(Photo Credit above: Dr. Tim Anderson)

Want to write a piece for 3DHEALS Expert Corner? Email us: info@3dheals.com

Artificial Intelligence (AI) is the leading field of science nowadays.
Machines can be programmed to learn and complete tasks without human supervision. 

In other words, artificial intelligence is a self-learning system that can work with specific problems and make independent intelligent decisions (Chace, 2018). AI is applied to various fields of modern technologies and can be implemented to the manufacturing industry. The most innovative way of production is additive manufacturing, especially 3D printing.

The best advantage of 3D printing is unsupervised complex manufacturing. Various polymer, metal, and biomaterials are used in engineering applications mainly to create products with unique shapes, multifunctional compositions, reliability, and high quality. The 3D printing includes various techniques, but the most useful and time-tested method is Stereolithography (SLA). 

STEREOLITHOGRAPHY

SLA is the 3D printing process in which ultraviolet laser shoots on the surface of a tank filled with the photopolymer liquid. Energy, transferred by the laser to the material, activates curing reaction, which leads to the solidifying of the pattern traced on the photopolymer. Next, the build platform moves on the distance equal to the thickness of the one layer (Varotsis, 2018). The next layer is cured joining the previous layer. This procedure is repeated until the object is finished. However, every manufacturing method has its own problems and the risk of product failure.

REASONS OF THE SLA FAILURE 

What can cause failure during the SLA 3D printing process?
There are several reasons including:

• Photopolymer material failure 
• The UV-laser wavelength or scanning intensity change
• Curing (photopolymerization) reaction violation

These problems can be solved with the help of artificial intelligence.

INTRODUCTION TO PHOTOPOLYMERIZATION

Material, which is used in SLA, is called photopolymer and the name of the curing reaction is photopolymerization.

Steps of reaction (Terselius, 1998):

1. Radical formation: Radicals are formed under the exposure of the UV light
2. Propagation: energized photoinitiators create potential bonds
3. Termination: ends of the polymer chains face each other resulting in the rapid growth of the polymer chain

In the end, active groups are not able to create new bonds anymore, which means that the polymer chain grows process is terminated. 

PHOTOPOLYMERIZATION REACTION PROBLEMS AND SOLUTIONS

Resin needs the right amount of energy to achieve solidification. If the material receives not enough UV energy or the laser is spending less time for the curing process, then the print will not have appropriate characteristics to meet the application requirements (Jennings, 2018). The most optimal solution is to decrease the speed of printing by modifying laser settings (Jennings, 2018).

The second problem is associated with the lack of an appropriate amount of energy needed for the curing process (Jennings, 2018). However, an  AI system can modify the settings of the laser by increasing energy gradually to avoid abrupt changes during the 3D printing process.

Strength of the photopolymer can be increased by adjusting the laser in two ways (Ligon, et al., 2017): 

• Lowering the penetration depth
• Increasing the amount of energy

The penetration depth is the depth to which laser penetrates the photopolymer material layer and defined by (Ligon, et al., 2017):

Dp=Penetration depth  ε=Molar extinction coefficient  I=Photoinitiator concentration 

Penetration depth is reduced by adding light absorbers which change the formula (Ligon, et al., 2017):

I;A=Extinction coefficient  I=Photoinitiator concentration  A=Concentration of the absorber

UV absorbers increase the building time, but they improve resolution and strength. Penetration depth reduction is extremely important for the improvement of the resolution allowing the creation of thinner layers.

Critical exposure Ec is the energy needed to start the solidification reaction, which is defined as (Ligon, et al., 2017):

Ec=Critical exposure E0=Energy amount on the surface Cd=Curing depthDp=Penetration depth 

Formulas can be used to create an equation system for further implementation in the program of adjustment of the UV laser settings to control the material curing reaction. The aim is to achieve the right amount of energy which is needed to obtain successful solidification and meet the optimal properties of the final product.

WORKING CONCEPT OF AI

Possible concept of AI implementation in SLA includes layer scanning system, collection of the information, analysis, and solution to fix the failure without interruption of the 3D printing process (Bharadwaj, 2018). 

The failure is fixed at the earliest stage. The machine identifies divergence from the design and solves the problem as soon as the failure starts to appear. Therefore, the sensitivity of the scanning system and the reaction of the machine defining the errors should be developed.

The other way to fix production failure is to create a 3D printer that could remove material from the failed region. 

Next, AI should analyze the problem and find another way to build the part without changing final product properties.

The possible set of equipment for the creation of such technology includes:

• SLA 3D printer
• Sensors
• Scanning cameras
• Focused laser beam
• Machine learning algorithm
• Software

The software for the machine learning system is created with the machine coding which is very primitive, but complex at the same time. Machine code is the native code that a machine can read and execute to complete the specific task (Rouse, 2018). 

Additive manufacturing is based on the adding material layer by layer. 3D printers do not remove material layers. However, the failure happens in the printed layer which can be removed, predicted or enhanced.

In the first case, the material should be removed with high accuracy. The tool which can be used for such an application is the focused laser beam (Peels, 2017). Technology requires a laser that can move in three-dimensional space. The laser should be parallel to the printed layers to cut each layer from the side. Problematic layers should be analyzed by the scanning system and removed, interrupting the printing for a very short period of time, then the printing process should be continued. Despite the fact that the print can fail many times, the product will be successfully finished, and the machine will learn a lot from the problematic print at the end. The learning algorithm is enclosed, and the program will be repeated until reaching a successful result.

AI systems should be created to analyze large sets of data and make an instant decision, whereas humans are not able to react faster than the computer, just in a few seconds.

The system should repeat the process of material removing before the machine will learn how to fix the problem. This procedure improves the fixing process with prediction analysis and decreases the probability of the failure.

CONCLUSION

Machine learning improves the printing quality reducing risks of failure and manufacturing waste. The recycling in the field of additive manufacturing should be minimized with zero waste production with the implementation of AI. Also, there are a lot of possible ways that can be developed to protect the printing data and digital security system due to AI technologies.

REFERENCES

Chace, C., 2018. Artificial Intelligence and the Two Singularities. 1st ed. Boca Raton: CRC Press.

Varotsis, A. B., 2018. Introduction to SLA 3D Printing. [Online]
Available at: https://www.3dhubs.com/knowledge-base/introduction-sla-3d-printing#author
[Accessed 12 June 2018].

Terselius, B., 1998. Introduction to Polymer Science. 1st ed. Kristianstad: Arkitektkopia S. Niklasson AB.

Jennings, A., 2018. 3D Printing Troubleshooting Guide: 41 Common Problems. [Online]
Available at: https://all3dp.com/1/common-3d-printing-problems-troubleshooting-3d-printer-issues/
[Accessed 6 October 2018].

Ligon, S. C. et al., 2017. Polymers for 3D Printing and Customized Additive Manufacturing. Chemical Reviews, 117(15), pp. 10212-10290.

Bharadwaj, R., 2018. Artificial Intelligence Applications in Additive Manufacturing (3D Printing). [Online]
Available at: https://www.techemergence.com/artificial-intelligence-applications-additive-manufacturing-3d-printing/
[Accessed 12 September 2018].

Rouse, M., 2018. Machine code (machine language). [Online]
Available at: https://whatis.techtarget.com/definition/machine-code-machine-language
[Accessed 17 September 2018].

Peels, J., 2017. Comparison of Metal 3D Printing — Part Two: Directed Energy Deposition. [Online]
Available at: https://3dprint.com/182367/directed-energy-deposition/
[Accessed 6 August 2018].

FIGURES

Figure 1. Schematic of an SLA 3D printer (Varotsis, 2018).https://www.3dhubs.com/knowledge-base/introduction-sla-3d-printing/#author

Figure 2. Polymerization (MIT, 2018).https://formlabs.com/fr/blog/guide-ultime-impression-3D-stereolithographie-sla/

Figure 3. Relation between laser intensity, voxel size, and success of polymerization (Ligon, 2017).https://www.ncbi.nlm.nih.gov/pubmed/28756658

Figure 4. Representation of gel curing profiles. Laser penetrates deeply but only lightly cross-links the gel. (Jim H. Lee, et al., 2001).https://www.princeton.edu/~cml/assets/pdf/0112lee_curing.pdf

About the Author:

Reino Iuganson graduate student from Arcada university with a Bachelor’s Degree in Materials Engineering (Helsinki, Finland). He started his work with the injection molding industry in 2016 at the Plastoco company, then developed a prototype for a product design project in collaboration between Arcada university and Laerdal medical company using CAD/CAM and 3D printing. He is currently working on the project Artificial Intelligence in 3D printing.

Related Articles:

When Artificial Intelligence Meets 3D Printing

From Academia: 3D Printing and Robotics, Stem cell coated Implants, Decentralized Mitigation of Pandemics

Mixed Reality and 3D Printing, Imperfections to Perfection?

From Academia: 3D Printed Pills, 4D Printed Structure, and how to property 3D print Chocolate

From Academia: 3D-Printed Aligner, Bioprinting for Mouth Ulcer, Vertebroplasty Guides

From Academia: 3D Printing and Robotics, Stem cell coated Implants, Decentralized Mitigation of Pandemics

The post Artificial Intelligence and 3D Printing appeared first on 3DHeals.

Want to write a piece for 3DHEALS Expert Corner? Email us: info@3dheals.com

There are several main reasons that frequently motivate the innovators:

• Do cool things that could not be done before (e.g. flying, electricity, etc.).
• Make life better by a magnitude of a million times, etc. and not just minor increments (e.g. discovery of antibiotics).
• Save time, labor, and money that would recreate the industrial revolution and new economies.

While the general public is fascinated with both artificial intelligence and 3D printing as powerful new technological tools, and their potential future impact in healthcare, there has not been any known “killer applications” that utilize AI to improve existing 3D printing applications, in or out of healthcare/life sciences. The easy answer could be that both technologies are still relatively new, or that people who focus on AI applications are not necessarily interested in 3D printing, and vice versa. Or, maybe it’s because we simply do not have enough solutions to problems at hand.

Some of the proposed ways AI can improve 3D printing include the following [1-7]:

• Improve prefabrication design process
• Defect/Failure Detection
• Real-Time 3D printing Control/Failure compensation
• Predictive Maintenance/Inventory
• Workflow (Cost) optimization
• Chemical reaction/photopolymerization using ML-based algorithm to maximize control (chemicals and energy input)

There is an interesting analogy that I came across from professor Hiroya Tanaka, [2] with the following image(Figure 1). This shows that the subject “3D printing” has the visible physical components (tip of the iceberg) and the much larger invisible components in the realm of software, including data science, advanced 3D modeling, 3D object storage and retrieval, and AI/ML/Deep learning. While this is in accordance with the belief that “software eats the world” by the Silicon Valley, I would argue that all of these components will be equally important to the achieve the theoretical promises 3D printing as a successful manufacturing alternative.

That said, it is still helpful to do a brief review of where we are in terms of the intersection of these two technologies. Hopefully, this article can inspire interesting discussions, and even better, some new startups that Pitch3D can host very soon.

Artificial Intelligence/Machine Learning/Deep Learning

Artificial intelligence is an “intelligence” that is demonstrated by machines, which can perceive its environment and take actions to maximize its chance of success through the “learning” and “problem-solving” process. Machine learning is the scientific study of algorithms and statistical models that computers use to perform a specific task without human instructions, relying on patterns and inference instead. There are unsupervised ML (no human input) and supervised ML (human input). Finally, deep learning, also known as hierarchical learning, is based on artificial neural networks. There are also supervised and unsupervised DL.

The relationships among the concepts of artificial intelligence, machine learning, and deep learning (using artificial neural networks) are best demonstrated in the following diagram. (There are more sub-categories within each of these concepts that interested readers can easily find on the internet.)

The Problems

It is my theory that inventors can be lucky, but the inventions are never accidental. Inventions that changed human history (e.g. robots, computers, 3D printers, microbiology) are results of the continuous search for answers over long periods of time, from different perspectives and angles, and sometimes only after thousands of years.

The current status of healthcare applications using 3D printing is not so favorable because of several reasons:

1. 3D printing is still expensive, not just from the hardware and material cost, but also labor cost, and waste due to print defects and failures.
2. Lack of efficient and affordable design software. This is, in particular, a problem for the healthcare sector.
3. 3D printing is unable to achieve affordable (customized) mass production due to workflow challenges.
4. Lack of good quality control processes and tools, especially for the heavily regulated healthcare sectors.

The list can go on.

However, challenges also present opportunities, and AI/ML seem to be potential solutions to these worthy problems because AI/ML do somethings better than humans in many ways:

• Computers are able to process large amounts of data, learn, and implement actions in a more consistent fashion.
• Computers require little resources to function (i.e. electricity, minimal to no need for human operation).
• Computers can function well even in a toxic or harsh environment. (e.g. high temperature, toxic fumes)
• “Skillset” (algorithms) can be more rapidly “learned” and disseminated in a consistent way than human learning.
• Computers can store and retrieve large amounts of information almost instantaneously.

That said, creating the right AI/ML algorithm to 3D printing is no easy task because of the following:

• Successful AI/MI for the 3D printing process requires extensive knowledge of the specific 3D printing technologies, including but not limited to the design process, control of machine components, material science, post-processing. For example, the strategies behind optimizing the SLA based 3D printing process [1] will be very different from laser sintering metal 3D printing. [4]
• Finding high-value problems based on the end goal of production.  Either it is focused on reducing wasted time or precious materials, or ensuring end product mechanical properties that could result in serious clinical outcomes.  [1]
• Data collection. For example, for 3D printed anatomical models, a good AI/ML product focusing on optimizing the segmentation process will significantly decrease the bottleneck effect of entering the field for many hospitals and clinics. However, the lack of such a product is because of a lack of enough training datasets. [3]
• Intrinsic limitations of existing monitoring systems. Researchers are currently using either photos or videos to train their AI/ML algorithms. Smoothly incorporating the monitoring systems without interrupting the printing process will be challenging. [1, 4, 5] However, such integration will be required to achieve “real-time” 3D printing monitoring and subsequent “fixing” or “failure compensation” of the prints. [1]
• Forming a successful team that can tackle problems along the entire 3D printing process from design to final product requires a group of people from different disciplines. [1] For example, to accomplish real-time SLA 3D printing support modification[Figure 3], Dr. Iuganson proposed in his thesis a team structure that would include the following:

1. 3D printing engineer
2. Sensors technician
3. Automation CT engineer
4. Laser and optics engineer
5. Machine learning specialist develops a set of steps for correction of the printing and generating supports if the problem is predicted.
6. Data scientist creates a code for the machine to change the design structure and generated supports
7. AI research scientist analyses and implements the information in the AI system to add a new feature of real-time control over the design and supports.

Now, imagine that everyone on this team has to understand what is going on and can also communicate effectively with one another!

The Solutions

Solutions seem to be coming, but just not here yet.

GE Additive, Sculpteo, Autodesk, and many more all appear to actively develop AI/ML-based solutions to optimize various value points of the 3D printing process. [6] Align Technology just announced a new AI/ML-based visualization/predictive tool SmileView based on 60 million patient datasets. (Align is also actively hiring AI/ML engineers.) [8] It is my hope that perhaps more entrepreneurs can venture into this exciting intersection of two powerful emerging technologies.

Perhaps this IS where we will find the “killer app” in 3D printing.

References:

1. Artificial Intelligence in 3D Printing (Thesis by Dr. Reino Iuganson)
2. Deep Learning for Advanced 3D Printing
3. The potential for machine learning algorithms to improve and reduce the cost of 3-dimensional printing for surgical planning (Trevor J. Huff, Parker E. Ludwig & Jorge M. Zuniga) ISSN: 1743-4440 (Print) 1745-2422 (Online) Journal homepage: https://www.tandfonline.com/loi/ierd20
4. Machine Learning “Fixes” 3D-Printed Metal Parts—Before They’re Built
5. Automated Process Monitoring in 3D Printing Using Supervised Machine Learning
6. Artificial Intelligence Applications in Additive Manufacturing (3D Printing)
7. Multi-Objective Optimization Design through Machine Learning for Drop-on-Demand Bioprinting
8. New Dental Product: SmileView from Align Technology

Related Articles:

From Academia: 3D Printing and Robotics, Stem cell coated Implants, Decentralized Mitigation of Pandemics

Five Reasons Cybersecurity Will Play a Critical Role in 3D Printing in Healthcare – Part 1

The Augmented Mind: How AR/VR will empower 3D Printing technology in bettering the real world.

Decentralized Healthcare — Part I. Cooler than Bitcoins, But What Is It?

The post When Artificial Intelligence Meets 3D Printing appeared first on 3DHeals.