Artificial Intelligence in Medicine

Artificial Intelligence appeared in the 1950s as a term to designate a set of novel methods and philosophical attitudes toward problem solving. In the 1980s it had a serious falling in popularity, which was the period of methodologies such as Intellgent Control, fuzzy systems, and neural networks, nowadays composing computational intelligence, these are numerical.based methodolgies, so far artifiical intelligence was mainly worried about symbolic-based methodologies. 

Some says that artificial intelligence possesses too much Is, this is to highlight the problems faced by the same in the past. There are several definitions. Russel and Norvig (2010) categorizes them into: thinking humanly, thinking rationally, acting humanly, or acting rationally. Computational intelligence, a competitor for attention, is placed into acting rationally. 

Computer is without a doubt the revolution of the millennium. Medicine is not different from the other sciences, it is nowadays somehow slave of equipaments, some doctors will not move even an eye without the proper machinary. All these systems cannot be run and controlled just based on linear models or linearizations as it is done often by mathematician. Computer scientists and engineers are more practical and they have certainly been taking advantage of the changes so far.

According to Fieschi (1990),  artificial intelligence, a strange phrase in which the two words taken

separately conjure up opposite meanings. Intelligence seems to us to be intimately associated with human behaviour. The idea of 'artificial', on the other hand, conjures up the idea of objects characteristically not natural but 'man-made'. To call 'artificial' a prime component of human nature seems a paradox. This term is badly chosen, particularly as the aim of artificial intelligence systems is to represent behaviour comparable to human behaviour. The same observation was done on Poole et al (1998). Further, Fieschi (1990) still pinpoint, artificial intelligence in medicine is going to take a very important place in the science of medical informatics.

In summary, artificial intelligence is changing, and it seems for better. Medicine is a field rich on tough problems, problems which solution could benefit several people. This field certainly will occupy the minds of several researches on the future. I am quite sure that computational intelligence will not be left behind, see for example Lam et al (2012).

References cited

RUSSELL, Stuart; NORVIG, Peter. Artificial Intelligence: A modern approach. Third edition. Prentice Hall Series in Artificial Intelligence: 2010.
Fieschi, M, Artificial Intelligence in medicine: expert systems, Translated by D Cramp, Spring-Science + Business Media, 1990.
David Poole; Alan Mackworth; Randy Goebel. Computational Intelligence: A Logical Approach. Oxford University Press. 1998.
Lam, HK; Ling, SH; Nguyen, HT (eds) (2012). Computational Intelligence and its applications: evolutionary Computation, Fuzzy Logic, Neural Network, and Support Vector Machine Technique. Imperial College Press.

External Links

- Conference on Artificial Intelligence in Medicine (Italy)

- Artificial Intelligence in Medicine (Portugal)

Systems Identification vs. Model identification

"However, still the problem of system identification has not been completely solved.
Consequently, nowadays new ideas and methods to solve the system identification
problem or parts of it are introduced." Keesman 2011.

We can say that the ultimate goal of science is building models, mathematical modeling is one of the means to achieve this goal, the promising is to apply more symbolic methodologies once this is the way we think. However, as always, life is not easy. Behind this challenge, new ones come out. One of the is system identification, sometimes referred to as model identification, apparently model identification is used more in Italy, further, it seems to be a broader concept, having system identification as a special case. 

The challenge that arises is based upon the fact that besides we might know how a system model looks like, in general we do not know the parameter values, for instance, what demands methods to find them from experiments. Maybe a nice way to see it is if you know the basics of music theory. Besides a symphony is complex, it is built upon simple chords, in general 8, but some musicians rely their works on just three chords; e.g. I, IV, and V. If you hear a song, you suppose to identify the chords, given you know them. However in real life it does not happen so easily. If you ask ten guitar players to transcribe a song you like, this is highly possible they will disagree, especially on subtle differences such as G and G7. In this case we can say they have failed to identify the system model, this is so an identifiability problem, they know the model in general, but details cannot be filtered out. 

In general, system identification consists of three basic steps: experiment design and data acquisition, model structure selection and parameter estimation, and model validation [1]. System identification deals with the problem of building mathematical models of dynamical systems based on observed data from the system [2]. 

Traditionally system identification is based on mathematics, however new trends are applying what can be kept by the name of computational intelligence such as neural networks.  

References cited

[1] Karel J. Keesman, System Identification: an introduction, Advanced books in control and signal processing, Spring, 2011.

[2] Lennart Ljung, System Identification: theory for user, Second edition, Practice hall information znd system science series, 1999. 

Pharmacokinetics and pharmacodynamics

"Pharmacokinetics and pharmacodynamics are the important fields of pharmaceutical sciences for investigating disposition profiles and the pharmacological efficacy of drugs in the body under various experimental and clinical conditions."

(Caldwell et

al., 1995 and Cocchetto and Wargin, 1980, cited by Kwon 2002).

Source: Pires et al 2014.

Source: Pires et al 2014.

References cited

Younggil Kwon, Handbook of Essential Pharmacokinetics, Pharmacodynamics and Drug Metabolism for Industrial Scientists, Kluwer Academic Publishers, 2002.

Caldwell J. et al., An introduction to drug disposition: the basic principles of absorption, distribution,

Cocchetto D. M. and Wargin W. A., A bibliography for selected pharmacokinetic topics, Drug Intel. Clin. metabolism and excretion, Toxicol. Pathol. 23: 102-114, 1995.

Pharmacol. 14: 769-776,1980.

JG Pires, R Maggio, C Manes, P Palumbo, On the importance of pharmacokinetics and pharmacodynamics in engineering sciences as an inter- and multidisciplinary field: an introductory analysis. SIMPEP 2014,

Health care: past and future

Systems biology and the new demands

A lack of system-level understanding of cellular dynamics has prevented any substantial increase in the number of new drugs available to the public and any increase in drug efficacy or eradication of any specific disease. In contrast, pharmaceutical companies are currently lacking criteria for selecting the most valuable targets, research and development (R&D) expenses skyrocket, and new drugs rarely hit the market and often fail in clinical trials, while physicians face an increasing wealth of information that needs to be interpreted intelligently and holistically (Hood 2004;  Kriete and Eils, 2006).

Evolution of the Modern Healthcare System

Before 1900, medicine had little to offer the average citizen, since its resources consisted mainly of

the physician, his education, and his “little black bag.” In general, physicians seemed to be in short

supply, but the shortage had rather different causes than the current crisis in the availability of healthcare professionals. Although the costs of obtaining medical training were relatively low, the demand for doctors’ services also was very small, since many of the services provided by the physician also could be obtained from experienced amateurs in the community. The home was typically the site for treatment and recuperation, and relatives and neighbors constituted an able and willing nursing staff. Babies were delivered by midwives, and those illnesses not cured by home remedies were left to run their natural, albeit frequently fatal, course. The contrast with contemporary healthcare practices, in which specialized physicians and nurses located within the hospital provide critical diagnostic and treatment services, is dramatic (Bronzino, 2006).

References cited: 

Joseph D. Bronzino, Biomedical Engineering Fundamentals, The Biomedical Engineering Handbook, Third Edition, Taylor&Francis, 2006. 

Leroy Hood, James R. Heath, Michael E. Phelps, Biaoyang Lin, Systems Biology and New Technologies Enable Predictive and Preventative Medicine, Science Viewpoint, October, Vol. 306, 2004.

Andres Kriete, Roland Eils, Computational Systems Biology, Elsevier Academic Press, 2006. 

Stochastic models in medicine and life science (requirement for 2nd year, talk)

Whether we investigate the growth and interactions of an entire population, the evolution of DNA sequences, the inheritance of traits, or the spread of disease, biological systems are marked by change and adaptation [1]. It is often said that biology is going to be the science of the 21st  century as physics was the science of the 20th [3,4]. Computers, and computer science ideas and techniques, are of course an important part of all these scientific and engineering activities [4].

Indeed this represents the challenges in biological modeling compared to traditional branches such as physics. In general the adaptations and changes are much faster than physical systems, what makes the modeling and analysis in most of the cases a formidable task.

When I first read a biology textbook, it was like reading a thriller. Every page brought a new shock. As a physicist, I was used to studying matter that obeys precise mathematical laws. But cells are matter that dances. Structures spontaneously assemble, perform elaborate biochemical functions, and vanish effortlessly when their work is done. Molecules encode and process information virtually without errors, despite the fact that they are under strong thermal noise and embedded in a dense molecular soup. The main message is that biological systems contain an inherent simplicity  [3] .

In the pharmaceutical industry, the incorporation of the disciplines of pharmacokinetics, pharmacodynamics, and drug metabolism (PK/PD/DM) into various drug development processes has been recognized to be extremely important for appropriate compound selection and optimization [2].

Conerstones of my research

1. Introduction

The first year was dedicated to: 1) achieving the minimal requirements in terms of credits; 2) gathering the maximum amount of knowledge. The academic activities was divided into mathematical and biomedical; and master-level disciplines, summer schools, readings, and short courses. The activities was either suggested by the advisors, prof. Palumbo and prof. Manes, or chosen by me. Some of the disciplines was followed on the hope to increase my theoretical background.

2. Master-level disciplines

• Controllo Ottimo. Prof E De Santis (UAQ): in theory important for optimum regimen design in medical treatments. As result, a talk in the IASI-CNR in June and a paper in the Symposium SIMPEP 2014.
• Farmacologia Speciale. Prof. R, Maggio (UAQ): this theory supposes to support me next year, drug regimen design. As result, a talk given in the department of Medicine (UAQ) and an awarded paper in SIMPEP 2014, a journal extension was proposed by SIMPEP and a book proposal was submitted;

3. Summer Schools

• Mathematical Models and Methods for Living Systems: it was a one-intensive week of studies, see http://web.math.unifi.it/users/cime/. Presentation of talk: On the mathematical modeling in gene expression estimation: an initial discussion on PBM and BM;
• Systems Biology and Systems Medicine: precision Biotechnology and Therapies: this was a one-intensive week of studies and computer simulations, tutorial and lessons. See: http://ucbf.lakecomoschool.org/. Presentation of poster: On the mathematical modeling in gene expression estimation: an initial discussion on PBM and BM;

4. Short courses

• Software Architecture: theory of how to design better software;
• Convergence theory for observers: Necessary, and Sufficient conditions: theory on the design of state reconstruction systems;
• Others: other courses were followed in the hope to find insights and methodologies.

5. Main references used

• S Lenhart, J T Workman, Optimal Control Applied to biological models, Chapman & Hall/ CRC, Mathematical and Computational Biology Series, 2007;
• Sara E Rosenbaum, basic pharmacokinetics and pharmacodynamics: an integrated textbook and computer simulations, John Wiley & Sons, 2011.

6. Most significant publications

• JG Pires, R Maggio, C Manes, P Palumbo, On the importance of pharmacokinetics and pharmacodynamics in engineering sciences as an inter- and multidisciplinary field: an introductory analysis. SIMPEP 2014, Bauru (São Paulo, Brasil), Online: http://www.simpep.feb.unesp.br/anais_simpep.php?e=9
• JG Pires, C Manes, P Palumbo, On the importance of optimal control theory in engineering sciences as a complementary and supplementary methodology to Operations Research: a case-study analysis. SIMPEP 2014, Bauru (São Paulo, Brasil), Online: http://www.simpep.feb.unesp.br/anais_simpep.php?e=9 

1. Introduction

The year of 2015, second year of the PhD pathway of the abovementioned student, was agreed to be dedicated to researches on the IASI-CNR (Gemelli Ospedale)[1]. The researches will consist of reading literatures and testing models published or propose new ones. The main topics is what we have called "The Big Glucose Model," which boils down to an attempt to enhance already existing mathematical and computational models for studying glucose control, in general the models are based just on insulin, the idea is to gather several different models based on other hormones or bio- molecules considered significant on the alteration of glucose levels in the human blood.

2. Courses to enroll

• Identificazione dei Modelli e Analisi dei Dati: this is based on state space models;
• Complementi di Automatica: this is based on kalman filter models and parameter stimation;

3. References

The references were not defined yet, it will be taken from an archive offered by De Gaetano, from the IASI-CNR Gemelli Ospedale.

A starting point could be:

• P. Palumbo, S. Ditlevsen, A. Bertuzzi, A. De Gaetano, Mathematical modeling of the glucose-insulin system: A review, Mathematical Bioscience, 2013;
• Ludovic J. Chassin, Malgorzata E. Wilinska, Roman Hovorka. Evaluation of glucose controllers in virtual environment: methodology and sample application, Artificial Intelligence in Medicine (2004) 32, 171—181

4. Final Remarks

Unfortunately a precise agenda for next year is complex, once it depends on my response to the project proposed by Andrea De Gaetano and the success on the research. In the first year I have finished all the prerequisites - two master-level disciplines and 18 credits of ad hoc activities - for avoiding conflicts with this part of my academic cycle.

5. Extra

One paper partially accepted: Biologia Sistêmica: um novo paradigma para as ciências biológicas e exatas “ou” Biologia Sistêmica e Inteligência Computacional. One paper invited to a journal, and three textbooks proposed to publish under invitation.

---

[1] CNR-IASI - Laboratorio di Biomatematica, UCSC – Largo A. Gemelli 8, 00168, Roma, Italy, Ph: +39 06 30155389       Fax: +39 06 3057845

http://www.biomatematica.it

See also: Information Engineering, Engenharia da informação VS engenharia de produção

Suggestion: Access the video from Youbube, then you will see a set of correlated videos, which can be very helpful. See scheme below.

Accessing Youtube from the current blog