BREAKING AGE-LONG

BARRIERS IN SCIENCE USING

ARTIFICIAL INTELLIGENCE
A webinar delivered at Pan Africa University- Life
and Earth Science Institute (including Health and
Agriculture)- PAULESI on 15th October 2021.

DR. PAUL A. ADEDEJI

University of Johannesburg
South Africa.
What comes to your mind when
you hear Artificial Intelligence?
The Scientific Revolution
• Ancient approaches to science are now been challenged and fast fading
away and digital techniques are taking up the stage.

• “heavier-than-air flying machines are impossible” – Lord Kelvin 1895

• The Aeroplane
• Telecommunication cannot be via a wireless means
• Mobile phone Global System of
Code Division Multiple Gross Packet Radio
Mobile Communication
Access (CDMA) Service (GPRS)
(GSM)

2G, 3G, 4G/LTE, 5G

(LTE- Long-term Evolution)
The Scientific Revolution
• Conventional approaches to science are now been
challenged and digital techniques are taking their
turns

• Physical Patient Consultation

• Tele-medicine – Long-distance patient-clinician contact
• Reduced queue lengths and waiting time at hospitals
• Increase use of time

• Expert Systems for intelligent medical diagnosis

• Rapid Interventions
Innovations in Science
• Digital Twin Technology
• A digital representation of a physical product, service or
process.
• The virtual twin collects data from the physical asset,
process or service into mathematical models that
simulate the original system.
Innovations in Science
• Haptics and Ultrahaptics Technology
• Science of Touch. Transmitting and understanding
information through touch.
• Haptics assisted surgical training

• Ultrahaptics Application
What does data really
mean?
• Numeric
• Alphanumeric or Texts
• Images
• Audio
• Video frames

• Data usefulness RE-defined

Machine Learning and Data
• Data is the crux of artificial intelligence and without it, artificial intelligence
becomes “artificial naivety”.
• Data forms the key ingredient of machine learning models.
• - these models learn the data.

The 4Vs of data: Large observational, machine

analyzed data [1]
• Volume
• Variety Experimentally/ intentionally collected
• Velocity data of a human scale focused on
• Veracity causation and understanding rather
than prediction [1]
What level of data use
am I?
Data-Centric

Data is the centre of

Operations/Activities
Data-Driven

Use data to make decisions at

strategic and operational
Data Informed levels

Data is collected
Data can be accessed
Use dashboards for
visualization
BIG DATA AND SMALL DATA
o Resolution and indexicality is tight and strong
o Potentially in-expensive [1]
BIG DATA o Highly exhaustive and the velocity of generation is significantly high. [3,
4]
o Many insights can be extracted from it [5].

o Comprises of representative data harvested from a population.

o Performs better for most statistical inference.
SMALL DATA
o Data quality is ensured.
o Quality outwits quantity where causation is to be inferred [2]

Data should be viewed as a valuable and re-usable resource.

Data quality is significant to model robustness.
Moving beyond laboratory experiments…
o If it is all about DATA, won’t the lab equipment of today
be redundant tomorrow?

o How is DATA influencing the research activities today?

DESCRIPTIVE USE DIAGNOSTIC USE PREDICTIVE USE PRESCRIPTIVE USE

What Happened? Why did it Happen? What will happen? What should happen?

AI AI AI
AI
Predictive Modelling in Science Today
• Predictive Models are intelligent models that learn the complex patterns
and relationship within a given input dataset to make accurate prediction
of future outcomes from unknown scenario.

• Classification of sperm cells

• Prediction of implantation outcome based on embryos
• Identifying Essential genes in organisms
• Predicting disease spread
• Waste Management investigation
• Intelligent Medical imaging
Predictive Modelling in Science Today
MACHINE LEARNING • Artificial Neural Networks
MODELS • Support Vector Learning Principles
Regression
• Hybrid models [7-8]
Regressive
Supervised Learning [6-
Models 10]
• Predicts a quantity
(discrete or continuous) Un-supervised Learning
• Artificial Neural Networks
Semi-supervised
• Decision Trees
learning
• Support Vector Machines
• Rule-based machine Deep Learning
Classification learning models
Models • Hybrid models [5-9]
• Predicts class/label
(Discrete)
Knowledge-
EXPERT SYSTEMS
base (rules)
 Supervised and Unsupervised Learning
• Supervised Learning • Unsupervised Learning

• Clustering techniques (Fuzzy c-mean (FCM), grid

• Support vector regression, ANN, Linear partition, subtractive clustering, hierarchical
Regression, Support vector machines, k- Nearest clustering, etc.), Self organizing maps (SOMs)
Neighbour (kNN), Naïve Bayes etc.
 Semi-Supervised and Reinforced Learning
• Semi-Supervised Learning • Reinforced Learning

• Training machine learning model to make series or decisions in a

complex situation.
• The model gets rewards/ penalties for its action and
• Often the principle of real-life cases this informs its further decision.
• Used in Games
 PLANT BREEDING

 Machine learning on different –omics datasets.

 Distinguishing different types of genomics regions
(localized and global level predictions)
Source: Ref. [28]  Investigating genome functions and metabolic pathways.

 At phenotype level, machine learning

can be used to predict growth pattern
based on related factors as inputs.-
supervised learning approach.

Ref. [15, 16]

Source: Ref. [30]
REPRODUCTIVE HEALTH
SCIENCES

• Predictive modeling of in-vitro fertilization (IVF) and

Intracytoplasmic sperm injection (ICSI) outcomes-
• Classification of sperm cells
• Embryo Selection
• Prediction of implantation outcome based on embryos
• Expert systems for prognosis and diagnosis
• Effective Selection of sperms and embryos for ART process
• Digitization of IVF process
Ref: [11-13]
GEOSCIENCES- PETROLEUM
GEOSCIENCES AND MINERAL
EXPLORATION GEOSCIENCES

 AI-based predictive modelling of petroleum reservoir

conditions. (e.g. prediction of permeability in petroleum
reservoirs)
 Prediction of geological-petrophyscial properties
 AI-based decision-making on mineral exploration (Data-driven
or knowledge-driven)

Ref. [17-20]
 ENVIRONMENTAL MANAGEMENT

In this field, AI can be applied to

• Waste Management
• Environmental protection and hazard control
• Air quality
Source: Ref. [21]
• Water resource management
• Environmental sustainability, etc.
(Data-driven and/or knowledge-based approach)
Ref. [24-26]

Source: Ref. [22]

 Better understanding of prevalent events VETERINARY (AVIAN MEDICINE AND
and its dynamics e.g. disease spread VETERINARY VACCINE PRODUCTION AND QUALITY
 Risk analysis and parasite control (proactive CONTROL)
and reactive levels
 Modelling complex socio-biological systems
 Spatio-temporal predictive modelling at
specific and generic levels.
 AI-based quality control for veterinary
vaccine production
 Value chain optimization using AI
techniques (e.g. evolutionary based
approaches)
 Genetics
 Data-driven decision making

Ref. [27, 28] Source: Ref. [27]

 Proteomics

MEDICINAL PLANTS RESEARCH AND

Metabolomics Transcriptomics

Machine
DRUG DEVELOPMENT
Learning

Better understanding of plant biology and

Lipidomics Genomics disease control.
AI-based process optimization
Automatic medicinal plant recognition
Prediction of variables of interest from
experimental data.
Modeling experimental data to reduce the
number of animal experiments.
AI-integrated bioinformatics
SPORT MANAGEMENT & POLICY
DEVELOPMENT
AI-based recruitment process
Training and Performance
Sport trauma prediction
Prediction of outcomes of sport
matches
Expert system for policy development

Ref. [29, 30]

Responsible AI Minimizes Un-
intended Bias
• Responsible AI is “the practice of
designing, developing, and
Promotes
deploying AI with good intention to Transparency
empower employees and
businesses, and fairly impact
customers and society” Privacy Protection
and Data Security

Community Benefit
Previous Works
Previous Works (2)
 OUR RESPONSIBILITIES
Equipping ourselves for the future tasks
First, be well-grounded in the basics of the
classical models.
Take that machine learning course!
 Udemy
 Coursera
Give yourself that push.
Taking the challenge to digitize our classical
way of experiments
Learn AI- Do not be left out
Training and re-training
Creating data banks of credible experimental
data. (E.g. Data in brief, data repositories)
Institutional Responsibility
Need for AI
Untapped
Opportunities
still lies in
artificial
intelligence for
scientific
growth…

29
Feedback
Was this session useful?

Do you have any questions or issues to

raise?

What areas in your field do you find AI

applicable? – Chat box

30
 References
1. R. Kitchin and G. McArdle, “What makes Big Data, Big Data? Exploring the ontological characteristics of 26 datasets,” Big Data Soc., vol. 3, no. 1, pp. 1–10,
2016, doi: 10.1177/2053951716631130.
2. J. J. Faraway and N. H. Augustin, “When small data beats big data,” Stat. Probab. Lett., vol. 136, pp. 142–145, 2018, doi: 10.1016/j.spl.2018.02.031.
3. T. R. Rao, P. Mitra, R. Bhatt, and A. Goswami, The big data system, components, tools, and technologies: a survey, vol. 60, no. 3. Springer London, 2018.
4. Y.-H. Kuo and A. Kusiak, “From data to big data in production research: the past and future trends,” Int. J. Prod. Res., vol. 7543, pp. 1–26, 2018, doi:
10.1080/00207543.2018.1443230.
5. T. Hillel, M. Bierlaire, M. Z. E. B. Elshafie, and Y. Jin, “A systematic review of machine learning classification methodologies for modelling passenger mode
choice,” J. Choice Model., vol. 38, no. August 2020, p. 100221, 2021, doi: 10.1016/j.jocm.2020.100221.
6. P. A. Adedeji, S. Akinlabi, N. Madushele, and O. O. Olatunji, “Wind turbine power output very short-term forecast : A comparative study of data clustering
techniques in a PSO-ANFIS model,” J. Clean. Prod., vol. 254, pp. 1–16, 2020, doi: 10.1016/j.jclepro.2020.120135.
7. P. A. Adedeji, S. Akinlabi, N. Madushele, and O. O. Olatunji, “Wind turbine power output very short-term forecast : A comparative study of data clustering
techniques in a PSO-ANFIS model,” J. Clean. Prod., vol. 254, pp. 1–16, 2020, doi: 10.1016/j.jclepro.2020.120135.
8. O. O. Olatunji, N. Madushele, P. A. Adedeji, and S. Akinlabi, “Digitalisation of biomass exploration: A case study of biomass feedstock classification,” in
American Society of Mechanical Engineers, Power Division (Publication) POWER, 2020, vol. 2020-Augus, doi: 10.1115/POWER2020-16772.
9. P. A. Adedeji, “Hybrid neurofuzzy investigation of short-term variability of wind resource in site suitability analysis : a case study in South Africa,” Neural
Comput. Appl., vol. 0123456789, 2021, doi: 10.1007/s00521-021-06001-x.
10. P. A. Adedeji, S. A. Akinlabi, and O. O. Olatunji, “Hybrid neurofuzzy wind power forecast and wind turbine location for embedded generation,” Int. J. Energy
Res., no. May, pp. 1–16, 2020, doi: 10.1002/er.5620.
11. A. Uyar, A. Bener, and H. N. Ciray, “Predictive Modeling of Implantation Outcome in an in Vitro Fertilization Setting,” Med. Decis. Mak., vol. 35, no. 6, pp. 714–
725, 2015, doi: 10.1177/0272989X14535984.
12. P. Hafiz, M. Nematollahi, R. Boostani, and B. N. Jahromi, “Predicting implantation outcome of in vitro fertilization and intracytoplasmic sperm injection using
data mining techniques,” Int. J. Fertil. Steril., vol. 11, no. 3, pp. 184–190, 2017, doi: 10.22074/ifs.2017.4882.
13. R. Wang et al., “Artificial intelligence in reproductive medicine,” Reproduction, vol. 158, no. 4, pp. R139–R154, 2019, doi: 10.1530/REP-18-0523.
14. Jamiepighin, “Transgenic crops: how genetics is providing new ways to envision agriculture,” 2003.
References (2)
15. M. Yoosefzadeh-Najafabadi, H. J. Earl, D. Tulpan, J. Sulik, and M. Eskandari, “Application of Machine Learning Algorithms in Plant Breeding: Predicting Yield From Hyperspectral
Reflectance in Soybean,” Front. Plant Sci., vol. 11, no. January, pp. 1–14, 2021, doi: 10.3389/fpls.2020.624273.
16. A. D. J. van Dijk, G. Kootstra, W. Kruijer, and D. de Ridder, “Machine learning in plant science and plant breeding,” iScience, vol. 24, no. 1, p. 101890, 2021, doi:
10.1016/j.isci.2020.101890.
17. F. I. Syed, A. AlShamsi, A. K. Dahaghi, and S. Neghabhan, “Machine Learning techniques to Model Geomechanics and Petrophysical Properties of Shale Reservoirs – A Systematic
Literature Review,” Petroleum, no. xxxx, 2021, doi: 10.1016/j.petlm.2020.12.001.
18. Y. Huang, T. D. Gedeon, and P. M. Wong, “Integrated neural-fuzzy-genetic-algorithm using hyper-surface membership functions to predict permeability in petroleum reservoirs,” Eng.
Appl. Artif. Intell., vol. 14, no. 1, pp. 15–21, 2001, doi: 10.1016/S0952-1976(00)00048-8.
19. M. Yousefi, E. J. M. Carranza, O. P. Kreuzer, V. Nykänen, J. M. A. Hronsky, and M. J. Mihalasky, “Data analysis methods for prospectivity modelling as applied to mineral exploration
targeting: State-of-the-art and outlook,” J. Geochemical Explor., vol. 229, no. April, 2021, doi: 10.1016/j.gexplo.2021.106839.
20. J. M. A. Hronsky and O. P. Kreuzer, “Applying spatial prospectivity mapping to exploration targeting: Fundamental practical issues and suggested solutions for the future,” Ore Geol. Rev.,
vol. 107, no. March, pp. 647–653, 2019, doi: 10.1016/j.oregeorev.2019.03.016.
21. SUNSTAR, “Environmental Management,” Sunstar Group, 2020. https://www.sunstar.com/sustainability/environment/management/ (accessed Jul. 25, 2021).
22. SHMZ, “Environmental Management,” ESG Management, 2021. https://www.shimz.co.jp/en/company/csr/environment/manage/ (accessed Jul. 25, 2021).
23. O. Adeleke, S. Akinlabi, T. C. Jen, and I. Dunmade, “Prediction of the heating value of municipal solid waste: a case study of the city of Johannesburg,” Int. J. Ambient Energy, 2020, doi:
10.1080/01430750.2020.1861088.
24. O. Adeleke, S. A. Akinlabi, P. A. Adedeji, and T. Jen, “Energy Content Modelling for Municipal Solid Waste Using Adaptive Neuro-Fuzzy Inference System ( ANFIS ),” Lect. Notes Mech.
Eng., pp. 177–185, 2000.
25. O. O. Olatunji, S. Akinlabi, N. Madushele, P. A. Adedeji, and I. Felix, “Multilayer perceptron artificial neural network for the prediction of heating value of municipal solid waste,” AIMS
Energy, vol. 7, no. July, pp. 944–956, 2019, doi: 10.3934/energy.2019.6.944.
26. O. Olatunji, S. Akinlabi, N. Madushele, and P. A. Adedeji, “Estimation of Municipal Solid Waste (MSW) combustion enthalpy for energy recovery,” EAI Endorsed Trans. Energy Web, vol. 6,
no. 23, pp. 1–9, 2019.
27. P. Ezanno et al., “Research perspectives on animal health in the era of artificial intelligence,” Vet. Res., vol. 52, no. 1, pp. 1–15, 2021, doi: 10.1186/s13567-021-00902-4.
28. Animal Health Institute, “Animal Medicines,” Animal Health Institute, 2020. https://ahi.org/future-of-animal-medicine/#remote-monitoring (accessed Jul. 25, 2021).
29. G. Fialho, A. Manhães, and J. P. Teixeira, “Predicting Sports Results with Artificial Intelligence - A Proposal Framework for Soccer Games,” Procedia Comput. Sci., vol. 164, pp. 131–136,
2019, doi: 10.1016/j.procs.2019.12.164.
30. G. Kakavas, N. Malliaropoulos, R. Pruna, and N. Maffulli, “Artificial intelligence: A tool for sports trauma prediction,” Injury, vol. 51, pp. S63–S65, 2020, doi: 10.1016/j.injury.2019.08.033.
 Thank you for Listening!

Dr Paul A. Adedeji
Department of Mechanical Engineering Science,
University of Johannesburg
South Africa

Email: [email protected]