and identification of diseased tissue and for the characterization of its morphological properties. In this work, we propose a hybrid scheme to classify non-tumor and tumor histological brain samples by hyperspectral imaging. The proposed approach is based on the identification of characteristic components in a hyperspectral image by linear unmixing, as a features engineering step, and the subsequent classification by a deep learning approach. For this last step, an ensemble of deep neural networks is evaluated by a cross-validation scheme on an augmented dataset and a transfer learning scheme. The proposed method can classify histological brain samples with an average accuracy of 88%, and reduced variability, computational cost, and inference times, which presents an advantage over methods in the state-of-the-art. Hence, the work demonstrates the potential of hybrid classification methodologies to achieve robust and reliable results by combining linear unmixing for features extraction and deep learning for classification.

Autism spectrum disorder detection using facial images: A performance comparison of pretrained convolutional neural networks.

Autism spectrum disorder (ASD) is a complex psychological syndrome characterized by persistent difficulties in social interaction, restricted behaviours, speech, and nonverbal communication. The impacts of this disorder and the severity of symptoms vary from person to person. In most cases, symptoms of ASD appear at the age of 2 to 5 and continue throughout adolescence and into adulthood. While this disorder cannot be cured completely, studies have shown that early detection of this syndrome can assist in maintaining the behavioural and psychological development of children. Experts are currently studying various machine learning methods, particularly convolutional neural networks, to expedite the screening process. Convolutional neural networks are considered promising frameworks for the diagnosis of ASD. This study employs different pre-trained convolutional neural networks such as ResNet34, ResNet50, AlexNet, MobileNetV2, VGG16, and VGG19 to diagnose ASD and compared their performance. Transfer learning was applied to every model included in the study to achieve higher results than the initial models. The proposed ResNet50 model achieved the highest accuracy, 92%, compared to other transfer learning models. The proposed method also outperformed the state-of-the-art models in terms of accuracy and computational cost.

Tensor Methods in Biomedical Image Analysis.

In the past decade, tensors have become increasingly attractive in different aspects of signal and image processing areas. The main reason is the inefficiency of matrices in representing and analyzing multimodal and multidimensional datasets. Matrices cannot preserve the multidimensional correlation of elements in higher-order datasets and this highly reduces the effectiveness of matrix-based approaches in analyzing multidimensional datasets. Besides this, tensor-based approaches have demonstrated promising performances. These together, encouraged researchers to move from matrices to tensors. Among different signal and image processing applications, analyzing biomedical signals and images is of particular importance. This is due to the need for extracting accurate information from biomedical datasets which directly affects patient's health. In addition, in many cases, several datasets have been recorded simultaneously from a patient. A common example is recording electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) of a patient with schizophrenia. In such a situation, tensors seem to be among the most effective methods for the simultaneous exploitation of two (or more) datasets. Therefore, several tensor-based methods have been developed for analyzing biomedical datasets. Considering this reality, in this paper, we aim to have a comprehensive review on tensor-based methods in biomedical image analysis. The presented study and classification between different methods and applications can show the importance of tensors in biomedical image enhancement and open new ways for future studies.

Advancing Accuracy in Multimodal Medical Tasks Through Bootstrapped Language-Image Pretraining (BioMedBLIP): Performance Evaluation Study.

BACKGROUND: Medical image analysis, particularly in the context of visual question answering (VQA) and image captioning, is crucial for accurate diagnosis and educational purposes. OBJECTIVE: Our study aims to introduce BioMedBLIP models, fine-tuned for VQA tasks using specialized medical data sets such as Radiology Objects in Context and Medical Information Mart for Intensive Care-Chest X-ray, and evaluate their performance in comparison to the state of the art (SOTA) original Bootstrapping Language-Image Pretraining (BLIP) model. METHODS: We present 9 versions of BioMedBLIP across 3 downstream tasks in various data sets. The models are trained on a varying number of epochs. The findings indicate the strong overall performance of our models. We proposed BioMedBLIP for the VQA generation model, VQA classification model, and BioMedBLIP image caption model. We conducted pretraining in BLIP using medical data sets, producing an adapted BLIP model tailored for medical applications. RESULTS: In VQA generation tasks, BioMedBLIP models outperformed the SOTA on the Semantically-Labeled Knowledge-Enhanced (SLAKE) data set, VQA in Radiology (VQA-RAD), and Image Cross-Language Evaluation Forum data sets. In VQA classification, our models consistently surpassed the SOTA on the SLAKE data set. Our models also showed competitive performance on the VQA-RAD and PathVQA data sets. Similarly, in image captioning tasks, our model beat the SOTA, suggesting the importance of pretraining with medical data sets. Overall, in 20 different data sets and task combinations, our BioMedBLIP excelled in 15 (75%) out of 20 tasks. BioMedBLIP represents a new SOTA in 15 (75%) out of 20 tasks, and our responses were rated higher in all 20 tasks (P<.005) in comparison to SOTA models. CONCLUSIONS: Our BioMedBLIP models show promising performance and suggest that incorporating medical knowledge through pretraining with domain-specific medical data sets helps models achieve higher performance. Our models thus demonstrate their potential to advance medical image analysis, impacting diagnosis, medical education, and research. However, data quality, task-specific variability, computational resources, and ethical considerations should be carefully addressed. In conclusion, our models represent a contribution toward the synergy of artificial intelligence and medicine. We have made BioMedBLIP freely available, which will help in further advancing research in multimodal medical tasks.