3D Bioprinting: A Solution To The Organ Shortage Crisis?

Ammielle WB
14 min readApr 30, 2021


Every 9 minutes, another person is added to the national organ transplant waiting list. At least 17 people die each day waiting for an organ transplant in the United States. In 2013, some 2.8 million patients were undergoing dialysis worldwide, even though organ transplants offer patients a better and longer life. The University of Newcastle suggests that 10 million people around the world require surgery to prevent corneal blindness which could be treated by human cornea transplants.

The demand for transplants far exceeds the supply of suitable organs. And the organ shortage crisis is growing worse every year as demand increases faster than supply. The need for donor kidneys in the United States is rising at 8% per year. Yet only 3 in 1,000 people die in a way that allows for organ donation.

This crisis has deprived thousands of patients of a better quality of life and caused a substantial increase in the cost of alternative medical care. In addition, this scarcity has caused considerable consequences, like the rise of an international black market in organ trade.

And even if a transplant candidate makes it to the top of the list, the risk of complications, such as organ failure and immunorejection, remains significant. Not only do physical pain and limitations during recovery need to be addressed, but any major surgery also has a significant effect on the patient’s financial, mental, and emotional well-being. The potential side effects and disadvantages of donating and receiving an organ are significant.

But what if we could skip the whole process? What if we no longer had to harvest organs from people? What if we could make on-demand organs with a patient’s own cells?

Similar to 3D printing, bioprinting is an additive manufacturing approach in which three-dimensional objects are built from a computer-aided design (CAD) or digital 3D model. But unlike 3D printing, bioprinters print with cells and biomaterials, creating organ-like structures that allow living cells to multiply.

The goal of bioprinting is to ultimately create anatomically-correct biological structures for clinical and commercial applications.

The bioprinting process mainly involves preparation, printing, maturation, and application and can be summarized in 3 key steps:

  1. Pre-bioprinting: From computed tomography (CT) and magnetic resonance imaging (MRI) scans, a digital file can be created for the bioprinter to read. Researchers prepare cells and mix them with bioinks to ensure that a tissue model can be successfully printed.
  2. Bioprinting: Researchers then load the cell-laden bioink into a cartridge and choose one or multiple print heads depending on the desired construct. Developing different types of tissue requires researchers to use different types of cells, bioinks, and equipment.
  3. Post-bioprinting: To create stable structures for the biological material, printed parts usually undergo mechanical and chemical stimulation. Finally, the cell-laden constructs are placed inside an incubator or bioreactor for cultivation.

Although there are several bioprinting methods, based on extrusion, inkjet, acoustic, or laser technologies, the typical process is as follows:

  • 3D imaging should provide a perfect fit of the desired tissue model, with exact dimensions and minimal or no necessary adjustments.
  • During 3D modeling, a blueprint is generated in high detail. Using CAD software, the blueprint may include layer-by-layer instructions for the bioprinter. Fine adjustments may be made at this stage to avoid the transfer of defects.
  • The process of bioink preparation combines living cells with a biomaterial, such as collagen, gelatin, hyaluronan, silk, alginate, or nanocellulose. Bioink provides the cells with the nutrients to survive and a scaffold on which they can grow. The complete substance is patient-based and function-specific.
  • 3D printing consists of depositing the bioink layer by layer, each having a thickness of 0.5 mm or less. The number of nozzles and the type of desired tissue determine the amount of deposit dispensed.
  • Finally, during solidification, the viscous liquid starts to hold its shape as more layers are continuously deposited. The process of blending and solidification is known as crosslinking and may be aided by UV light, specific chemicals, or heat.

What’s the matter with bioinks?

Formulating the right bioink is difficult. To ensure correct functionality, an ideal bioink should possess the mechanical, rheological (flow/deformation), and biological properties of the target tissues.

Bioinks must support the adhesion, proliferation, and differentiation of living cells by providing a desirable environment while also meeting the requirements of printing, such as viscoelasticity, shear thinning, and printability.

Think of designing bioinks as using a soundboard. There are so many switches, dials, and knobs that influence the sound produced. You have to experiment, adjust, and try until you find the desired result. But biology is incredibly complex, and in bioinks, the environmental stimuli and forces applied during printing and their impact on the viability and biological performance of the cells must be well understood. It is nearly impossible to perform individual experiments for an infinite number of cases with different bioprinting conditions and bioink formulations. Finding the bioink that supports the desired cellular processes and biological functions of a construct is therefore an enormous challenge.

Here’s what experts had to say when asked “What are the biggest challenges in bioprinting?”:

“Regardless of the technique used to deposit living mammalian cells (of which there are many), the biggest challenge to bioprinting is the ‘ink’ — i.e. the physicochemical composition and properties of the extracellular material used as the vehicle to carry and deposit the cells. This material dictates both cell survival (including ability to cope with the stressors of printing — shear, compression, decompression, pH, T°C, osmolarity, light energy, etc.) and subsequent behaviour — proliferation, quiescence, differentiation, death (all forms), transdifferentiation, and ultimately ability to form multicellular tissue.”

“If the ink is inappropriate, secondary parameters are unimportant, because the process will not deliver the outcome you want!”

Leavesley, David. (2018). Re: What are the biggest challenges in bioprinting?. Retrieved from: https://www.researchgate.net/post/What_are_the_biggest_challenges_in_bioprinting/5b17906cc4be9355e741c987/citation/download.

“To find the optimal bioink!”

Hammad, Mira. (2018). Re: What are the biggest challenges in bioprinting?. Retrieved from: https://www.researchgate.net/post/What_are_the_biggest_challenges_in_bioprinting/5b17aba7eb87035d4d580029/citation/download.

The printability of bioink, the control of multiple cell types for co-culture, and long-term cell culture performance study.”

Ding, Houzhu. (2018). Re: What are the biggest challenges in bioprinting?. Retrieved from: https://www.researchgate.net/post/What_are_the_biggest_challenges_in_bioprinting/5b771a18979fdc4d2d1acd94/citation/download.

“The bioink composition and the physicochemical properties should match both for printing and the cell survival and functionality. If the bioink is printable at the above said parameters by maintaining the cells viable, then you can take further for differentiation, maturation in bioreactor and the studies you wish to do. The porosity of the bioprinted construct is very important to keep it functional, which takes care of the diffusion of nutrients, oxygen and metabolites. As a whole, various factors influence the bioprinting process and they are interlinked.”

Guru, Janani. (2018). Re: What are the biggest challenges in bioprinting?. Retrieved from: https://www.researchgate.net/post/What_are_the_biggest_challenges_in_bioprinting/5b92b8e884a7c1522d3630a0/citation/download.

Furthermore, bioinks need to be standardized to enable their use in various bioprinting applications. New models and standards must be developed to compare bioprinting processes, to evaluate the properties of different biomaterials, and to standardize the use of bioinks. Standardization of the evaluation of the bioink and of its properties is a critical need in bioprinting.

Additionally, much like different filaments are used to tackle different jobs in regular 3D printers, we want to develop a diverse catalog of biomaterials and bioinks for a wider range of applications.

While significant strides have been made in the field of bioprinting in the last decade, the commercial availability and clinical application of bioprinting has been limited by the lack of appropriate bioinks.

What’s the status quo?

In a recent paper published in the International Society of Biofabrication, Machine learning-based design strategy for 3D-printable bioink: elastic modulus and yield stress determine printability, researchers prepared bioinks, measured their rheological properties, printed cell-laden constructs, and built a machine learning to predict printability by ink composition. Their bioinks were formulated from collagen, fibrin, and hyaluronic acid. A rheometer was used to measure stress-strain relationships and understand the flow/deformation properties of the bioinks. All constructs were observed under a scanning electron microscope at various magnifications. Using extrusion-based bioprinting, it was possible to measure printability as layers of bioink were deposited. When human dermal fibroblasts were added to the bioprinted structures, a cellular viability test was conducted to evaluate the bioink’s biocompatibility and ability to support cell growth. ImageJ, an image processing program, was used to assess the shape fidelity of the printed pattern.

The relative least general generalization algorithm was employed for analysis, where 19 samples were used for modeling and 6 samples for the validation of the prediction algorithm. Despite the variety of bioink components and compositions, a universal relationship was found; results of their analysis indicated that printable bioink should have a high elastic modulus for improved shape fidelity and that extrusion is possible below the critical yield stress.

And while this paper demonstrates an important application of ML in controlling the mechanical strength of bioinks and meeting the requirements of high shape fidelity, the clinical use of these bioinks is still limited because this model doesn’t support predictive tissue engineering. Beyond optimizing bioink formulations to meet the printability requirements, there’s a need to evaluate the biological performance of the construct, such as cell viability, cell proliferation, and cell differentiation.

In another paper that was published this year, Hierarchical Machine Learning for High-Fidelity 3D Printed Biopolymers, it was possible to predict the shape fidelity of bioprinted constructs with alginate. Physical relationships between variables were used to link various parameters to print score. Their dataset was generated from high- and low-fidelity alginate constructs by varying the print input parameters and assessing the dimensional similarity between the original CAD designs and the constructs.

The variables of the bottom consist of system variables, also known as predictors, that are controlled in the lab. The physical variables on the middle layer describe the print system and depend on the bottom-layer predictors. The top layer is concerned with the system response and print score, calculated by comparing the fidelity of prints to the CAD file.

In a review published this year, Application of Machine Learning in 3D Bioprinting: Focus on Development of Big Data and Digital Twin, an outline of the future of 3D bioprinting and a forecast of the impact of ML is provided. As ML can be leveraged to drastically reduce the number of experiments by thousands of possible combinations, the impact of optimizing bioinks and bioprinting processes seems promising.

The rheological and mechanical properties of bioink, such as viscosity, elasticity, yield strength, and shear modulus, have been studied in detail using mathematical and analytical models, but the biological function of printed constructs has not been fully evaluated. While previous studies have performed a simplistic analysis of cell viability to assess the impact of biomaterials and the printing process on superior cellular functions, the biological performance and properties of bioinks need to be further analyzed.

Vitality is dedicated to the research and development of bioinks, contributing to the advancement of bioprinting and bringing the reality of lab-grown organs one step closer. We leverage machine learning to predict the role of biomaterials in biological functions and processes, specifically angiogenesis, the formation of new blood vessels from pre-existing ones.

Our pipeline includes 4 steps to predicting the relationships between bioinks and the biological function of the printed constructs.

We’re proposing a machine learning analytics model that can help scientists, researchers, and experts by revealing insights into the relationship between a bioink formulation and the function of a printed tissue.

With this model, we’ll be able to accelerate the R&D of bioink formulas and consequently reduce the costs.

Step 1: Formulating bioinks

To ensure a constant flow of oxygen and nutrients inside bioprinted tissues, vascular networks must be developed. And if we want to print vascular networks, a vascular bioink needs to provide a 3D environment that is conducive to cellular processes.

We’ll be focusing on formulating bioinks that favour the formation of new blood vessels from pre-existing ones, also known as angiogenesis. As the surrounding environment of cells is remodelled to grow blood vessels, cells can spread and vasculature can form, indicating angiogenesis. Because the biomaterials in the bioink mimic the surrounding environment, they must provide cells with specific physical and biochemical signals for vessel sprouting and splitting during tissue development in vivo.

The biomaterials used in our bioinks are:

  • Collagen
  • Elastin
  • Matrigel
  • Fibrin
  • Alginate
  • Chitosan
  • Agarose
  • Hyaluronic acid

Varying concentrations of these biomaterials can be used to develop bioinks that have different advantages and limitations.

Vitality’s platform develops, optimizes, and evaluates bioinks to find ones that promote angiogenesis.

Step 2: Bioprinting the tissues

Once we have developed a series of bioink formulations, it’s time to biofabricate multiple vascularized models and tissues for clinical transplantation. Each bioink formulation can be used to develop a vascularized tissue with distinct properties and processes.

Because extrusion or pressure assisted bioprinting is one of the most common methods used to promote angiogenesis and vasculature formation, we’ll be using extrusion-based bioprinters. In this technique, bioink is deposited using a pressure-based system to directly fabricate vascular-like networks.

Not only does the bioink depend on the intended mechanical properties and biological function, but its behaviour is also determined by unique printing parameters depending on the bioprinting technique. For the case of extrusion-based bioprinting, the printing outcome is dictated by varying factors, such as the printing resolution, nozzle diameter, apparent viscosity, nozzle length, and moving speed.

However, Vitality is open to trying various bioprinting techniques to fabricate 3D tissues with more desirable biological functions. Using the characteristics of the materials constituting bioinks and the bioprinting parameters, Vitality intends to predict the probability of successful and viable constructs. For angiogenesis, vascular formation could then be studied under a range of different conditions and techniques to improve constructs.

Step 3: Determining their properties

Determining the individual factors that control the complex process of angiogenesis is very challenging. We’re interested in properties, such as tubulogenesis (tube formation) and cellular proliferation and migration, to analyze and compare angiogenesis in printed tissues.

For example, if we want to quantitatively analyze the angiogenic response of tissues, we can count the number of vessels that are developed and measure the length, caliber, or density of the new vessels. To measure cell growth, we can observe cell cycle progression in the tissues. In other cases, we want to study particular mechanisms or processes in angiogenesis by monitoring the behaviour of cells in the controlled microenvironment.

Moreover, an ideal bioink should have essential physicochemical and physiological properties that lead to:

  1. The generation of tissue constructs with adequate mechanical strength and robustness, while retaining the mechanics that mimic tissue;
  2. Adjustable gelation and stabilization to aid the bioprinting of structures with high shape fidelity;
  3. Biocompatibility and biodegradability to avoid a toxic or immunological response and to mimic the natural microenvironment of the tissue;
  4. Suitability for chemical modifications to meet tissue-specific needs; and
  5. The ability for large-scale production with minimal batch-to-batch variation.

Angiogenesis assays are not only used to test the efficacy of pro- and anti-angiogenic agents and factors, but they also quantify the complex process. For the evaluation and quantification of angiogenesis, Vitality intends to run many assays and tests in order to obtain quantitative data for its dataset.

Step 4: Training a ML model

Using an analytical model such as a decision tree classifier, we can analyze angiogenesis in our printed constructs. Thanks to the quantitative data we collected in the previous step, our decision trees can analyze the impact of biomaterials and printing processes on the biological performance of our bioink. As ML models do best, we can then identify patterns in the data to determine bioinks that promote angiogenesis.

Why decision trees? Decision trees are simple but embraced for their unique high interpretability in comparison to black box models, where it’s difficult to discern exactly what processes the model went through to determine final outputs, and ability to run on limited data. Both of these are ideal for our R&D to find bioinks that stimulate angiogenesis in printed tissues.

Considered to be one of the best and most used supervised learning methods, it uses a decision tree as a predictive model to go from observations about an item, represented in the branches, to conclusions about an item’s target value, represented in the leaves. When deciding what to wear in the morning, you probably run conditional evaluations similar to decision tree learning in your head to factor in the weather, your daily activities, and more.

With that additional context in mind, here’s how Vitality leverages decision trees on data about bioinks. Consider the following table of data below:

From our data, the decision tree models will…

1. Infer specific conditions or decision rules;

2. Compare the value of specific conditions and the value of a datapoint to filter through the information; and

3. Reach conclusions about the degree of angiogenesis in a bioprinted tissue.

As specific conditions are learned, relationships between a bioink’s formulation and a printed tissue’s biological performance, specifically angiogenesis, can be discovered.

Next steps…

Invented in 2003, bioprinting is still in its infancy. Currently, bioprinting can be used to print tissue models to research drugs and pills.

But the 3D bioprinting of organs for the human body will be revolutionary. This technology has the potential to make medical care faster, more effective, and more personalized.

With 3D bioprinters in the clinical setting, we could bypass the problems associated with organ transplants, such as long donor waiting lists, immunorejection of the new organ, life-threatening medical complications, or the high price tag of an organ.

By increasing the likelihood of organ transplant acceptance, the cost of the operation could decrease and future surgeries due to complications could be avoided. While a kidney transplant costs about $300,000, a 3D bioprinter can cost as little as $10,000, and costs are expected to drop further as the technology evolves in the coming years. Implementing 3D bioprinting in our healthcare system will allow us to save millions of lives, both at low cost and risk. At the forefront of bioink development, Vitality is contributing to the advancement of bioprinting and bringing the reality of lab-grown organs closer to fruition by predicting the biological performance of bioprinted tissue.

🚀 Here’s what Balaji S. Srinivasan, the former CTO of Coinbase and a previous General Partner at Andreessen Horowitz, had to say about bioprinting:


We would like to thank Prof. Darcy Wagner, Nicholas Karaiskos, Caner Dikyol, Maria Stang, Jaci Bliley, Emma Davoodi, Maxwell Nagarajan, Prof. Chee Kai Chua, Andrew Hudson, Erica Comber, and Ankita Gupta for meeting with us and for providing helpful advice and feedback.

Thank you so much for reading our article, 3D Bioprinting: A Solution To The Organ Shortage Crisis?, and learning about bioprinters and bioinks!

Feel free to check out our work at Vitality:

🌐 Website

📄 One-pager

📼 Video