Artificial Intelligence in Health E-Book

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch

Seitenzahl: 298

Veröffentlichungsjahr: 2024

Beschreibung

Undeniable, inescapable, exhilarating and breaking free from the exclusive domain of science, artificial intelligence has become our main preoccupation. A major generator of new mathematical thinking, AI is the result of easy access to information and data, as facilitated by computer technology. Big Data has come to be seen as an unlimited source of knowledge, the use of which is still being fully explored, but its industrialization has swiftly followed in the footsteps of mathematicians; today's tools are increasingly designed to replace human beings, which comes with social and philosophical consequences. Drawing on examples of scientific work and the insights of experts, this book offers food for thought on the consequences and future of AI technology in education, health, the workplace and aging.

Leseprobe

Technological Prospects and Social Applications Set

coordinated byBruno Salgues

Volume 7

Artificial Intelligence in Health

Edited by

Marianne Sarazin

First published 2024 in Great Britain and the United States by ISTE Ltd and John Wiley & Sons, Inc.

Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms and licenses issued by the CLA. Enquiries concerning reproduction outside these terms should be sent to the publishers at the undermentioned address:

ISTE Ltd27-37 St George’s RoadLondon SW19 4EUUK

www.iste.co.uk

John Wiley & Sons, Inc.111 River StreetHoboken, NJ 07030USA

www.wiley.com

© ISTE Ltd 2024The rights of Marianne Sarazin to be identified as the author of this work have been asserted by her in accordance with the Copyright, Designs and Patents Act 1988.

Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s), contributor(s) or editor(s) and do not necessarily reflect the views of ISTE Group.

Library of Congress Control Number: 2023948023

British Library Cataloguing-in-Publication DataA CIP record for this book is available from the British LibraryISBN 978-1-78630-889-4

Author Presentation

Cyriak Azefac

Cyriak Azefac is a 2017 computer engineering graduate specializing in machine learning at the Institut Mines-Télécom (IMT) at Atlantique Bretagne-Pays de la Loire. With a PhD in engineering sciences from the Centre Ingénierie et Santé (CIS) at the École nationale supérieure des mines de Saint-Étienne (ENSM.SE), he also holds a position as a machine learning engineer at EOVI MCD. His research interests include intelligent habitats, health, frailty, machine learning and data mining.

François-Xavier Clément

François-Xavier Clément is an alumnus of the Faculté Libre de Philosophie Comparée in Paris, an IPC and an auditor at the Institut des Hautes Études de Défense Nationale (IHEDN). He has worked successively as a teacher, educational executive, school head and diocesan director. He is currently involved in a number of educational projects in which he deploys virtuous leadership in the field of education.

Aline Courie-Lemeur

Associate Professor at the Institut Supérieur du Management (ISM) at the University of Versailles Saint-Quentin-en-Yvelines, she is also responsible for the master’s in Human Resources Management. Her research and publications focus on strategic management, innovation management and network management and governance, with a particular interest in the healthcare field. She is in charge of managing research projects linked to healthcare structures.

Marysa Germain

Marysa Germain is a public health physician and head of the medical information department at the Groupement des Hôpitaux de l’Institut Catholique of Lille. Specialized in health data processing, she is involved in the development of technologies for the automated processing of information contained in computerized patient records.

Gilles Rouet

Gilles Rouet, PhD in history and associate professor of economics and management, has been a university professor of management science and director of ISM-IAE Versailles-Saint-Quentin-en-Yvelines, Université Paris Saclay, since 2015. He has also been Professor of International Relations at Matej Bel University, Banska Bystrica, Slovakia, since 2009. He has coordinated several European projects in Russia (1993–2004) and Georgia (2009–2010) and served as cooperation attaché at the French Embassies in Slovakia (2004–2008) and Bulgaria (2010–2014). He is currently a member of the editorial board of the journal Hermes, CNRS, and of the board of directors of the association AIRMAP. He leads the Human Resources, Culture and Communication axis of the Larequoi research team.

Bruno Salgues

Bruno Salgues is Director of Studies at the Institut Mines-Télécom. He currently works as a teacher-researcher in the ICT and Health program in Montpellier. Previously, from 1989 to 2010, he was a teacher-researcher in the management sciences department at the Institut national des télécommunications, now Télécom SudParis and Télécom business school. From 1981 to 1989, he was in charge of “services” at the Forecasting Department of the French Ministry of the Economy and Finance. Bruno Salgues has been and still is a shareholder in several companies in the information technology sector, some of which he currently sits on their Boards of Directors or Supervisory Boards. He is also an expert in these fields and a frequent speaker at conferences.

Marianne Sarazin

Marianne Sarazine is a medical doctor specializing in public health and holds a doctorate in life sciences from the Centre ingénierie et santé at the École nationale supérieure des Mines in Saint-Étienne, focusing on the modeling of predictive aging scores. She is currently head of the medical information department (hospital database management) at Groupe Mutualiste in Saint-Étienne, after working for 12 years in the medical information department at the Centre hospitalier in Firminy. Since 2013, she has also been an associate in the I4S department of the Centre ingénierie et santé, specializing in optimizing healthcare organization, and since 2006, regional manager of the Sentinelles network, UMRS 1136 INSERM, specializing in epidemic modeling.

Marc Soler

Marc Soler is a parietal surgeon, associate member of the French National Academy of Surgery and obtained a DEA in biomedical engineering (UTC Compiègne). He has had an online medical practice for over 15 years (website with blog, encrypted health data transmissions, RGPD, use of patient monitoring apps). He is the editor and co-host of the blog innovationesante.fr. He is an ODPC trainer who carried out the first training courses dedicated to connected health on mondpc.fr (June 2016). He collaborates with the French National Academy of Surgery, industry and higher education to promote the use of new information and communication technologies in healthcare.

Preface

The great guru of science is setting an indelible stamp on the world: artificial intelligence! Unquestionable, inescapable, exhilarating for minds (and wallets), even beyond the realm of science, it has become the talk of every society, from the most exclusive circle to the smallest grocery store in Auvergne, surpassing in notoriety all our world-famous singing stars. It is enough to confuse any artistic agent or publicist. Intelligence is a concept hitherto reserved for living beings, and now, for more than a decade, our great “science” reasoner has been telling us that this concept can be applied to machines, machines that are themselves built by man! So, man is the architect of his own replacement? In a world where epidemics have not succeeded in eliminating humans, where meteorites are still too small to stop them from journeying around the Sun, and where even wars and ideologies have not been able to settle the issue, they would finally be able to find the object of their own demise! And this is where collective hysteria finds the breeding ground for its fear and causes spirits to swell around this concept: artificial intelligence (AI).

Even if intelligence is a suggestive notion, based on the observation that humans and living beings’ capacities for discernment and understanding can be amalgamated into a single concept, the fact remains that it has determined and organized human and animal societies for centuries. With a balance and insight that were admittedly somewhat questionable at times, it has nevertheless succeeded in bringing society to its present point of development. This original fact in our galaxy (at least until proven otherwise) deserves some praise and the assurance of a reassuring continuity. But what can we say about this intelligence, dear to our Cartesian minds, except that it is very difficult to give a precise definition. Multifaceted, it has specialized and diversified, following paths of generous multiculturalism where all minds have found a place. Artists revel in the intelligence of the senses, engineers in the intelligence of demonstration, doctors in the intelligence of the human body, and craftsmen in the intelligence of everyday life, and animals have finally adapted their own intelligence to ours by accepting domestication. The concept of intelligence has evolved since ancient Greece. During this period, it was most often referred to as noûs (νοῦς), more rarely nous or noos, which can be translated as spirit, intellect, reason. For Plato, it most often refers to the most divine part of the soul, intelligence. In other words, intelligence is an emanation of a god who manages the soul and thought. Over the following centuries, this idea was reinforced with the construction of monotheistic religions, instilling the concept of a superior intelligence, God, as guarantor of man’s intelligence. In numerous verses, the Bible links man’s intelligence with his attachment to God (Proverbs 9:10: “The fear of the LORD is the beginning of wisdom, and knowledge of the Holy One is understanding”; Job 28:28: “And unto man he said, Behold, the fear of the Lord, that is wisdom; and to depart from evil is understanding”). In the 5th century, Saint Augustine confirmed this concept (“Believe and you will understand; faith precedes, intelligence follows”). Heir to a strong Christian culture, he retained the idea of the individual soul – “I, the soul” (animus, Confessions, X, 9, 6) – characterized essentially by its relationship with God the Creator: the soul is “capax Dei”, meaning “capable of God” (The Trinity, XIV, 4, 6, 8, 11). It embodies God, in whose image it was created. This vision would continue until the Enlightenment, when man, as an individual, was detached from God and took on a predominant role in the structuring of thought. Jean-Jacques Rousseau made man master of his own destiny, and went so far as to assert that “human intelligence has its limits, and not only can a man not know everything, he cannot even know in its entirety the little that other men know”. In other words, everything comes down to learning, which frees man from the yoke of God. The development of medicine and an increasingly pragmatic science reinforced this idea of a man-centered world. The heart lost its status as the place where the soul rests, to the benefit of the brain. Doctors such as Robert Bentley Todd, Jean-Martin Charcot, John Hughlings Jackson and Alois Alzheimer set out to dissect this supreme locus of thought. Neurons and their network organization became the representatives of human intelligence and the model for what our new century calls artificial intelligence! Idolized, perhaps excessively so, it remains an interesting idea to explore, having already applied its reasoning to concrete achievements and notions, of which this book is about to present a few examples and lines of thought. The aim of these examples is to give concrete expression to this new approach throughout life, particularly in the field of healthcare.

November 2023

PART 1Growing with Artificial Intelligence

Introduction to Part 1

Growing with artificial intelligence is akin to bottle-feeding with tablets, smartphones and similar iPhones, and expecting the world to digest data for us.

Where is the nearest diaper store? How do I educate my child? Where is the best place to buy the right vegetables? All questions delivered at the click of a button, thanks to Google’s massive algorithm factory. There is no need to look up to follow the signs, no need to consult a book to find the information you need, no need to test the vegetables to find out where they are best grown and no need to use your intuition to make your choices. “Google thinks for you!”

It is an immeasurable comfort, relieving any responsibility for mistakes and preventing guilt from gnawing away at the small area of reflection left by this assisted upbringing.

Cogito ergo sum, in the language of oblivion, “I think therefore I am” in the language of Molière and “Google thinks for me and I am what it tells me to be” in today’s language. And in tomorrow’s language: Google thinks for itself, and I no longer exist!

Without lapsing into the dinosaur syndrome, this Google meteorite could be the object of our destruction. We are thinking beings, and our thinking must be cultivated. Without cultivation, the land lies fallow and we can no longer be nourished. The mind is a piece of land that needs to be ploughed, seeded, watered and cultivated so that it can push its capacity to reason, perceive and create to the fullest. And the survival of our species depends on creation! But to cultivate, you have to learn, and learning requires understanding, which means making an effort. It is the equivalent of fetching all the pieces of a jigsaw puzzle to piece together the image you have chosen. This choice is itself dictated by our environment and what we have done with it, but on the scale of a local microcosm that kneads our originality and personality into it. When this microcosm becomes the whole world, with Google as a sanctuarized Gaia, it is more complicated than ever, on a human scale, to act upon it, and uniformity threatens the human species, and its lack of creativity with it! Education therefore has a key role to play in curbing this standardization. It is supposed to enable the mind to incorporate all that its senses allow it to know about the world, by prioritizing it and providing it with enough critical sense to judge the relevance of what its mind receives from the outside world. In this way, man will be able to use this artificial intelligence for his own creativity, instead of swallowing it like an alcoholic swallowing whisky or cologne!

Artificial intelligence can be an interesting concept and technique for probing certain mysteries and enriching skills, but you still have to “be a man my son!”

Note

Introduction written by Marianne SARAZIN.

1From Human to Artificial Intelligence

1.1. The different forms of intelligence

1.1.1. Human intelligence typologies

DEFINITION.– Etymology of the word intelligence: Borrowed from the Latin intelligentia, “faculty of perceiving, understanding, intelligence”, derived from intellĕgĕre (“to discern, grasp, understand”), composed of the prefix inter- (“between”) and the verb lĕgĕre (“to gather, choose, read”). Etymologically, intelligence consists of making a choice, a selection.

1.1.2. Artificial intelligence (AI) and human intelligence

One of the main challenges of artificial intelligence is that it can only simulate one or a limited number of forms of human intelligence! There are many forms of intelligence, as shown in Figure 1.1.

1.1.3. Object intelligence and human assistance

The human mind can use conventional statistical modeling to plot graphs with dozens of variables, but the decision resulting from these calculations remains more complex. A computer with artificial intelligence software can test and then sift through millions of variables with incredible speed, helping human intelligence to increase its analytical capacity.

Figure 1.1.The different forms of intelligence

1.2. History of “artificial” intelligence1

1.2.1. Mechanical forms

The first phase of this process was the desire to “mechanize” man and their thought processes. The idea was to represent man as a machine, in order to design learning machines.

The first works are attributed to the Catalan Ramon Lulle (1232–1315). Theological predicates, subjects and theories were organized in geometrical figures. These were considered perfect: circles, squares and triangles. By activating a mechanism consisting of dials, levers and cranks, a wheel could be turned, and the propositions of the thesis moved on guides to position themselves on points. These points were used to affirm that the proposition was positive, or true, or, conversely, negative, and therefore false. With the attributes of God and nine questions, the machine proved the existence of God!

Giordano Bruno2, Nicolas de Cues3, Athanasius Kircher4 and Gottfried Leibniz5 were influenced by the work of Ramon Lulle. Descartes (1596–1650) drew a parallel between the animal–machine of this period and the man–machine. The animal is nothing more than a perfected machine, a clockwork of metal parts and springs.

This era is represented by Maillard’s6 famous “swan”, capable of swimming (1733), and Jacques Vaucanson7 “duck”, which simulated digestion (1738). Julien Offray de la Mettrie (1709–1751) published a work entitled: L’Homme-machine8 (The Man–Machine).

Figure 1.2.Vaucanson’s duck, 1738 (model by Riskin Jessica). Illustration by a 19th-century inventor of his own version of a mechanical digestive duck. An arrow indicates where the main action takes place. Based on Chapuis and Édouard Gélis, Le Monde des Automates (The World of Automata).

1.2.2. The desire to model neurons and cybernetics

DEFINITION.– Physiologist: The physiologist conducts research into the functions of the body’s physiological systems: cardiovascular, digestive, musculoskeletal, nervous, respiratory, reproductive and urinary. He examines how these systems function, under both normal and pathological conditions. After identifying the various systems, he integrates them and builds models ranging from the molecular and cellular scales to those of tissues, organs and the whole organism.

DEFINITION.– Logician: specialist in logic as a discipline.

Warren Sturgis McCulloch, physiologist (1898–1969), joined forces with William Pitts, logician (1923–1969), to carry out work in the field of intelligence9. These two actors wanted to define a way of calculating nervous activity. McCulloch refers to Descartes in these works, as well as to Leibniz, whom he considers to be their precursors. They speak of artificial neurons and the perceptron. There was no learning mechanism. They just wanted to perceive “things”. Both authors refer to Turing and Church.

This work influenced the cyberneticists, whose leading figures were Norbert Wiener10 (1894–1964) and John von Neumann. They used an old Greek word to designate a new science. The relationship between Pitts and Weiner began as that of a teacher and a student before becoming tumultuous. It was for this reason that Pitts sank into alcohol and burned all his thesis documents11.

DEFINITION.– Cybernetics is the unification of the nascent fields of automation, electronics and mathematical information theory, as an entire theory of control and communication, both in animals and in machines.

The term cybernetics, from the Greek κῠβερνήτης (kubernêtês) meaning pilot or governor, was proposed in 1947 by American mathematician Wiener to promote this new discipline. The work of Wiener and von Neumann12 (1903–1957) was published in English and French. Von Neumann is considered the father of computer architecture and was also an economist, mathematician and physicist. In von Neumann’s 1945 report on building an intelligent machine, the only article cited is by McCulloch and Pitts. Today, the word cybernetics is considered obsolete. It was considered “satanic” by the Russians during the Stalin era, before Soviet researchers took up the discipline in the 1950s. Today, this root has resurfaced in other words such as cybersecurity, cyber spy or cyber physical system.

1.2.3. The arrival of computers

The Electronic Discrete Variable Automatic Computer (EDVAC) in 1949, proposed by von Neumann, was one of the very first electronic computers to work only in binary, while the Electronic Numerical Integrator and Computer (ENIAC), based on the German Z3, was less intelligent, but calculated with decimals.

The EDVAC consists of 6,000 vacuum tubes and 12,000 diodes, consumes 56 kW, occupies a surface area of 45.5 m² and weighs 7,850 kg. The ENIAC is a machine made up of 17,468 vacuum tubes, 7,200 crystal diodes, 1,500 relays, 70,000 resistors, 10,000 capacitors and around 5 million hand-welded connections. It weighs 30 tons and has a surface area of 167 m². Its power consumption is 150 kW.

The strength of these early systems was their speed of calculation, which is shown in Table 1.1 for a multiplication of two 10-digit numbers. These first calculators were built for military purposes, for instance, to calculate a ballistic trajectory. Physicians and pharmacists saw a use for these machines, notably in drug galenics and drug dosage calculations.

Table 1.1.Evolution of calculation speeds

Man alone

ENIAC

Current machines (2018)

5 minutes

1 millisecond

30 nanoseconds

Turing (1912–1954) is the father of connectivism. His report Intelligent Machinery dates from 194813. His aim was to organize a disorganized machine.

DEFINITION.– Connectivism is Turing’s idea that knowledge is distributed through a network of connections and, therefore, that learning consists of the ability to build and navigate these networks, and that this can be implemented in a machine.

In the 2000s, educationalists resurrected connectivism, which was subsequently reused by artificial intelligence. For pedagogues, connectivism is a theory of learning in the digital age, a response to the limitations of behaviorism, cognitivism and constructivism. It sought to explain the effects that technology has on the way we live, communicate and learn. This more contemporary approach was developed by George Siemens and Stephen Downes14. There is a kind of reflexivity between the way machines learn and the way educators want us to learn with machines.

Learning was particularly highlighted by the work of Frank Rosenblatt (1928–1971) and the creation of the real machine capable of recognition: the perceptron15. The approach was that of a probabilistic model. The MARK1 Perceptron was built on the principle of multi-layer linear threshold response arrays. It comprised a grid of 400 photocells. This matrix simulated a retina, and was used to discover objects presented to it. Rosenblatt’s algorithm is quite simple. The first step is to create a binary supervised classification of a variable as a function of a predictor. The second step is a stochastic search for a linear operator to act as a separator. Each time a new pair, individual and observation, is misclassified, the model is modified, which corresponds to the evolution of a frontier. Several runs are made on the database until convergence is reached.

This approach poses several problems:

It is possible not to have convergence if the data are not separable.

If the situation is linearly separable, but there are infinitely many solutions, the calculation may become impossible.

The case of “exclusive or” (XOR) is not allowed as it is not linearly separable.

Minsky and Paper have shown that the XOR case is not possible in this approach, as it cannot be represented by a neuron16. More complex systems must be used. This text helped to discredit artificial intelligence, and in particular the use of neural networks. For 10 years, these approaches have been sidelined by researchers.

1.2.3.1. Expert systems and the 1980s renaissance

The period 1970–1980 was a dark age for artificial intelligence before a revival in the 1980s. Connectionism was out of the limelight before reappearing in the 2000s. The novelty of the 1980s was expert systems and fuzzy set theory. Neural networks were reborn with Hopfield.

The Hopfield neural network is a model that is both recurrent and discrete-time. It uses a connection matrix that must be symmetrical and zero on the diagonal. This double constraint is one of the system’s limitations: only one neuron is updated per time unit, making it an asynchronous system. Proposed by physicist John Hopfield in 1982, this approach is at the root of the revival of neural networks. At the same time, gradient backpropagation algorithms appeared.

1.2.3.2. The 2010 revival

Vladimir Vapnik and Alexandre Cervonenkis implemented a mathematical theory of learning. They then demonstrated the criteria for convergence. Cervonenkis died young in 2014, and Vapnik immigrated to the United States to work at Bell-Labs, then NEC and Facebook. They are considered the fathers of the artificial intelligence revival.

The debates at the beginning of the period can be summarized in Table 1.2.

Table 1.2.Artificial intelligence debates since 2010

Path A

Path B

Statistics

Learning

Connexionism

Symbolic artificial intelligence

The idea is that we can predict without understanding. The model can be better if we do not use models based on the past.

The idea of machine learning is based on the use of both paths and the separation of learning sets and test sets.

1.2.4. Different uses of artificial intelligence in healthcare

Different approaches to artificial intelligence are possible in healthcare.

1.2.4.1. Neural networks

Neural networks are among the most highly developed forms of artificial intelligence today, for a number of reasons:

Users seem to understand what is going on. It is like a brain, but in software, which is not reality.

Neural networks were invented in the 1940s. They are not the newest techniques, so solution designers have experience and salespeople have arguments.

1.2.4.2. Intelligent agents

The notion of the intelligent agent had its heyday in the late 1990s before being replaced by learning methods.

DEFINITION.– An intelligent agent (AI) is a computer program that operates autonomously, perceiving its environment in the form of events, and acting on its environment with a precise objective defined by programming.

Increasingly, this type of software is complementing other AI technologies.

1.2.4.3. Knol extraction

DEFINITION.– A Knol is a simple unit of information. Knol is a combination of Knowledge and Mol (an abbreviation of mole, a unit of measurement for the quantity of chemical matter). A Knol is therefore a small amount of knowledge (a unit of knowledge), i.e. an article that you wish to share with other users.

Knol extraction collects “Knols” of text following a question. Knols can be extracted using a variety of technologies. In its Knol extraction of form, the simplest takes the form of what the name implies, “an 8-legged octopus”.

Knol extraction is used by search engines such as Google, for example, when we see “direct answers” in the results. It can also be used for counting purposes. For example, a “Google Flu” application visualized the development of influenza in certain countries, based on user queries.

1.2.4.4. Self-constructed ontologies

DEFINITION.– An ontology is the structured set of terms and concepts representing the meaning of a field of information, whether through the metadata of a namespace, or the elements of a knowledge domain.