Machine Learning in Tissue Engineering

Virtual histopathology methods in medical imaging - a systematic review

Virtual histopathology is an emerging technology in medical imaging that utilizes advanced computational methods to analyze tissue images for more precise disease diagnosis. Traditionally, histopathology relies on manual techniques and expertise, often resulting in time-consuming processes and variability in diagnoses. Virtual histopathology offers a more consistent, and automated approach, employing techniques like machine learning, deep learning, and image processing to simulate staining and enhance tissue analysis. This review explores the strengths, limitations, and clinical applications of these methods, highlighting recent advancements in virtual histopathological approaches. In addition, important areas are identified for future research to improve diagnostic accuracy and efficiency in clinical settings.

Predicting ultrasound wave stimulated bone growth in bioinspired scaffolds using machine learning

For conditions like osteoporosis, changes in bone pore geometry even when porosity is constant have been shown to correlate to increased fracture risk using techniques such as dual-energy x-ray absorptiometry (DXA) and computed tomography (CT). Additionally, studies have found that bone pore geometry can be characterized by ultrasound to determine fracture risk, since certain pore geometries can cause stress concentration which in turn will be a source for fracture. However, it is not yet fully understood if changes in pore geometry can be detected by ultrasound when the porosity is constant. Therefore, this study develops an unsupervised machine learning model classifying pore geometry between bioinspired and quadrilateral pore scaffolds with constant porosity using experimental ultrasound wave transmission data. Our results demonstrate that differences in pore geometry can be detected by ultrasound, even at constant porosity, and that these differences can be distinguished in an unsupervised manner with machine learning. For traumatic bone injuries and late-stage osteoporosis where fracture occurs, tissue scaffolds are used to aid the healing of fractures or bone loss. The scaffold design is optimized to match material properties closely with bone, and healing can be enhanced with ultrasound stimulation. In this study we predict the combined effects of ultrasound parameters, such as wave frequency and mode of displacement, and scaffold material properties on bone tissue growth. We therefore develop an unsupervised machine learning clustering model of bone tissue growth in the scaffolds using finite element analysis and bone growth algorithms evaluating effects of pore geometry, scaffold materials, ultrasound wave type and frequency, and mesenchymal stem cell distribution on bone tissue growth. The computational predictions of tissue growth agreed within 10% of comparable experimental studies. The data corresponding to pore geometry, mesenchymal stem cell distribution, and scaffold material demonstrate distinct clusters of total bone formation, while ultrasound frequency and mesenchymal stem cell distribution show distinct clusters in bone growth rate. These variables can be tuned to tailor the scaffold design and optimize the required amount and rate of bone growth to meet a patient's specific needs.

Hematoxylin and Eosin Architecture Uncovers Clinically Divergent Niches in Pancreatic Cancer

Pancreatic ductal adenocarcinoma (PDAC) represents one of the only cancers with an increasing incidence rate and is often associated with intra- and peri-tumoral scarring, referred to as desmoplasia. This scarring is highly heterogeneous in extracellular matrix (ECM) architecture and plays complex roles in both tumor biology and clinical outcomes that are not yet fully understood. Using hematoxylin and eosin (H&E), a routine histological stain utilized in existing clinical workflows, we quantified ECM architecture in 85 patient samples to assess relationships between desmoplastic architecture and clinical outcomes such as survival time and disease recurrence. By utilizing unsupervised machine learning to summarize a latent space across 147 local (e.g., fiber length, solidity) and global (e.g., fiber branching, porosity) H&E-based features, we identified a continuum of histological architectures that were associated with differences in both survival and recurrence. Furthermore, we mapped H&E architectures to a CO-Detection by indEXing (CODEX) reference atlas, revealing localized cell- and protein-based niches associated with outcome-positive versus outcome-negative scarring in the tumor microenvironment. Overall, our study utilizes standard H&E staining to uncover clinically relevant associations between desmoplastic organization and PDAC outcomes, offering a translatable pipeline to support prognostic decision-making and a blueprint of spatial-biological factors for modeling by tissue engineering methods.

Artificial Intelligence (AI): A Potential Game Changer in Regenerative Orthopedics-A Scoping Review

Background and Aims

Regenerative orthopedics involves approaches like stem cell therapy, platelet-rich plasma (PRP) therapy, the use of biological scaffold implants, tissue engineering, etc. We aim to present a scoping review of the role of artificial intelligence (AI) in different treatment approaches of regenerative orthopedics.

Methods

Using the PRISMA guidelines, a search for articles for the last ten years (2013-2024) on PubMed was done, using several keywords. We have discussed the state-of-the-art, strengths/benefits, and limitations of the published research, and provide a useful resource for the way ahead in future for researchers working in this area.

Results

Using the eligibility criteria out of 82 initially screened publications, we included 18 studies for this review. We noticed that the treatment responses to regenerative treatments depend on several factors; hence, to facilitate better comprehensive and patient-specific treatments, AI technology is very useful. Machine learning (ML) and deep learning (DL) are a few of the most frequently used AI techniques. They use a data-driven approach for training models to make human-like decisions. Data are fed to the ML/DL algorithm and the trained model makes classifications or predictions based on its learning.

Conclusion

The area of regenerative orthopedics is highly sophisticated and significantly aids in providing cost-effective and non-invasive treatments to patients suffering from orthopedic ailments and injuries. Due to its promising future, the use of AI in regenerative orthopedics is an emerging and promising research field; however, its universal clinical applications are associated with some ethical considerations, which need addressing.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s43465-024-01189-1.

Constraint based Bayesian optimization of bioink precursor: a machine learning framework

Current research practice for optimizing bioink involves exhaustive experimentation with multi-material composition for determining the printability, shape fidelity and biocompatibility. Predicting bioink properties can be beneficial to the research community but is a challenging task due to the non-Newtonian behavior in complex composition. Existing models such as Cross model become inadequate for predicting the viscosity for heterogeneous composition of bioinks. In this paper, we utilize a machine learning framework to accurately predict the viscosity of heterogeneous bioink compositions, aiming to enhance extrusion-based bioprinting techniques. Utilizing Bayesian optimization (BO), our strategy leverages a limited dataset to inform our model. This is a technique especially useful of the typically sparse data in this domain. Moreover, we have also developed a mask technique that can handle complex constraints, informed by domain expertise, to define the feasible parameter space for the components of the bioink and their interactions. Our proposed method is focused on predicting the intrinsic factor (e.g. viscosity) of the bioink precursor which is tied to the extrinsic property (e.g. cell viability) through the mask function. Through the optimization of the hyperparameter, we strike a balance between exploration of new possibilities and exploitation of known data, a balance crucial for refining our acquisition function. This function then guides the selection of subsequent sampling points within the defined viable space and the process continues until convergence is achieved, indicating that the model has sufficiently explored the parameter space and identified the optimal or near-optimal solutions. Employing this AI-guided BO framework, we have developed, tested, and validated a surrogate model for determining the viscosity of heterogeneous bioink compositions. This data-driven approach significantly reduces the experimental workload required to identify bioink compositions conducive to functional tissue growth. It not only streamlines the process of finding the optimal bioink compositions from a vast array of heterogeneous options but also offers a promising avenue for accelerating advancements in tissue engineering by minimizing the need for extensive experimental trials.

Programming mammalian cell behaviors by physical cues

In recent decades, the field of synthetic biology has witnessed remarkable progress, driving advances in both research and practical applications. One pivotal area of development involves the design of transgene switches capable of precisely regulating specified outputs and controlling cell behaviors in response to physical cues, which encompass light, magnetic fields, temperature, mechanical forces, ultrasound, and electricity. In this review, we delve into the cutting-edge progress made in the field of physically controlled protein expression in engineered mammalian cells, exploring the diverse genetic tools and synthetic strategies available for engineering targeting cells to sense these physical cues and generate the desired outputs accordingly. We discuss the precision and efficiency limitations inherent in these tools, while also highlighting their immense potential for therapeutic applications.

Modelling and targeting mechanical forces in organ fibrosis

Few efficacious therapies exist for the treatment of fibrotic diseases, such as skin scarring, liver cirrhosis and pulmonary fibrosis, which is related to our limited understanding of the fundamental causes and mechanisms of fibrosis. Mechanical forces from cell-matrix interactions, cell-cell contact, fluid flow and other physical stimuli may play a central role in the initiation and propagation of fibrosis. In this Review, we highlight the mechanotransduction mechanisms by which various sources of physical force drive fibrotic disease processes, with an emphasis on central pathways that may be therapeutically targeted to prevent and reverse fibrosis. We then discuss engineered models of mechanotransduction in fibrosis, as well as molecular and biomaterials-based therapeutic approaches for limiting fibrosis and promoting regenerative healing phenotypes in various organs. Finally, we discuss challenges within fibrosis research that remain to be addressed and that may greatly benefit from next-generation bioengineered model systems.

Optimization by mixture design of chitosan/multi-phase calcium phosphate/BMP-2 biomimetic scaffolds for bone tissue engineering

The modulation of cell behavior during culture is one of the most important aspects of bone tissue engineering because of the necessity for a complex mechanical and biochemical environment. This study aimed to improve the physicochemical properties of chitosan/multi-phase calcium phosphate (MCaP) scaffolds using an optimized mixture design experiment and evaluate the effect of biofunctionalization of the obtained scaffolds with the bone morphogenetic protein BMP-2 on stem cell behavior. The present study evaluated the compressive strength, elastic modulus, porosity, pore diameter, and degradation in simulated body fluids and integrated these responses using desirability. The properties of the scaffolds with the best desirability (18.4% of MCaP) were: compressive strength of 23 kPa, elastic modulus of 430 kPa, pore diameter of 163 μm, porosity of 92%, and degradation of 20% after 21 days. Proliferation and differentiation experiments were conducted using dental pulp stem cells after grafting BMP-2 onto scaffolds via the carbodiimide route. These experiments showed that MCaP promoted cell proliferation and increased alkaline phosphatase activity, whereas BMP-2 enhanced cell differentiation. This study demonstrates that optimizing the composition of a mixture of chitosan and MCaP improves the physicochemical and biological properties of scaffolds, indicating that this solution is viable for application in bone tissue engineering.

Trends in bioactivity: inducing and detecting mineralization of regenerative polymeric scaffolds

Due to limitations of biological and alloplastic grafts, regenerative engineering has emerged as a promising alternative to treat bone defects. Bioactive polymeric scaffolds are an integral part of such an approach. Bioactivity importantly induces hydroxyapatite mineralization that promotes osteoinductivity and osseointegration with surrounding bone tissue. Strategies to confer bioactivity to polymeric scaffolds utilize bioceramic fillers, coatings and surface treatments, and additives. These approaches can also favorably impact mechanical and degradation properties. A variety of fabrication methods are utilized to prepare scaffolds with requisite morphological features. The bioactivity of scaffolds may be evaluated with a broad set of techniques, including in vitro (acellular and cellular) and in vivo methods. Herein, we highlight contemporary and emerging approaches to prepare and assess scaffold bioactivity, as well as existing challenges.

Fusion and classification algorithm of octacalcium phosphate production based on XRD and FTIR data

The present manuscript tested an automated analysis sequence to provide a decision support system to track the OCP synthesis from [Formula: see text]-TCP over time. Initially, the XRD and FTIR signals from a hundredfold scaled-up hydrolysis of OCP from [Formula: see text]-TCP were fused and modeled by the curve fitting based on the significantly established maxima from the literature and nine features extracted from the fitted shapes. Afterward, the analysis sequence enclosed the machine learning techniques for feature ranking, spatial filtering, and dimensionality reduction to support the automatic recognition of the synthesis stages. The proposed analysis pipeline for OCP identification might be the foundation for a decision support system explicitly targeting OCP synthesis. Future projects will exploit the suggested methodology for pinpointing the OCP production over time (including the intermediary phases present in the OCP formation) and for evaluating whether biological variables might be merged with biomaterial properties to build a unified model of tissue response to the implant.

Development of Bioactive Scaffolds for Orthopedic Applications by Designing Additively Manufactured Titanium Porous Structures: A Critical Review

We overview recent findings achieved in the field of model-driven development of additively manufactured porous materials for the development of a new generation of bioactive implants for orthopedic applications. Porous structures produced from biocompatible titanium alloys using selective laser melting can present a promising material to design scaffolds with regulated mechanical properties and with the capacity to be loaded with pharmaceutical products. Adjusting pore geometry, one could control elastic modulus and strength/fatigue properties of the engineered structures to be compatible with bone tissues, thus preventing the stress shield effect when replacing a diseased bone fragment. Adsorption of medicals by internal spaces would make it possible to emit the antibiotic and anti-tumor agents into surrounding tissues. The developed internal porosity and surface roughness can provide the desired vascularization and osteointegration. We critically analyze the recent advances in the field featuring model design approaches, virtual testing of the designed structures, capabilities of additive printing of porous structures, biomedical issues of the engineered scaffolds, and so on. Special attention is paid to highlighting the actual problems in the field and the ways of their solutions.

Rapid prediction of lab-grown tissue properties using deep learning

The interactions between cells and the extracellular matrix are vital for the self-organisation of tissues. In this paper we present proof-of-concept to use machine learning tools to predict the role of this mechanobiology in the self-organisation of cell-laden hydrogels grown in tethered moulds. We develop a process for the automated generation of mould designs with and without key symmetries. We create a large training set withN = 6400 cases by running detailed biophysical simulations of cell-matrix interactions using the contractile network dipole orientation model for the self-organisation of cellular hydrogels within these moulds. These are used to train an implementation of thepix2pixdeep learning model, with an additional 100 cases that were unseen in the training of the neural network for review and testing of the trained model. Comparison between the predictions of the machine learning technique and the reserved predictions from the biophysical algorithm show that the machine learning algorithm makes excellent predictions. The machine learning algorithm is significantly faster than the biophysical method, opening the possibility of very high throughput rational design of moulds for pharmaceutical testing, regenerative medicine and fundamental studies of biology. Future extensions for scaffolds and 3D bioprinting will open additional applications.

Emerging Bioprinting for Wound Healing

Bioprinting has attracted much attention due to its suitability for fabricating biomedical devices. In particular, bioprinting has become one of the growing centers in the field of wound healing, with various types of bioprinted devices being developed, including 3D scaffolds, microneedle patches, and flexible electronics. Bioprinted devices can be designed with specific biostructures and biofunctions that closely match the shape of wound sites and accelerate the regeneration of skin through various approaches. Herein, a comprehensive review of the bioprinting of smart wound dressings is presented, emphasizing the crucial effect of bioprinting in determining biostructures and biofunctions. The review begins with an overview of bioprinting techniques and bioprinted devices, followed with an in-depth discussion of polymer-based inks, modification strategies, additive ingredients, properties, and applications. The strategies for the modification of bioprinted devices are divided into seven categories, including chemical synthesis of novel inks, physical blending, coaxial bioprinting, multimaterial bioprinting, physical absorption, chemical immobilization, and hybridization with living cells, and examples are presented. Thereafter, the frontiers of bioprinting and wound healing, including 4D bioprinting, artificial intelligence-assisted bioprinting, and in situ bioprinting, are discussed from a perspective of interdisciplinary sciences. Finally, the current challenges and future prospects in this field are highlighted.

Bioprinted Hydrogels for Fibrosis and Wound Healing: Treatment and Modeling

Three-dimensional (3D) printing has been used to fabricate biomaterial scaffolds with finely controlled physical architecture and user-defined patterning of biological ligands. Excitingly, recent advances in bioprinting have enabled the development of highly biomimetic hydrogels for the treatment of fibrosis and the promotion of wound healing. Bioprinted hydrogels offer more accurate spatial recapitulation of the biochemical and biophysical cues that inhibit fibrosis and promote tissue regeneration, augmenting the therapeutic potential of hydrogel-based therapies. Accordingly, bioprinted hydrogels have been used for the treatment of fibrosis in a diverse array of tissues and organs, including the skin, heart, and endometrium. Furthermore, bioprinted hydrogels have been utilized for the healing of both acute and chronic wounds, which present unique biological microenvironments. In addition to these therapeutic applications, hydrogel bioprinting has been used to generate in vitro models of fibrosis in a variety of soft tissues such as the skin, heart, and liver, enabling high-throughput drug screening and tissue analysis at relatively low cost. As biological research begins to uncover the spatial biological features that underlie fibrosis and wound healing, bioprinting offers a powerful toolkit to recapitulate spatially defined pro-regenerative and anti-fibrotic cues for an array of translational applications.