THE USA JOURNALS
THE AMERICAN JOURNAL OF ENGINEERING AND TECHNOLOGY (ISSN
–
2689-0984)
VOLUME 06 ISSUE06
24
https://www.theamericanjournals.com/index.php/tajet
PUBLISHED DATE: - 26-06-2024
https://doi.org/10.37547/tajet/Volume06Issue06-04
PAGE NO.: - 24-31
HARNESSING ARTIFICIAL INTELLIGENCE FOR REAL-
TIME QUALITY ASSURANCE IN MEDICAL DEVICE
MANUFACTURING
Phani Chandra Barla
Principal Quality Engineer, Senseonics Inc 20451 Seneca Meadows Parkway, Germantown, MD
20876-7005
Dr. Laina Karthikeyan
Professor, Department of Biological Sciences, New York City College of Technology, 285 Jay St,
Brooklyn, NY 11201
INTRODUCTION
In the field of medical device manufacturing,
ensuring the quality and reliability of the products
is of utmost importance to safeguard patient safety.
Defects or malfunctions in medical devices can
have severe consequences, ranging from
compromised patient care to potential harm or
even loss of life. Therefore, the detection of defects
during the manufacturing process is critical to
prevent faulty devices from reaching the market.
Traditional methods of defect detection in medical
device manufacturing often rely on manual
inspection by human operators. While these
methods have proven effective to some extent, they
are inherently limited by human subjectivity,
fatigue, and the potential for human error.
Moreover, the complexity and intricacy of modern
medical devices make it increasingly challenging
for human inspectors to identify subtle defects or
anomalies. To overcome these limitations and
enhance defect detection capabilities, artificial
intelligence (AI) has emerged as a transformative
RESEARCH ARTICLE
Open Access
Abstract
THE USA JOURNALS
THE AMERICAN JOURNAL OF ENGINEERING AND TECHNOLOGY (ISSN
–
2689-0984)
VOLUME 06 ISSUE06
25
https://www.theamericanjournals.com/index.php/tajet
technology in the manufacturing industry. AI-
powered defect detection systems leverage
advanced
algorithms
and
computational
techniques to analyze large volumes of data and
automatically identify and classify defects in
medical devices. The application of AI in defect
detection offers several advantages over
traditional approaches. AI algorithms can process
vast amounts of data with speed and precision,
enabling the detection of even the most subtle
defects that may go unnoticed by human
inspectors.
Additionally,
AI
systems can
continuously learn and improve their defect
recognition abilities through machine learning
techniques, ensuring higher accuracy and
adaptability over time.
In today’s rapidly evolving pharmaceutical and
medical device industries, maintaining high
standards of quality assurance is paramount. As
technology continues to advance, innovative
solutions such as artificial intelligence (AI) and
automation are revolutionizing the validation
process. The world is being changed at a dizzying
rate by artificial intelligence (AI). Artificial
intelligence (AI) has entered many facets of our
daily lives, from self-driving vehicles to virtual
assistants. The quality field is one area where AI
has the potential to make a big difference. Utilizing
AI for quality management allows firms to detect
and resolve quality issues early on, guaranteeing
that products and services live up to customer
expectations.
Artificial intelligence (AI) is a subfield of computer
science that focuses on the application of
sophisticated algorithms and processing abilities
to the problem of extracting useful insights from
large datasets, with a particular emphasis on the
context of the fourth industrial revolution. With the
advent of the Internet of Things (IoT), production
will see a marked improvement in efficiency,
quality, management ease, and transparency. How?
By utilizing Industry 4.0-based smart factories that
integrate physical and cybertechnologies. One of
the most important ways to make factory
automation systems smarter is to use sensors and
AI [3,4].
In this new era of Digital Transformation, quality is
no longer about raw data, but the way we process
the data and the insights we extract from it. There
is no doubt that the combination of Artificial
Intelligence (AI) and Quality Management is not
just a dream but is already reshaping the way we
do business today. It is a game-changer.
The expansion of numerous sectors and the
improvement of national economies are highly
dependent on the development and improvement
of sensor technology, particularly as it pertains to
Industry 4.0. To collect data and put it to good use,
manufacturing organizations and supply chains
need access to modern, inexpensive sensor
technology. The most common kinds of sensors
include those that measure location, flow,
temperature, pressure, and force. Motorsport,
healthcare, manufacturing, the military, and
agriculture are just a few of the many industries
that depend on them frequently. Improving
productivity through the use of automation is the
goal of Industry 4.0 [5,6]. Recent advances in
healthcare using AI methods have sparked a heated
debate about whether AI doctors will one day
replace human doctors [7,8].
LITERATURE REVIEW
The possibility of computers fully replacing human
doctors is quite improbable for the near future.
However, AI might substantially help doctors make
better clinical decisions. A few medical fields, like
radiology, may even be able to do away with human
judgment altogether thanks to AI [9,10]. The recent
successful deployment of AI in healthcare has been
made possible by the fast improvement of big data
analysis techniques and the increasing accessibility
of healthcare data. In order to answer important
clinical issues and extract useful information from
the massive amounts of data, sophisticated AI
algorithms are being used. With this information,
clinical decision-making can be improved [11,12].
The ever-evolving manufacturing, medical device,
pharmaceutical, and food and beverage industries
must maintain the highest standards of quality at
all times [13,14]. As a game-changing technology,
artificial intelligence (AI) is revolutionizing the
quality management system (QMS) industry.
Quality Control (QC) and Quality Assurance (QA)
are being transformed by technology.
THE USA JOURNALS
THE AMERICAN JOURNAL OF ENGINEERING AND TECHNOLOGY (ISSN
–
2689-0984)
VOLUME 06 ISSUE06
26
https://www.theamericanjournals.com/index.php/tajet
To ensure that products and services meet or
exceed customer expectations and established
standards, quality management is essential in
many sectors [15]. As technology advances, the
integration of artificial intelligence (AI) is
revolutionizing traditional business practices. With
the introduction of AI, quality assurance (QA) has
undergone a radical transformation [16]. It would
be foolish to ignore the substantial benefit that AI
provides in terms of streamlining and improving
testing procedures. To gain a substantial
competitive edge, businesses need to know how to
incorporate artificial intelligence (AI) into their
testing procedures. Quality assurance (QA) teams
may improve their efficiency by moving away from
manual testing and toward more advanced
autonomous testing technologies, which are
discussed in this article, which offer a thorough
study of AI's possibilities in QA [17].
Intelligent 3D printing of personalized
medicines
There has been a digital revolution in the past
quarter of a century, beginning with the
introduction of wireless internet and continuing
with the ubiquitous usage of smartphones around
the world, cloud computing, and social media.
These revolutionary inventions were initially
designed and used by intelligent humans.
The modern catalog of pharmaceutical 3D
printing technologies
Printing nearly every kind of medication is now
possible because to pharmaceutical 3DP, which is a
combination of separate technologies. To grasp the
potential of AI in pharmaceutical 3DP, one must
first acknowledge the diversity and difficulties of
the available approaches, as well as their benefits
and drawbacks when applied to specific
medications and excipients.
Alternative
optimization
techniques
to
machine learning in 3D printing
The intricacy of pharmaceutical 3DP means that
developing new drugs through trial and error is not
only inefficient but also runs the risk of producing
subpar results. Decisions about printing method
and formulation components are examples of
macro-level choices that must be considered while
creating a new 3DP drug.
METHODOLOGY
Data-Driven Decision-Making
Quality assurance technologies powered by AI can
instantly sift through mountains of data, revealing
patterns and trends that can guide business
choices. As a result, quality management becomes
more proactive, which speeds up the process of
finding and fixing problems. Utilizing AI-powered
quality
assurance
technologies,
a
car
manufacturing company, for instance, can examine
real-time data from multiple sources, including
production lines, customer feedback, and historical
records. The business may improve its quality
management strategy and make better judgments
by seeing patterns, insights, and trends and acting
swiftly to resolve any problems that may arise.
Automated Quality Control:
Businesses may automate quality control
operations with the help of AI algorithms, which
reduces human error and increases productivity.
This guarantees a higher degree of accuracy in
finding defects and inconsistencies in goods and
services while simultaneously saving time and
resources. Using AI algorithms, a pharmaceutical
corporation may automate the production process
inspection of pills and capsules with a precision of
99.999%. To provide safer and more dependable
pharmaceuticals, the corporation must decrease
human error and increase efficiency to guarantee a
better degree of precision in spotting defects and
inconsistencies.
Predictive Quality Management:
Artificial intelligence (AI) applied to quality
management can look at past data and find trends
that could cause difficulties in the future. Ensuring
a constant level of quality and boosting client
happiness are both made possible when
organizations handle potential issues before they
become significant. Applying AI to quality control
at a food processing plant, for instance, could help
spot trends in past data that could indicate
impending problems with product quality. With
this kind of planning, the plant can head off
problems before they even start, which keeps
THE USA JOURNALS
THE AMERICAN JOURNAL OF ENGINEERING AND TECHNOLOGY (ISSN
–
2689-0984)
VOLUME 06 ISSUE06
27
https://www.theamericanjournals.com/index.php/tajet
quality consistent and makes customers satisfied.
AI in Quality Management: A Case Study
With the help of an AI-driven quality management
system, a prominent car manufacturer was able to
raise the bar on product excellence. Data from
production lines, suppliers, and customer feedback
were all inputted into the system by use of artificial
intelligence algorithms. Potential flaws and quality
process aberrations were detected by the system,
allowing for the implementation of preventive
steps. Manufacturers can enhance their quality
control procedures with the help of the system's
real-time monitoring and continuous improvement
features. Manufacturers saw a 30% drop in quality-
related expenses and a 20% uptick in happy
customers because of it.
Manufacturing Excellence with AI:
Maintaining a high level of consistency and
accuracy is critical in the manufacturing industry.
When it comes to improving and optimizing each
stage of the production process, AI is important.
The next generation of quality management
systems, driven by artificial intelligence, can help
manufacturers achieve quality control like never
before. Computer programs powered by artificial
intelligence can instantly sift through mountains of
data in search of trends and outliers that a human
eye could miss. Early defect discovery, less waste,
and higher overall product quality are all results of
its pattern recognition and anomaly spotting
capabilities.
On top of that, cloud-based quality management
systems make it easy for diverse production units
and regions to work together and share data. This
guarantees that no matter how far apart locations
are, quality requirements will always be satisfied.
The end product is an AI-powered, highly
integrated manufacturing environment, that takes
quality to new levels.
Medical Devices: A Leap Forward in Precision
Accuracy is of the utmost importance in the
medical devices sector. Artificial intelligence is
changing the medical device manufacturing
industry with its large dataset processing and
learning capabilities. Quality management systems
powered by AI guarantee that all stages, from
design and prototyping to production and post-
market surveillance, are carried out to the highest
standards of quality.
Medical device producers must adhere closely to
the FDA's strict rules. With the help of AI, not only
are these regulatory standards more easily met, but
the clearance procedure is also faster and more
accurate. Artificial intelligence (AI) allows medical
device companies to speed up product
development without sacrificing quality or safety.
Pharma
and
Lifesciences:
Accelerating
Innovation and Compliance
AI is revolutionizing the pharmaceutical and life
sciences industry, which is characterized by a close
relationship between innovation and compliance.
There are a lot of moving parts and strict quality
controls involved in creating and releasing new
medications to the market. AI-powered quality
management systems streamline this process by
taking over mundane, repetitive jobs, freeing up
researchers and scientists to concentrate on new
ideas. Furthermore, AI ensures compliance in an
area where following FDA standards is absolutely
essential. The likelihood of non-compliance issues
is greatly reduced since it guarantees that every
phase of drug research and manufacture conforms
with regulatory criteria. The approval procedure is
accelerated and pharmaceutical companies'
reputations are protected.
Food and Beverages: Ensuring Safety and
Consistency
In order to ensure the safety and happiness of
consumers, the food and beverage business must
maintain consistent quality. Precision and real-
time monitoring are essential in this sector, and AI
offers both to QMS. Using AI, we can guarantee that
every stage, from sourcing raw materials to
packaging and distribution, is done to the highest
quality standards.
With the use of cloud-based quality management
system solutions, businesses can monitor the
whereabouts of all their products and ingredients
as they move through the supply chain. In the event
of a quality issue, this traceability is vital for
avoiding extensive recalls and protecting the
THE USA JOURNALS
THE AMERICAN JOURNAL OF ENGINEERING AND TECHNOLOGY (ISSN
–
2689-0984)
VOLUME 06 ISSUE06
28
https://www.theamericanjournals.com/index.php/tajet
reputation of the brand.
The application of AI in QA also includes predictive
maintenance for production machinery, which
lessens the likelihood of unanticipated failures that
can lower product quality. This preventative
method improves operating efficiency and
guarantees product uniformity.
The Power of Cloud-QMS
The AI-driven quality management system (QMS)
cloud revolution is supported by cloud-based
solutions. With this unified platform, stakeholders
can save, analyze, and collaborate on data in real-
time, regardless of their location. Businesses with a
worldwide supply chain or numerous production
facilities will benefit greatly from this degree of
interconnection.
Data security and integrity are guaranteed by the
cloud-QMS strategy, which also improves
cooperation. Companies may rest certain that their
critical quality data is safeguarded from breaches
or loss, thanks to strong encryption and backup
measures. Because of this, both consumers and
regulatory agencies have more faith in the sector.
RESULTS AND DISCUSSION
The rapid identification of respiratory pathogens,
such as parainfluenza virus, influenza virus,
respiratory syncytial virus and/or adenovirus,
especially during a winter period, is of great
importance for the prevention of nosocomial
spread, for monitoring infected patients, and for
improved clinical management. Rightfully so, there
is added pressure for clinical laboratories to be
able to provide rapid and sensitive applications for
the testing of these respiratory pathogens,
particularly for immunocompromised individuals,
the young and the elderly.
A rapid diagnostic test should be able to provide
the clinician with a result within a short time frame,
to allow for prompt and appropriate therapeutic
action. Reverse transcriptase-PCR assay (RT-PCR)
is the gold standard for diagnosing viral infections
including the COVID-19 because it can identify the
infectious agent, severe acute respiratory
syndrome coronavirus 2 (SARS-CoV-2). An unusual
need for polymerase chain reaction (PCR) testing
in diagnostic laboratories across the globe was
noted as COVID-19 rapidly propagated from
person to person during the pandemic. For the
purpose of reducing the workload of healthcare
and laboratory professionals, several AI-aided
detection models have been created to quickly and
accurately diagnose SARS-CoV-2 using RT-PCR. A
deep learning model called qPCRdeepNet was
proposed by Alouani et al. It uses a deep
convolutional neural network to analyze
fluorescence readings taken during COVID-19 RT-
PCR.
The goal of the model is to increase test specificity
and detect false positive results. Along with this,
Lee et al. created a deep learning model using the
long-term short memory (LSTM) technique. This
model was fed raw data of fluorescence levels from
each of the 40 RT-PCR test cycles.
In their analysis of patients' clinical data, blood test
findings, and chest CT imaging data, the authors
found that RT-PCR diagnosis time was decreased.
Similarly, the authors automatically classified RT-
PCR data as positive, weak-positive, negative, or re-
run using an AI-based detection and classification
system for COVID-19 RT-PCR diagnosis that
utilized fluorescence data and amplification curves.
Additionally, Villarreal-González et al. classified
4230 RT-PCR curves from patient data into
positive, early, no, and abnormal amplifications
using various ML models. The top model achieved
rapid diagnosis while reducing false positives by
detecting atypical profiles in PCR curves caused by
contamination or artifacts. It has also been utilized
to detect SARS-CoV-2 variations.
During the pandemic, researchers used RT-PCR
data to train an ML algorithm that relies on the
number of cycles (cycle threshold, Ct). The
technique's proponents posited that different virus
variants can be detected by identifying patterns in
THE USA JOURNALS
THE AMERICAN JOURNAL OF ENGINEERING AND TECHNOLOGY (ISSN
–
2689-0984)
VOLUME 06 ISSUE06
29
https://www.theamericanjournals.com/index.php/tajet
Figure 1. Number of approved (USA) and CE-marked (Europe) AI/ML-based medical devices
between 2015 and 2019
the Ct values of PCR-positive samples. The use of
laser-scribed graphene (LSG) sensors in
conjunction with a biosensing platform for gold
nanoparticles (AuNPs) was also employed by
Beduk et al. to detect SARS-CoV-2 variations using
a Dense Neural Network (DNN) algorithm.
Streamlining the interpretation of RT-PCR tests
and reducing the need for human intervention in
laboratory practice have both been greatly aided
by the AI-driven SARS-CoV-2 diagnosis. From 2015
to 2019, figure 1 also displayed the number of
AI/ML-based medical devices that were approved
in the US and CE-marked in Europe. See the
healthcare industry's use of AI from 2021
–
2030 in
Figure 2. Several aspects of healthcare operations
can greatly benefit from the use of AI, such as the
management of chronic diseases, the automation
and optimization of workflows, the early detection
of risks, and the improvement of patient care.
THE USA JOURNALS
THE AMERICAN JOURNAL OF ENGINEERING AND TECHNOLOGY (ISSN
–
2689-0984)
VOLUME 06 ISSUE06
30
https://www.theamericanjournals.com/index.php/tajet
Figure 2. AI in healthcare from 2021 to 2030
CONCLUSIONS
Applications of artificial intelligence are inevitably
going to become a part of modern healthcare.
These applications have a high potential to assist
caretakers and decision-makers in the areas of
laboratory and imaging diagnosis, antimicrobial
stewardship,
discovery
of
antimicrobials,
microbiome-based translational interventions,
infectious disease surveillance, prediction, and
prevention. The widespread digitization of medical
records, which has made data more accessible, as
well as the advancements in computer power have
been extremely helpful and will continue to be
essential for the research and development that
will take place in the sector in the future. Despite
the fact that artificial intelligence is usually
considered to be a danger to "common"
employment, its incorporation into healthcare
should be viewed as an opportunity for enhanced
patient care and infection management, higher
survival, improved staffing and resource allocation,
and decreased costs in healthcare systems.
REFERENCES
1.
Harrer S, et al. Artificial intelligence for clinical
trial design. Trends Pharmacol. Sci. 2019;
40:577
–
591. doi: 10.1016/j.tips. 2019.05.005.
[PubMed] [CrossRef] [Google Scholar].
2.
Li BT, et al. Reimagining patient-centric cancer
clinical trials: a multi-stakeholder international
coalition. Nat. Med. 2022; 28:620
–
626. doi:
10.1038/s41591-022-01775-6.
[PubMed]
[CrossRef] [Google Scholar]
3.
Dagenais S, et al. Use of real-world evidence to
drive drug development strategy and inform
clinical trial design. Clin. Pharmacol. Ther.
2022; 111:77
–
89. doi: 10.1002/ cpt.2480.
[PMC free article]
THE USA JOURNALS
THE AMERICAN JOURNAL OF ENGINEERING AND TECHNOLOGY (ISSN
–
2689-0984)
VOLUME 06 ISSUE06
31
https://www.theamericanjournals.com/index.php/tajet
4.
Liu R, et al. Evaluating eligibility criteria of
oncology trials using real-world data and AI.
Nature. 2021; 592:629
–
633.
5.
Ithapu VK, et al. Imaging-based enrichment
criteria using deep learning algorithms for
efficient clinical trials in mild cognitive
impairment. Alzheimers Dement. 2015;
11:1489
–
1499. doi: 10.1016/j.jalz. 2015.01.10.
[PMC free article] [PubMed] [CrossRef] [Google
Scholar]
6.
Ezzati A, et al. Machine learning predictive
models can improve efficacy of clinical trials for
Alzheimer’s disease. J. Alzheimers Dis. 2020;
74:55
–
63. doi: 10.3233/JAD-190822. [PMC
free article] [PubMed] [CrossRef] [Google
Scholar]
7.
Mohan A, et al. A machine-learning derived
Huntington’s disease progression model:
insights for clinical trial design. Mov.
8.
de Jong J, et al. Towards realizing the vision of
precision medicine: AI based prediction of
clinical drug response. Brain. 2021; 144:1738
–
1750. doi: 10.1093/brain/awab-108. [PMC
free article] [PubMed] [CrossRef] [Google
Scholar].
9.
Hassanzadeh H, et al. Matching patients to
clinical trials using semantically enriched.
10.
Alexander M, et al. Evaluation of an artificial
intelligence clinical trial matching system in
Australian lung cancer patients. JAMIA.
11.
Haddad T, et al. Accuracy of an artificial
intelligence system for cancer clinical trial
eligibility screening: retrospective Pilot Study.
JMIR Med. Inform. 2021;9:e27767.
12.
Beck JT, et al. Artificial intelligence tool for
optimizing eligibility screening for clinical
trials in a large community cancer center. JCO
Clin. Cancer Inform. 2020; 4:50
–
59. doi:
10.1200/CCI.19.00079. [PubMed] [CrossRef]
[Google Scholar]
13.
Kim J, et al. Review of the performance metrics
for natural language systems for clinical trials
matching. Stud. Health Technol. Inform. 2022;
290:641
–
644. [PubMed] [Google Scholar]
14.
Unlearn works with pharma sponsors, biotech
companies,
and
academic
institutions.
https://www.businesswire
.com/news/home/20220419005354/en
15.
European Medicines Agency releases for public
consultation its draft policy on the publication
and access to clinical-trial.
16.
Abramson A, et al. A flexible electronic strain
sensor for the real-time monitoring of tumor
regression. Sci. Adv. 2022; 8:eabn6550. doi:
10.1126/sciadv.abn6550. [PMC free article]
[PubMed] [CrossRef] [Google Scholar]
17.
Warnat-Herresthal S, et al. Swarm learning for
decentralized and confidential clinical machine
learning. Nature. 2021;594:265
–
270. doi:
10.1038/s41586-021-03583-3. [PMC free
article] [PubMed] [CrossRef] [Google Scholar
18.
18.
Lipson, S.M., Gordon, R.E., Ozen, F.S.,
Karthikeyan, L., Stotzky, G. Effect of cranberry
and grape juices on tightjunction function and
structural integrity among rotavirus-infected
money
kidney
epithelial
cell
culture
monolayers. Food Environ. Virol. 3, 2011; 46-
54.
