Artificial intelligence

Presented by,
Pratiksha C Chandragirivar
M pharma 2nd sem
Dept. of pharmaceutics
HSK COP Bagalkot
Facilitated to,
Dr. Laxman Vijapur
Assistant professor
Dept. of pharmaceutics
HSK COP Bagalkot
1

01
02
03
04
05
06
Basics of Artificial
Intelligence
Introduction
Artificial Inteligence in
Pharmaceuticals
Robotics
07
Future potential
Challenges faced by AI
References
2

 It is intelligence, demonstrated by machines, unlike natural intelligence displayed
by humans
 This field was founded by John McCarthy in 1956
 An Artificial Intelligence (AI) runs on clear cut set of instruction called algorithm
 In pharmaceutical field we generally use models like Artificial neural network
(A.N.N.) and genetic algorithm (GA) etc
Artificial Intelligence in Drug
Delivery Systems
3

Artificial
intelligence
Machine
learning
Artificial
neural
network
Deep
learning
Any technique that enable machines to
solve task in a way like human do
Algorithms that allow computer to
learn from example without being
explicitly programmed
Brain-inspired machine learning models
A subset of machine learning which uses
ANN as models and automatically builds
hierarchy of data representations
5

Automated
pharmaceutical
manufacturing
Robotics
Interpret data in form
which can be understood
by human
Contd…
Delivery Systems
6

Artificial Intelligence
Machine
Language
A.N.N. Cognitive computing Natural Language
processing
Supervised Unsupervised Reinforcement
Delivery Systems
7

Machine Language Neural Network
 Machine learning algorithms almost always
require structured data
 Results are often unsatisfactory without
human intervention
 They relatively require less data than deep
learning
 Deep learning can work on unstructured
 Deep learning networks do not require
human intervention
 Deep learning requires much more data
than a traditional machine learning
algorithm
5
Delivery Systems
8

 The pharmaceutical product development process is a multivariate optimization problem. It
involves the optimization of formulation and process variables.
 A desirable formulation for one property is not always desirable for the other
characteristics.
 ANN represents a promising modeling technique especially for data sets having the kind of nonlinear
relationships, which are frequently encountered in pharmaceutical processes.
 ANNs can identify and learn correlative patterns between input and output data pairs. Once
trained, they may be used to predict outputs from new sets of data.
 AI is currently being used in range of pharmaceutical related fields like – drug discovery,
development of drug delivery system etc.
Artificial Intelligence in
Drug Delivery Systems
9

Advantage Limitations
 It can repeatedly perform same,
monotonous task without getting tired.
 If coded properly than it causes less error
than human.
 It can analyze huge amount of data.
 Can automatically manage scientific data
and records without error.
 High cost Creation of artificial intelligence
requires huge costs as they are very complex
machines.
 Their repair and maintenance require huge costs.
 Highly erroneous results obtained when amount
of data present is less.
 AI lacks in practical common sense (visual common
sense reasoning and common sense knowledge)
making them harder to use in certain scenarios.
7
Delivery Systems
10

Drug discovery and drug development
Pre-formulation
Formulations development
Quality by design (QBD)
Clinical trials
1.
2.
3.
4.
5.
8
Delivery Systems
11

In finding new
molecular target
• Predictive models helps in
rationalizing drug discovery process.
In finding hits and leads
•
• QSAR modelling helps in identifying
Leads, based on relation between
Structure and Biological properties.
In drug repurposing
• ANNs could classify complex drug
action mechanisms on the pathway
level.
In automated drug
design
• The algorithms performs in-depth
analysis of the chemical synthesis
data to give optimal synthetic
pathways.
McInnes C et al, 2007, Curr. Op. Chem.Biol, 494-520
1. Role of artificial intelligence in drug designing
Delivery Systems
12

 The AI enabled computational tools and systems can guide in selection of optimal
molecule to advance into next stage.
 The last stages of clinical studies exhibit a higher elimination of candidate molecules.
This is due to lack of understanding of lack of information in the early stages of drug
discovery.
 A Rule of “5Rs” which are the most crucial factors determining the success or failure of
drug Research and Development depends. These are - Right Target, Right Tissue, Right
Safety, Right Patient and Right Commercial Potential.
 With the help of virtual screening and in-silico ADMET prediction on can
obtain many useful information's like –
o physico-chemical property
o Better Drug-target model
o invivo information's
2. Artificial intelligence in pre-formulation studies
Delivery Systems
13

 Formulation involves multiple disciplines like physical chemistry, physics at micro to macro level
properties and setting coordination in them makes formulation development a tricky and specialized
task on which the success of the company and the drug product depends.
 The AI based methods when mainstreamed can help maximize the output while reducing the
operational cost.
Switching from
lab/ pilot plant
to industrial
level
production
lead changes
in various
parameters.
This leads to
variation in
production
and risk of
wastage of
time and
resources.
AI can analyze
this factors
beforehand
and can
optimize this
process.
3. In formulation developments
Delivery Systems
14

 AI can identify all the critical quality attributes and critical process parameter.
 Traditionally, pharmaceutical manufacturing has been accomplished using batch processing with quality
control testing conducted on samples of the finished product.
 The US Food and Drug Administration (FDA) announced amendments to its “Current Good Manufacturing
Practices” (CGMP) in 2002 with an emphasis on establishing a 21st century outlook on pharmaceutical
manufacturing and to utilize advances in science and technology to establish a more systematic science and
risk-based approach to the development of pharmaceutical products.
 ANNs, GAs, and fuzzy logic are fundamental computer-based systems that are utilized to design
an experimental space for the required data during manufacturing processes. They can help in
identifying CQA and CQP .
4. Role of artificial intelligence in QbD
Delivery Systems
15

 Clinical trial failures are mainly due to poor recruiting and selecting techniques, as well as an
inability to effectively monitor patients.
 Selecting and recruiting patients to participate in clinical trials is often the main cause of trial
delays, with 86 percent of all trials failing to meet enrollment deadlines.
 AI tools can help enhance patient selection by reducing population heterogeneity, choosing
patients who are more likely to have a measurable clinical endpoint, and identifying a
population more capable of responding to treatment.
 It can Collect the data from multiple clinical trial centres simultaneously and analyze them.
 It Will assure authenticity, quality and integrity of data.
5. Role of artificial intelligence in clinical trial
Delivery Systems
16

 Drug delivery technologies modify drug release profile, absorption, distribution and
elimination for the benefit of improving product efficacy and safety, as well as patient
convenience and compliance.
 ANNs and fuzzy logic algorithm for designing of formulations helps to analyze the
effects of formulation components on the release characteristics and optimize drug
formulations.
 Findings have shown the influential role of the chemical compositions, molecular
volume, lipophilic-hydrophilic balance, length of the hydrocarbon chain, and
hydrocarbon volume of co-surfactants in delivery system. These parameters can
be analyzed for the creation of optimum environment.
Delivery Systems
17

 It has been observed that often, insufficient information is available to determine the
pharmacokinetics of a drug or which drug will have a desired effect for an individual patient.
 ANNs represent a novel model-independent approach to the analysis of pharmacokinetic
(PK)-pharmacodynamics (PD) data. ANNs have been shown to be flexible enough to predict
PD profiles accurately for a wide variety of PK-PD relationships.
 ANNs are useful for optimizing modified-release products including the solid dosage forms.
General neural network (GRNNs) and connectionist models help to predict drug stability and
release profile.
1. Evaluation and prediction of effectiveness of drug delivery method
Delivery Systems
18

 Various kind of AI like ANN, MLP has been to control and optimize release rate and
other factors for development of transdermal patches as well as implant
 For obtaining an appropriate drug release profile, the ability of
ANN model to determine the optimal levels of formulation
parameters, was evaluated.
 Besides providing shorter formulation design process, drug release profile from the
implant device and transdermal system are accurately modeled by ANN model and
the results were so close to the experimentally obtained values indicating the model
effectiveness.
2. Role of artificial intelligence in implants and trans-dermal system
Delivery Systems
19

3. Optimization of drug release profile of matrix tablet and
controlled release formulation using Elman Dynamic ANN
 The network consists of an input layer, two hidden layers and an output layer of
neurons. Input layer contains four neurons (information about weight ratio of matrix
forming material, compression force, solidity and tablets tensile strength).
 MLP neural network optimized for modeling of drug
release is generally used.
Artificial Intelligence in
Drug Delivery Systems
20

 2 layers of hidden layer consist of pre-optimized functions and weight.
 Weights in an ANN are the most important factor in converting an input to
impact the output. This is similar to slope in linear regression, where a weight is
multiplied to the input to add up to form the output.
 Whereas function defines a perticular set of quality to be observed and
analyzed.
 After passing through hidden layers of A.N.N , input is converted into suitable
output which help us in determine overall effect of the previous parameter.
Contd…
Delivery Systems
21

INTRODUCTION
The International Organization for Standardization gives a definition of robot in ISO 8373: "An
automatically controlled, reprogrammable, multipurpose, manipulator programmable in three or
more axes, which may be either fixed in place or mobile for use in industrial automation applications."
 Reprogrammable: whose programmed motions or auxiliary functions may be changed without
physical alterations;
 Multipurpose: capable of being adapted to a different application with physical alterations;
 Physical alterations: alteration of the mechanical structure or control system except for changes of
programming cassettes, ROMs, etc.
 Axis: direction used to specify the robot motion in a linear or rotary mode. The Robotics Institute of
America defines a robot as Re-programmable multi-functional manipulator designed to move
materials, parts, tools, or specialized devices through variable programmed motions for the
performance of a variety of tasks.
23

THREE LAWS OF ROBOTICS
1. A robot must obey orders given to it by human beings except where such orders
would conflict with the First Law.
2. A robot must protect its own existence as long as such protection does not conflict
with the First or Second Law.
3. A robot may not injure a human being or, through in action, allow a human being to
come to harm.
24

ADVANTAGES OF PHARMACEUTICAL ROBOTS
 1. Accuracy: Robotic systems are more accurate and consistent than their human counterparts.
 2. Tirelessness: A robot can perform a 96 man-hour project in 10 hours with more consistency and
higher quality results.
 3. Reliability: Robots can work 24 hours a day, seven days a week without stopping or tiring.
 4. Return on investment (ROI): There is quick turn-around with ROI. Plus, with the increase in quality
and application speed, there are the benefits of increased production possibilities.
 5. Affordability: With the advancements in technology and affordable robotics becoming available at
less cost, more pick and place robotic cells are being installed for automation applications.
 6. Production: With robots, throughput speeds increase, which directly impacts production. Because
robots have the ability to work at a constant speed without pausing for breaks, sleep, vacations, they
have the potential to produce more than a human worker.
25

 7. Quality: Robots have the capacity to dramatically improve product quality. Applications are
performed with precision and high repeatability every time. This level of consistency can be hard
to achieve any other way.
 8. Speed: Robots work efficiently, without wasting movement or time. Without breaks or
hesitation, robots are able to alter productivity by increasing throughput.
 9. Flexibility: Packaging applications can vary. Robots are easily reprogrammed. Changes in
their End of Arm Tooling (EOAT) developments and vision technology have expanded the
application-specific abilities of packaging robots.
 10. Safety: Robots increase workplace safety. Workers are moved to supervisory roles, so they no
longer have to perform dangerous applications in hazardous settings.
 11. Savings: Greater worker safety leads to financial savings. There are fewer healthcare and
insurance concerns for employers. Robots also offer untiring performance which saves valuable
time. Their movements are always exact, so less material is wasted.
 12. Redeployment: The flexibility of robots is usually measured by their ability to handle multiple
product changes over time, but they can also handle changes in product life cycles.
26

 13. Reduced chances of contamination: Removing people from the screening process reduces the
potential for contamination and the potential for dropped samples when handling them in
laboratories. Robotics performs these tasks much faster with more precision and accuracy.
 14. Cost: Paybacks for the purchase of robotic equipment in the pharmaceutical industry, given
the fairly high hourly labor rates paid to employees, number of production shifts, and the low cost
of capital. A typical robot installation, complete with accessories, safety barriers, conveyors, and
labor, could cost around $200,000. If that robot were to replace four manual workers each earning
approximately $30,000 per year, the robot would be paid for through salary savings alone in a
little more than a year and a half.
 15. Work continuously in any environment: Another advantage in the laboratory is that robots
are impervious to many environments that would not be safe for humans. A robot can operate
twenty-four hours a day, seven days a week without a dip in accuracy
27

DISADVANTAGES OF PHARMACEUTICAL ROBOTS
1. Expense: The initial investment of robots is significant, especially when business owners are
limiting their purchases to new robotic equipment. The cost of automation should be calculated in
light of a business' greater financial budget. Regular maintenance needs can have a financial toll as
well.
2. Dangers and fears:
Although current robots are not believed to have developed to the stage where they pose any threat
or danger to society, fears and concerns about robots have been repeatedly expressed in a wide range
of books and films. The principal theme is the robots' intelligence and ability to act could exceed that
of humans, that they could develop a conscience and a motivation to take over or destroy the human
race.
3. Expertise: Employees will require training in programming and interacting with the new robotic
equipment. This normally takes time and financial output.
4. Return on investment (ROI): Incorporating industrial robots does not guarantee results. Without
planning, companies can have difficulty achieving their goals.
5. Safety: Robots may protect workers from some hazards, but in the meantime, their very
presence can create other safety problems. These new dangers must be taken into consideration
28

ROBOTS USED IN PHARMACEUTICAL INDUSTRY
1. Pharmaceutical Container Replacement Robot:
This autonomous robot is capable of navigating tight spaces at factories for the purpose of
transporting containers used in the pharmaceutical manufacturing process.
The robot can automatically connect itself to large containers (or cases packed with products)
weighing up to 200 kilograms (440 lbs) for transport. The robot only needs to be charged once per
day, it can be freely programmed and customized to suit the manufacturing process, and it is safe and
easy to use on existing production lines.
29

2. Cylindrical Robot for High Throughput Screening
 ST Robotics presents a new 4-axis cylindrical robot for DNA screening in applications such as
forensic science, drug development, bacterial resistance, and toxicology.
 The R19 is a totally new design that may be supplied as a precise 4-axis robot, or as a simple 2-
axis plate mover. It is usually mounted on a track, which can be up to five meters long,
surrounded by various laboratory instruments. The robot moves samples from instrument to
instrument according to a protocol decided by the user. Advanced drives create swift and
smooth motion while maintaining quiet operation in the lab environment.
 Like all Sands Technology robots, the new R19 is a totally reliable workhorse, tested to ISO 9000
quality assurance.
 The KUKA KR 1000 Titan is the company's latest product and with its heavy weight capabilities
has earned an entry in the Guinness Book of Records. The KR 1000 Titan is the world's first
industrial robot that can lift a payload of 1000 kilograms with a reach of 4000 mm and will be
handling a Chrysler Jeep body. The Titan is ideally suited to handle heavy, large or bulky work
pieces. The heavyweight champion was developed for sectors such as the building materials,
automotive and foundry industries.
30

 This robotic food and pharmaceutical handling system features a visual tracking system and a pair
of multi-axis robot arms that each can accurately pick up 120 items per minute as they move
along a conveyor belt. The arms can work non-stop 24 hours a day, are resistant to acid and
alkaline cleaners, and feature wrists with plastic parts that eliminate the need for grease. The
sanitary design provides the cleanliness required of machines tasked with handling food and
medicine. With a proven record of success in reducing manufacturing costs and improving
quality, about 150 systems have been sold to manufacturers worldwide since October 2006.
31

3. Six-Axis Robots suit Class 1 Clean Room Applications:
Running on Smart Controller (TM) CX controls and software platform, Adept Viper (TM) s650 and
Adept Viper (TM) s850 bring precision motion and 6-axis dexterity to clean room assembly,
handling, testing, and packaging applications.
32

4. Space Saving Ceiling Mounted Robot:
Adept Technology has introduced a ceiling-mounted version of its s800 series Cobra robot. The
inverted robot offers high-speed packaging and assembly with a wider reach, while leaving a much
clearer working area. The new robot offers several advantages over its predecessor, which is floor-
mounted and traditionally sits beside the conveyor belt or packing line. While the Cobra s800
Inverted Robot has a reach of 800mm, the same as the previous floor-mounted model, being
mounted on the ceiling above the conveyor effectively doubles this reach. The machinery can also
be supplied with a vision system of up to four cameras, which identify the position of products on
the conveyor belt and link back to the robot so it can accurately pick up and orientate the product
for assembly or packaging
33

5. Metal Detector Targets Pharmaceutical Industry:
Incorporating Quadra Coila system, Goring Kerr DSP Rx screens pills and capsules at out
feed of tablet presses and capsule filling machines. It offers adjustable in feed heights
from 760-960 mm and angular adjustments of 20-40°. System features open-frame design
and polished, stainless steel finish. For maximum hygiene, pneumatics and cables are
contained within unit stand. Mounting bars have round profiles to remove risk of debris
and bacteria traps
34

APPLICATIONS OF ROBOTS PHARMACEUTICAL
INDUSTRY
1. Research and Development (R&D)
Robots now also play an essential role in the development of new drugs. In high throughput screening
(H.T.S.) for instance, millions of compounds are tested to determine which could become new drugs.
 There is a need for the use of robotics to test these millions of compounds.
 The use of robotics can speed this process up significantly, just as they can any other process where a
robot replaces a person completing any repetitive task.
2. Control Systems
Most robots have onboard controllers that communicate with other machines' programmable logic
controllers (PLCs) or with personal computers (PCs) networked to the line.
 Robot controller is an industrial VME bus controller that connects to PCs for networking and for
graphical user interfaces.
35

3. Laboratory Robotics
This new technology allows human talents to be concentrated on sample selection and submittal, and
scrutiny of the resulting data, rather than monotous tasks that lead to boredom and mistakes.
The desired results of this automation are of course better data and reduced costs. Using laboratory
robotics, new experimental procedures are eliminating human tedium and miscalculation in washing
and transferring.
This includes experiments in radioactive, fluorescent, and luminescent analysis Laboratory robotics is
being increasingly applied in pharmaceutical development to help meet the needs of increasing
productivity, decreasing drug development time and reducing costs.
Three of the most common geometries for laboratory robots are: Cartesian (three mutually
perpendicular axes); cylindrical (parallel action arm pivoted about a central point); and
anthropomorphic (multijointed, human-like configuration).
36

4. Sterilization and Clean Rooms
Robotics can be adapted to work in aseptic environments. Clean room robots have features that
protect the sterile environment from potential contamination. These features include low flake
coatings on the robotic arm, stainless steel fasteners, special seal materials, and enclosed cables.
Clean room robots reduce costs by automating the inspection, picking and placing, or loading and
unloading of process tools.
Benefits of robot use in the clean room include:
1. Robots minimize scrap caused by contamination.
2. Robots reduce the use of clean room consumables such as bunny suits.
3. Robots reduce scrap by minimizing mishandled or dropped parts.
4. Training costs and clean room protocol enforcement are minimized.
5. Robots reduce the amount of costly clean room space by eliminating aisles and access ways
typically required for human clean room workers.
Robots can also be enclosed in mini environments. This permits relaxed cleanliness throughout the
remainder of the plant.
37

5. Packaging Operations:
Packaging processes, like other pharmaceutical operations, benefit from the speed and
repeatability that automation brings.
Robotics in particular provides flexibility and accuracy. In some packaging applications such as
carton loading, robotics also performs more efficiently than dedicated machines.
Pharmaceutical packaging machines are often custom designed to handle specific product
configurations such as vials.
38

Challenges faced by AI
 One of the challenges of using AI in the pharmaceutical industry is the availability of
resources and access. A potential solution to this is to simplify AI models so that users can
input data without complication.
 Another is the social and institutional understanding of AI. Due to the various ways in
which AI can be modelled, it can be hard to explain how it works, making adoption even
harder. Although a model can be mathematically proven, its reasonings can be difficult to
articulate.
 Purchasing companies requesting the documentation and explanation for the decisions
that AI makes, especially when they can affect an individual or private citizen.
 Techniques and skills are very complicated.
39

Developments of Nano robots for Drug
Delivery
Artificial Intelligence in Personalized
Medicine
Delivery Systems
40

 Development nanorobots systems has provided the
possibility of fabricating implantable robots for performing
a variety of tasks including the controlled delivery of drugs
or genes.
 Because of the remarkable advances in nanotechnology, increasing interest has been attracted
towards the development of nanorobots which are integrated with internal or external power
supply, sensors, and AI.
 Bio-nanorobot rule structure includes the navigation rules,
collision avoidance, target identification rule, detection and
attachment rule, drug delivery rule, mission complete rule,
and activation of flush-out mode leading to the excretion of
bio-nano-robots.
 several issues have remained challenging including those related
to the fabrication process, controlling interactions with complex
biological environments, and biocompatibility.
Delivery Systems
41

 medicine dosing has been relatively generic, adopting a universal approach for treatment
of an entire population of patients with a particular condition.
 Personalized medicine moves away from this ‘trial and error’ approach. Rather than
considering all patients with a particular condition as being the same, personalized
medicine represents a more refined approach that takes into account some disease
heterogeneity.
 By combining and analyzing information from groups of patients, personalized medicine
attempts to determine the most effective treatment plan for a particular individual.
 Given how important data-intensive assays are to revealing appropriate intervention
targets and strategies for treating an individual with a disease, AI can play an important
role in the development of personalized medicines.
Delivery Systems
42

References
 Russell S.J. and Norvig P., Artiﬁcial Intelligence : A Modern Approach, 3rd Ed , 20-28
 http://www.insightpharmareports.com/Affiliated-Reports/FirstWord/Artificial-Intelligence-in-the-Pharma-Industry
 https://zendesk.com/deep-learning-vs-machine-learning-a-simple-explanation-47405b3eef08
 Agatonovic-Kustrin S. 2000. J Pharmaceut Biomed. 22(5):717-727.
 Macalino SJY et al, 2015, Arch. pharm. res., 38(9):1686-1701.
 Li Y et al, 2015, Eur. J. Pharm. Biopharm. , 1163-1176
 Ramesh K, Gupta S, Ahmed S, Kakkar V.2016 IJBSBT 8(6):11-20.
 Aksu B et al, 2013, Pharm. dev. technol., 236-245.
 https://www.england.nhs.uk/healthcare-science/personalisedmedicine
 Dong L et al 2006. Proceedings IEEE ICRA 2006, pp 1396-1401.
 Agatonovic-Kustrin S, Beresford R 2000. J Pharmaceut Biomed. 22(5):717-727.
 www.clinicalinformatics.com
 Neat GW, Kaufman H, Roy RJ 1989. IEEE Control Syst. Lett. 9(4):20-24.
 Petrović J, Ibrić S, Betz G, Đurić Z 2012. Int J Pharm. 428(1-2):57-67.
 https://www.europeanpharmaceuticalreview.com/article/107772/optimising-artificial-intelligence-in-the-pharmaceutical-industry/
 https://www.pharmatutor.org/articles/pharmaceutical-industrial-applications-robots-current-scenario-recent-review?page=1
47

Artificial intelligence

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Artificial intelligence

Similar to Artificial intelligence (20)

More from Pratiksha Chandragirivar

More from Pratiksha Chandragirivar (18)

Recently uploaded

Recently uploaded (20)

Artificial intelligence