TYPE Original Research

PUBLISHED 26 May 2023

DOI 10.3389/fmicb.2023.1179312

Antimicrobial resistance and

OPEN ACCESS machine learning: past, present,
and future
EDITED BY
Alberto Antonelli,
University of Florence, Italy

REVIEWED BY
Ahmed S. Khairalla, Faiza Farhat 1, Md Tanwir Athar 2, Sultan Ahmad 3,4,
University of Regina, Canada Dag Øivind Madsen 5* and Shahab Saquib Sohail 6*
Marco Coppi,
University of Florence, Italy 1
Department of Zoology, Aligarh Muslim University, Aligarh, India, 2 Department of Pharmacognosy and
*CORRESPONDENCE
Pharmaceutical Chemistry, College of Dentistry and Pharmacy, Buraydah Colleges, Buraydah,
Dag Øivind Madsen Al-Qassim, Saudi Arabia, 3 Department of Computer Science, College of Computer Engineering and
dag.oivind.madsen@usn.no Sciences, Prince Sattam bin Abdulaziz University, Al-Kharj, Saudi Arabia, 4 Department of Computer
Shahab Saquib Sohail Science and Engineering, University Center for Research and Development (UCRD), Chandigarh
shahabssohail@jamiahamdard.ac.in University, Mohali, Punjab, India, 5 School of Business, University of South-Eastern Norway, Hønefoss,
Norway, 6 Department of Computer Science and Engineering, Jamia Hamdard University, New Delhi,
RECEIVED 03 March 2023 India
ACCEPTED 15 May 2023
PUBLISHED 26 May 2023

CITATION Machine learning has become ubiquitous across all industries, including the
Farhat F, Athar MT, Ahmad S, Madsen DØ and relatively new application of predicting antimicrobial resistance. As the first
Sohail SS (2023) Antimicrobial resistance and
bibliometric review in this field, we expect it to inspire further research in this area.
machine learning: past, present, and future.
Front. Microbiol. 14:1179312. The review employs standard bibliometric indicators such as article count, citation
doi: 10.3389/fmicb.2023.1179312 count, and the Hirsch index (H-index) to evaluate the relevance and impact of the
COPYRIGHT leading countries, organizations, journals, and authors in this field. VOSviewer and
© 2023 Farhat, Athar, Ahmad, Madsen and Biblioshiny programs are utilized to analyze citation and co-citation networks,
Sohail. This is an open-access article
distributed under the terms of the Creative
collaboration networks, keyword co-occurrence, and trend analysis. The
Commons Attribution License (CC BY). The United States has the highest contribution with 254 articles, accounting for over
use, distribution or reproduction in other 37.57% of the total corpus, followed by China (103) and the United Kingdom (78).
forums is permitted, provided the original
author(s) and the copyright owner(s) are
Among 58 publishers, the top four publishers account for 45% of the publications,
credited and that the original publication in this with Elsevier leading with 15% of the publications, followed by Springer Nature
journal is cited, in accordance with accepted (12%), MDPI, and Frontiers Media SA with 9% each. Frontiers in Microbiology is the
academic practice. No use, distribution or
reproduction is permitted which does not
most frequent publication source (33 articles), followed by Scientific Reports (29
comply with these terms. articles), PLoS One (17 articles), and Antibiotics (16 articles). The study reveals a
substantial increase in research and publications on the use of machine learning
to predict antibiotic resistance. Recent research has focused on developing
advanced machine learning algorithms that can accurately forecast antibiotic
resistance, and a range of algorithms are now being used to address this issue.

KEYWORDS

antimicrobial resistance, antibiotic resistance, machine learning, deep learning,

bibliometric analysis, healthcare

1. Introduction
Since Fleming’s ground breaking discovery of the first antibiotic, Penicillin, in 1928, a new
era of medicine emerged, leading to the development of other antibiotics that are now commonly
used to combat bacterial infections. Nevertheless, the recurring use of these antibiotics has
resulted in the emergence of drug-resistant bacteria, posing a major global threat. In many cases,
clinicians must administer broad-spectrum antibiotics based only on symptoms, or wait for
lengthy laboratory test results before providing the appropriate treatment. These antibiotic-
resistant bacteria are becoming more prevalent every day, leading to a significant public health

Frontiers in Microbiology 01 frontiersin.org

Farhat et al. 10.3389/fmicb.2023.1179312

issue and placing an immense financial burden on the healthcare Research question Objective
system. Additionally, research suggests that the recent COVID-19 RQ1 Which authors lead the research on To identify the most prolific
pandemic has led to an increase in antimicrobial resistance (AMR) machine learning applications in authors and their
due to the widespread use of disinfectants and antibiotics (Wilson AMR and what are their collaborating network
et al., 2020). The World Health Organization (WHO) and the US collaborating network?
Centers for Disease Control (CDC) estimate that there are at least 2
RQ2 Which articles are cited the most and To identify the most cited
million cases of antibiotic-resistant infections globally each year, with
which journals are publishing the publication and the most
23,000 of them resulting in fatalities, and the global cost of medical
most on the prediction of AMR contributing journals
care ranging from $20 to $35 billion (Centers for Disease Control and
using machine learning algorithms? publishing in the field
Prevention, 2013). By 2050, the yearly death toll from infections
brought on by bacteria resistant to antibiotics would reach 10 million, RQ3 Which organizations and countries To identify the organizations

exceeding the annual death toll from cancer (Murray et al., 2022). A contribute most to the scientific and countries mostly

2014 study found that minor bacterial infections during hip production? working on the subject

replacement surgery could increase the fatality risk by as much as 30% RQ4 What are the trending keywords and To identify the trending
(Berstock et al., 2014). which keywords are most cited in the topics and keywords in the
To reduce drug abuse, it is essential to quickly and precisely literature on the use of machine related research filed
identify strains that are resistant to antibiotics. Detecting learning in AMR identification or
antimicrobial resistant (AMR) infections is often a time-consuming prediction?
and tedious process because some organisms are exact copies of RQ5 What is the significant research To know the type of
each other or cannot be grown in the lab. Due to advancements in works going on related to AMR in research, methods applied,
computer-aided drug design and novel in silico methods, the the view of machine learning applications and results
creation of antibacterial drugs recently underwent a paradigm applications? obtained
shift. Recent studies have demonstrated the efficacy of machine
learning methods in predicting AMR in a variety of bacterial
strains (Liu et al., 2020). Additionally, it would be extremely time-
efficient if machine learning algorithms could be used instead of
traditional culturing studies to identify the various mechanisms 2. Results
underlying antimicrobial resistance, such as efflux pumps, target
modifications, and enzymatic inactivation, and predict resistance In this section we have elaborated the different analyses for
in bacterial strains. After being trained on whole-genome bibliometric study, well supported by appropriate diagrams and
sequencing, a number of machine-learning algorithms, including tables. These analyses include trend analysis of publications, types
support vector machines (SVM), logistic regression models (LR), of documents published and the renowned publishers, most
and random forests (RF), have been shown to demonstrate contributing journals, most contributing countries, most
excellent accuracy for predicting AMR (Yang et al., 2018; Liu et al., contributing institutions, most prolific authors and their scientific
2020). Deep learning algorithms’ effectiveness in predicting new contribution over time, collaboration network analysis, citations
antibiotics, AMR genes, and AMR peptides has also recently been analysis, co-citation and bibliographic coupling of authors, and
proven (Arango-Argoty et al., 2018; Stokes et al., 2020). co-occurrence of keywords.
In addition to empirical research, review articles, including
concise reviews and systematic reviews, have recently been published
(Arango-Argoty et al., 2018; Weis et al., 2020; Li et al., 2021; Lv et al., 2.1. Trend analysis of publications
2021; Kim et al., 2022; Popa et al., 2022), but no bibliometric analysis
has been conducted on this specific topic. This study offers a Figure 1 displays the trends in article production from 2000 to
quantitative perspective based on bibliometric data in light of the 2022. It was noted that before 2012, there was an average of just 1
significance of the research on machine learning uses in AMR. The study published annually on the topic of machine learning to combat
researcher in this field would be able to choose the research direction AMR. This is despite the fact that the field’s publication began in 2000.
for potential future projects with the help of a thorough bibliometric With an average of up to 7 papers published per year between 2013
analysis of the machine learning applications used in AMR because and 2015, the annual publications showed a relatively slow progression.
the bibliographic study not only highlights the most illustrious An inflection point can be seen in 2017, when there were more
researchers, institutions, and countries in a particular field, but also publications than 20. After that, there were consistently more
represents the most popular research topics in that field (Yu et al., publications, particularly from 2019 onward, with more than 80% of
2020). Through an analysis of collaborations between authors, the papers published in the previous 4 years. Such a growth profile
institutions, and countries, it is possible to establish a bibliometric and demonstrates how this area is gaining ground and impact in the
intellectual foundation for a specific area of interest (Farhat et al., scientific academia. After 2019, the number of articles published
2023). According to our research, this is the first bibliometric analysis annually quadrupled to 122 on average, reaching a peak in 2022 (183
on the topic that we are aware of, so we hope that it will encourage articles/year). We can deduce that machine learning approaches are
researchers to think about fresh directions for their future work in the becoming increasingly promising in the identification or prediction
field. To achieve the aim of the study, research questions were prepared of AMR and would continue to flourish in the near future based on
with clear objectives as follows: the trend of growth that has been observed.

Frontiers in Microbiology 02 frontiersin.org

Farhat et al. 10.3389/fmicb.2023.1179312

FIGURE 1
Trend analysis of article published/year.

2.2. Types of documents published and the 2.4. Most contributing countries
renowned publishers
The total number of articles contributed by 74 different countries
Among the 676 publications that were obtained, the majority is 676, with Table 2 displaying the top 10 countries based on article
of them were empirical papers, constituting 78.7% (532 articles) of count. The United States has the highest contribution, with 254
the corpus. Review articles made up 12.87% (87 articles), while articles, accounting for over 37% of the entire corpus. The
proceeding papers accounted for 7.25% (49 publications), totaling United Kingdom (78 articles, 11.53%) and China (103 articles,
98.82% of the publications (as depicted in Figure 2A). Despite a 15.23%) rank second and third, respectively, in terms of contribution.
considerable number of review articles published, no bibliometric In addition, the United States has a more considerable global academic
analysis has been released so far. Presently, out of the 58 publishers, impact than any other country, as demonstrated by the highest citation
only six have contributed to 60% of the corpus. Among them, count (5466). Canada ranks second in citations with 1879. Moreover,
Elsevier leads with 15% (104 articles), followed by Springer Nature technologically advanced countries such as the United Kingdom,
with 12% (81 articles), MDPI (61 articles), and Frontiers media SA China, and Germany have made significant contributions, with
(57 articles), each contributing 9% of all published articles (as corresponding citation counts of 1250, 1166, and 1000, respectively.
illustrated in Figure 2B).

2.5. Most contributing institutions

2.3. Most contributing journals
A total of 1288 institutions have contributed to the 676
The presence of 676 publications in 310 different journals publications, with the University of California participating in the
highlights the extensive diversity of literature on machine learning most papers (35). Harvard University (except school of medicine)
and its application in AMR. As demonstrated in Table 1, the top 10 (29), Harvard Medical School (21), the US Department of Energy
journals in terms of article count make up 25.14% of the corpus (21), and Imperial College London (20) make up the top 5 universities
and 21.43% of the total citations. Among these, in terms of based on article count (Table 3). Harvard Medical School has received
publishing output, Frontiers in Microbiology has emerged as the the most citations, cited 2148 times, followed by Harvard University
most productive journal with 33 published articles, followed by (1359) (All schools except Medicine) and the University of California
Scientific Reports (29 articles), PLoS One (17 articles), and (1310). In terms of average citations per article, Harvard Medical
Antibiotics (16 articles) respectively. However, in terms of citation School takes the top position with 102.21, followed by the
impact, Scientific Reports is leading the pack with 591 citations, Massachusetts Institute of Technology with 74.4.
while Frontiers in Microbiology has received 393 citations,
positioning it as the second-most cited journal. If we rank the
journals based on their H-index (using https://www.scimagojr. 2.6. Most prolific authors and their
com/journalrank.php), Bioinformatics ranks first with an H-index scientific contribution over time
of 415, PLoS One secures the second position with an H-index of
367, and Antimicrobial Agents and Chemotherapy comes in third A total of 3913 authors have contributed to the 676 articles about
with an H-index of 269. the subject of the study. Based on the number of articles published on

Frontiers in Microbiology 03 frontiersin.org

Farhat et al. 10.3389/fmicb.2023.1179312

FIGURE 2
(A) Types of articles published. (B) Top 6 contributing publishers.

TABLE 1 Top 10 leading journals on the basis of article published.

Journal Article Citation Average citation H-index Publisher

count count per article (ACPA)
Frontiers in Microbiology 33 393 11.9 166 Frontiers Media SA

Scientific Reports 29 591 20.37 242 Springer Nature

PLoS one 17 281 16.5 367 Public Library of science

Antibiotics 16 52 3.25 47 MDPI

BMC Bioinformatics 15 168 11.2 218 BMC

Briefings in Bioinformatics 15 121 8.06 121 Oxford University Press

mSystems 15 121 8.06 54 American society for Microbiology

Antimicrobial Agents and Chemotherapy 11 239 21.72 269 American society for Microbiology

PeerJ 10 41 4.1 83 PeerJ Inc.

Bioinformatics 9 290 32.22 415 Oxford University Press

the topic, the top 10 authors are depicted in Figure 3. Wang, Hsin-Yao The current study aimed to evaluate the degree of intellectual
is ranked first with 11 articles, while Huleihel, Mahmoud, and Salman, collaboration between authors, institutions, and countries.
Ahmad follow closely with 10 articles each. Additionally, Lapidot,
Itshak; Chung, Chia-Ru; Davis, James J.; Horng, Jorng-Tzong; and Lu, 2.7.1. Author collaboration
Jang-Jih have contributed 9 publications on machine learning The author collaboration network analysis revealed a significant
applications in AMR. Looking at the authors’ overall scientific output, collaboration network consisting of 158 productive authors, with at
Wang, Hsin-Yao, Chung, Chia-Ru, Horng, Jorng-Tzong, Lu, Jang-Jih, least three articles each, connected by a total of 83 collaborations and
and Lee, Tzong-Yi have been the most active authors in the past a link strength of 203 (Figure 4). The overall link strength indicates the
2 years, while Huleihel, Mahmoud, and Salman, Ahmad have strength of all collaborations between a particular researcher and
contributed the majority of articles from 2018 to 20, as shown in other researchers. Davis James J has the highest link strength of 6 and
Figure 3. is the closest collaborator to Nguyen Marcus and Olson Robert. The
second-most collaborative partners, with a link strength of 5 each, are
Shukla Maulik and Davis James J, as well as Shukla Maulik and Olson
2.7. Collaboration network analysis Robert. Davis James J and Nguyen Marcus have worked with the most
authors, a total of 14. Additionally, eight authors, including Shukla
Extensive collaboration between academics and researchers at all Maulik; Olson Robert; Mao Chunhong; Wattam Alice R; Vanoeffelen
levels is a prominent feature of academic research and is often used as Margo; Yoo Hyunseung; Aytan-Aktug Derya; and Brettin Thomas,
an indicator for evaluating research collaboration (Trotta et al., 2022). have exhibited strong collaboration networks by collaborating with 12

Frontiers in Microbiology 04 frontiersin.org

Farhat et al. 10.3389/fmicb.2023.1179312

TABLE 2 Top 10 countries on the basis of article published.

(Figure 6). The United States was found to lead the world in
Country Article Total Average citation collaborative networks, with the highest number of collaborations
count citations per article (ACPA) with 45 countries. The United Kingdom came in second with 43
USA 254 5,466 21.51 collaborators, followed by Germany (42), Italy (40), and China (39).
Among the collaborating partners of the United States, China had the
China 103 1,166 11.32
highest number of publications with 18, followed by Canada and the
UK 78 1,250 16.02
United Kingdom with 13 each. The second-most frequent
Canada 40 1,879 46.97 collaborators, with 11 link strengths each, were the United States,
Australia 38 543 14.28 Denmark, and Australia. The United States and Spain were third with
Germany 38 1,000 26.31
a total link strength of 10.

India 34 340 10

Spain 33 553 16.75 2.8. Citations analysis

Switzerland 28 469 16.75

Italy 25 383 15.32 Citation analysis is a technique used to assess the impact and
importance of a research work by counting the number of times it has
TABLE 3 Top 10 institutions on the basis of article published. been cited by other authors (Patience et al., 2017). This method is
commonly used in bibliometric research to measure the
Organization Article Total Average interconnectedness of publications based on their citation relationships.
count citations citation Using VOSviewer for citation analysis, it was found that 246 articles
per article
received at least 10 citations, while 26 papers received more than 50
University of California 35 1,359 38.82 citations, and 21 articles received over 100 citations. The most
Harvard University (All 45.17 frequently cited work, with a total of 500 citations, is the empirical
schools except Medicine) 29 1,310 study by Stokes et al. (2020). Tyers and Wright (2019), Cherkasov et al.
Harvard Medical School 21 2,148 102.21 (2009), Arango-Argoty et al. (2018), and Fjell et al. (2009) are the next
most cited authors, with 317, 274, 257, and 212 citations, respectively
United States Department of 35.71
(Figure 7). The top 10 most cited articles are listed in Table 4.
Energy 21 750

Imperial College London 20 436 21.8

Chinese Academy of Sciences 18 210 11.66 2.9. Co-citation and bibliographic coupling
Massachusetts Institute of 74.4 of authors
Technology 15 1,116

Cornell University 14 312 22.28

Co-citation analysis is a bibliometric method that measures the
frequency with which two documents are cited together in other
University of London 13 278 21.38
documents. When two documents are cited in the reference list of a
University of Oxford 13 243 18.69 third document, they are considered to be co-cited, and the strength
of their co-citation relationship indicates their semantic relatedness.
distinct authors. Furthermore, Xia FangFang and Stevens Rick have Bibliographic coupling, on the other hand, is an alternative method
each collaborated with 11 authors. that determines the degree to which two works address the same topic
by identifying whether they cite one or more of the same references.
2.7.2. Institutions The strength of the bibliographic coupling relationship is determined
In VOSviewer, the collaboration analysis of research institutions by the number of shared citations between the two works, and it can
revealed that the largest collaborating network was formed by 158 help researchers identify earlier, relevant, and similar studies
prominent institutions that had published at least three articles, with (Figures 8, 9).
a total of 606 linkages and 926 link strength (Figure 5). The In a co-citation analysis conducted using VOSviewer, it was found
Massachusetts Institute of Technology has the highest number of that 382 authors were co-cited at least 10 times. Tuchin, W. and
collaborators (37), followed by Weill Cornell Medical College (25), Genina, E.A. had the highest number of co-citations, with 216 link
Oxford University (24), Wellcome Sanger Institute, and University of strengths, followed by Yount, NY; Lee, E.Y.; and Yeaman, MR; with 85
Maryland (22), Harvard University, Pasteur Institute of Korea, and 78 link strengths, respectively. Lee EY and Yeaman MR came in
University of Minnesota, and Astar (21). The Chinese University of fifth place, with a link strength of 72. In the co-citation network
Hong Kong, Asia University, and the Chang Gung Memorial Hospital analysis, Breiman, I was found to have the most co-cited authors, with
are the closest partners, each with 7 link strengths. Harvard Medical 324, followed by Kilkauer T and Li H with 312 and 294 distinct
School has 6 link strengths with the Massachusetts General Hospital co-cited authors, respectively. As for bibliographic coupling, Huleihel,
and Harvard University. Mahmoud, and Salman Ahmad had the strongest link strength of
1571, indicating a potential similarity between the two works. Lapidot,
2.7.3. Countries Itshak, followed by Huleihel, Mahmoud, and Salman Ahmad, had
In the analysis of collaboration among contributing countries, 52 1409 link strength each, while Salman Ahmad; Riesenberg, Klaris; and
productive countries with at least three articles each were identified Sharaha, Uraib had a link strength of 1406 each.

Frontiers in Microbiology 05 frontiersin.org

Farhat et al. 10.3389/fmicb.2023.1179312

FIGURE 3
Top 10 authors on the basis of article count and author’s scientific production over time.

2.10. Co-occurrence of keywords: keywords were categorized into 18 clusters, with the largest cluster
centered around the term “machine learning,” along with other
The analysis of keyword co-occurrence in a corpus provides relevant algorithmic terms and specific AMR techniques (see
insight into the most commonly used keywords and research Figure 10). The second-largest cluster focused on machine learning
trends in a particular field, such as the study of AMR prediction applications for AMR prediction, while the third-largest cluster
through machine learning (Yu et al., 2020; Farhat et al., 2023). emphasized various prediction models and techniques.
Using the VOSviewer program, 3061 keywords were initially The overlay visualization of the keyword evolution over time
identified, but only 83 relevant keywords related to machine (Figure 11) revealed that older machine learning techniques, such as
learning for AMR prediction were retained for analysis. These “supervised learning,” “Bayesian network,” and “support network

Frontiers in Microbiology 06 frontiersin.org

Farhat et al. 10.3389/fmicb.2023.1179312

FIGURE 4
Collaboration network of authors.

FIGURE 5
Collaboration network of institutions.

machine,” were commonly used at the start of the study. However, indicating a shift towards improving AMR gene prediction accuracy.
after 2018, newer techniques like “deep learning,” “artificial neural These trends suggest that researchers are exploring various methods
network,” and “computer-aided drug design” gained popularity, to combat the growing problem of AMR and develop novel strategies.

Frontiers in Microbiology 07 frontiersin.org

Farhat et al. 10.3389/fmicb.2023.1179312

FIGURE 6
Collaboration network of countries.

FIGURE 7
Citation analysis of articles published.

3. Discussion Wiegand et al., 2008). The fact that the machine learning models gain
the AMR information from the data makes them stand out among
In the last 2 years, machine learning has been applied in various others that do not require a prior knowledge. Given that the model
healthcare branches (Mufti et al., 2021; Fatima et al., 2022; Naaz et al., can be interpreted, scientists can use these models to not only
2022; Rashid Irshad et al., 2023) among which the growing problem anticipate AMR but also to identify previously unidentified AMR
of AMR is one of the trending research areas (Cockerill, 1999). pathways. Machine learning algorithms have just started to be used in
Traditional antimicrobial susceptibility testing procedures call for the field of AMR from 2000 and until 2006, very discrete publications
isolating bacteria from human specimens using culture techniques, emerged. It is only after 2016 when the studies are more focused in
after which the recovered bacteria are subjected to various antibiotic this particular area.
concentrations in numerous assays to determine which concentration Bibliometric assessments can assist to forecast future trends in a
limits growth (Wiegand et al., 2008; Nguyen et al., 2018). These certain academic topic by providing a high-level overview of its
methods are time-consuming and expensive (Cockerill, 1999; current state (Trotta et al., 2022). Therefore, in this study,

Frontiers in Microbiology 08 frontiersin.org

Farhat et al. 10.3389/fmicb.2023.1179312

TABLE 4 Top 10 cited articles with their citation count, document type, their source, and country of origin.

No. Title Total Article journal Country of

citation type first author
1 A deep learning approach to antibiotic discovery 500 Empirical Cell USA

2 Drug combinations: a strategy to extend the life of antibiotics in the 21st century 317 Review Nature Canada

3 Use of artificial intelligence in the design of small peptide antibiotics effective against 274 Empirical ACS chemical Canada
a broad spectrum of highly antibiotic-resistant superbugs Biology

4 DeepARG: a deep learning approach for predicting antibiotic resistance genes from 257 Empirical Microbiome USA
metagenomic data

5 Identification of novel antibacterial peptides by chemoinformatic and machine 212 Empirical Journal of Medicinal Canada
learning Chemistry

6 Antibiotic-induced alterations of the murine gut microbiota and subsequent effects 167 Empirical mBio USA
on colonization resistance against Clostridium difficile

7 Bacterial metabolism and antibiotic efficacy 160 Review Cell Metabolism USA

8 Emerging technologies for molecular diagnosis of sepsis 147 Review Clinical Biology USA
Reviews

9 Deep learning improves antimicrobial peptide recognition 141 Empirical Bioinformatics USA

10 A white-box machine learning approach for revealing antibiotic mechanisms of action 136 Empirical Cell USA

we investigated publications on machine learning applications in trailing only the United States. The United Kingdom, China, and
AMR with a focus on the countries, organizations, journals, authors, Germany have also made notable contributions, with corresponding
and trending keywords related to this subject. A total of 676 journal citation counts of 1250, 1166, and 1000. Collaboration networks are
articles, available in the Web of Science database, were published from essential in advancing the field of study, and the United States has
2000 to 2022. The bibliometric data revealed that as of January 2023, collaborated with 45 countries, including China, Canada, and the
3913 active authors from 74 different countries had contributed to this United Kingdom, among others. The United Kingdom, Germany, Italy,
area of research. There were 1288 organizations involved in publishing and China also have extensive collaborative networks, with 43, 42, 40,
these articles in 310 different journals. A tremendous rise in this area and 39 collaborators, respectively, which aids in global information flow.
of research represents the increasing interest in machine learning According to the number of articles, University of California,
application in AMR. Here, we discovered that over 74 countries have Harvard University, Harvard Medical School and the US
made contributions to the field and the number of papers devoted to Department of Energy and Imperial College London are the top five
this topic is increasing every year. In light of these findings, we predict organizations engaged in research on the topic. When institutions
that numerous in-depth research examining various machine learning are examined for collaboration, it is found that the closest partners
algorithms to predict AMR in different species using the publicly are either those in the same country or those in nearby countries.
available databases would be published in the upcoming years. Initiatives to forge closer ties could aid this field’s future
The topic of applying machine learning in AMR is highly advancement due to the comparatively low level of collaboration
researched, with 676 publications spread out over 310 journals, across the institutes. The exchange of information would
suggesting a merging of the disciplines of life science and computer be improved and advancement in this field of study would
science. This is evident from the fact that the journals publishing on be encouraged with more international collaboration.
this topic belong to both fields. The 310 journals are published by 58 When the author’s overall scientific output is carried out it is
different publishers, but only six of them, namely Elsevier, Springer observed that Huleihel, Mahmoud, and Salman, Ahmad have
Nature, MDPI, Frontiers Media SA, American Society for contributed the majority of articles between 2018 to 20 while Wang,
Microbiology, and Oxford University Press, account for 60% of the Hsin-Yao, Chung (11 articles), Chia-Ru, Horng, Jorng-Tzong, Lu,
publications. Notably, the top 10 journals with the most publications Jang-Jih, and Lee, Tzong-Yi are the five most active authors from the
are mostly owned by these publishers, except for Elsevier, which is the previous 2 years with 9 articles each. However, Davis James J, Nguyen
top publisher but does not feature in the top 10 journals. Among the Marcus, Shukla Maulik, Olson Robert, and others have the strongest
top 10 journals, the American Society for Microbiology and Oxford network of collaboration.
University Press each contribute two, while Springer Nature, MDPI, Keyword co-occurrence analysis paves a way for researchers as
and Frontiers Media SA each contribute one. they navigate across literature and highlights significant research
Measuring a country’s academic influence can be done by interests and topics in particular field. An illustration of co-occurrence
examining the total number of papers produced and the total number network of keywords was generated with the relevant topics to
of citations it receives. The United States leads the world in academic machine learning in AMR. The most important and heavily weighted
influence in the field of machine learning and AMR, with 254 keywords in this network are probably research hotspots of the
publications and 5466 citations, accounting for over 37% of the total concerned subject, where there is still a demand for research on these
corpus. Canada, despite having only the fourth-highest number of topics and the related research directions. The evolution of keywords
articles, received the second-highest number of citations (1879), through time may be seen by overlay visualization analysis, and it also

Frontiers in Microbiology 09 frontiersin.org

Farhat et al. 10.3389/fmicb.2023.1179312

FIGURE 8
Co-citation analysis of authors.

FIGURE 9
Bibliographic coupling analysis of authors.

displays keywords that have received more attention from researchers, which proved highly effective in predicting the activity of virtual
indicating they are significant active research areas that need further peptides against multidrug-resistant “Superbugs.” Fjell et al. (2009)
investigation. The co-occurrence study shows that different machine reports the successful in silico screening for potent antibiotic peptides
learning algorithms are being used to either detect AMR or find new which were found to have activities comparable or superior to those
alternatives. Many methods, including supervised learning, ensemble of conventional antibiotics Tyers and Wright (2019) and Arango-
learning, Bayesian hyperparameter optimization, quantitative Argoty et al. (2018) propose strategies for addressing antibiotic
structure–activity relationships (QSAR), and support vector machines resistance through combinations of antibiotics and non-antibiotic
(SVM), have been around for a while Stokes et al. (2020). activity-enhancing compounds and through the use of deep learning
The top five cited articles highlight different approaches to approaches for comprehensive global monitoring of antibiotic
addressing the urgent need for new antibiotics in the face of increasing resistance genes in environmental media. Overall, these studies
antibiotic resistance. For example Stokes et al. (2020), Tyers and demonstrate the potential of machine learning and other innovative
Wright (2019), and Cherkasov et al. (2009) are primarily focused at approaches for discovering novel antibiotics and combating antibiotic
using machine learning techniques to identify novel antibacterial resistance. Additionally, efforts to curb the overuse and misuse of
molecules or peptides. Halicin a structurally distinct molecule with antibiotics and promote responsible antibiotic stewardship are also
broad-spectrum antibacterial activity against resistant pathogens is crucial in addressing the global challenge of antibiotic resistance
reported in Stokes et al. (2020). Cherkasov et al. (2009) describes the Several machine learning algorithms, including decision support
development of quantitative in silico models of antibiotic activity systems, random forest, rapid detection, decision trees,

Frontiers in Microbiology 10 frontiersin.org

Farhat et al. 10.3389/fmicb.2023.1179312

FIGURE 10
Co-occurrence of total keywords.

FIGURE 11
Overlay visualization of keywords.

Frontiers in Microbiology 11 frontiersin.org

Farhat et al. 10.3389/fmicb.2023.1179312

high-throughput screening, and multivariate analysis, have been keywords. The pattern of articles published from 2000 to 2022 is
applied in AMR detection (Schubert et al., 2015; Macesic et al., 2020; shown in a trend analysis of publications for each year. By employing
Melo et al., 2021; Wang et al., 2022; Yasir et al., 2022). The growth of the the full counting approach, which only displays the elements related
machine learning mechanism, through the use of deep learning models with one another, co-authorship analysis, citation analysis,
such as artificial neural networks, has advanced machine learning- bibliographic coupling is all carried out. Additionally, co-occurrence
driven AMR research (Cherkasov et al., 2009; Fjell et al., 2009; Arango- of total keywords, author’s keywords, and indexed keywords are also
Argoty et al., 2018; Popa et al., 2022). Current methods include performed to analyze the research trend and trending topics in
automated antibiotic discovery, logistic regression models, and k-means machine learning approaches used in AMR. Larger circles denote
clustering (Veltri et al., 2018; Barlandas-Quintana and Martinez- more partnerships and stronger ties between people who collaborate
Ledesma, 2020; Valizadehaslani et al., 2020; Marini et al., 2022). These more regularly in various illustration maps of the collaboration
algorithms train the machine learning system to identify unique network of authors, institutions, and countries while larger circle
features and resolve the complicated network, enabling computer-aided denotes more citations and links with other co-cited partners in the
drug design in the field of AMR (Sinha et al., 2018; Tyers and Wright, citation and bibliographic coupling maps. The keyword map using the
2019). These findings provide a foundation for identifying current complete counting method groups related keywords together and
scientific hotspots and guiding future research endeavors. gives equal weight to each co-occurrence link, making the circles
around the terms with higher frequency larger.

4. Methodology
5. Conclusion
4.1. Data extraction
The present study utilized literature obtained from the Web of
We used the Web of Science database in January 2023 to find Science database to showcase the advancements made in scientific
articles with the search queries “antimicrobial resis*” or “antibacterial knowledge of machine learning and AMR from 2000 to 2022. The
resis*” AND “machine learning.” The identified articles that contain study offers a comprehensive analysis of the existing literature,
the search query in the title, abstract, or keywords are chosen for the highlighting significant authors and scientific collaborations, as well
bibliometric review study. Only English-language journal articles that as identifying the most common topics and keywords using keyword
were published up through January 2023 were included in the search co-occurrence analysis and overlay visualization. The results provide
results. Thereafter, 681 articles were retrieved. Then, manual screening valuable information on current journals, authors, and extensively
was performed to improve the quality of the data by reading the explored topics in this field, making it a useful guide for individuals
complete texts and looking at the articles’ content to weed out the interested in contributing to this discipline. The study also shows a
superfluous ones. Consequently, 676 articles were chosen for the significant increase in the amount of literature on the subject since
present analysis. The retrieved articles are evaluated using the 2018, primarily due to the introduction of new machine learning
following criteria: Organization, Countries/Regions, Journals, Total algorithms for AMR detection. These advancements have improved
Citations, Number of Publications Per Year, Keywords, and Number the accuracy of methods by utilizing various machine learning
of Publications Per Year. The complete records are downloaded for techniques and deep learning algorithms. This bibliometric study not
bibliometric analyses and imported into the Biblioshiny only emphasizes the current research directions but also suggests that
(Bibiliometrix) and VOSviewer software (Eck and Waltman, 2010; diverse methods may be feasible in the future, enhancing the
Aria and Cuccurullo, 2017; Moral-Muñoz et al., 2019). predictive efficiency and accuracy of AMR prediction through
Various indicators have been utilized in the literature during machine learning.
bibliometric analysis, including total article count, total citation count, The present analysis revealed useful information such as:
total link strength, average citations per article (ACPA), and Hirsch
index (H-index). These metrics are commonly used in bibliometric • Approximately 60% of the publications in the field of machine
studies, with the H-index being a widely recognized measure of learning application in AMR are published by Elsevier, Springer
research quality and quantity for authors and research avenues. ACPA Nature, MDPI, Frontiers Media SA, American Society for
is also widely accepted as a measure of research impact for individual Microbiology, and Oxford University Press.
works, authors, and publication avenues. Citation and co-citation are • The leading countries in this field are the United States,
useful for exploring the scientific impact and themes of the study United Kingdom, and China. The majority of the leading
under consideration, and co-authorship and co-occurrence have also institutes are located in the US and United Kingdom, with the
been investigated to analyze scientific collaboration. All of these University of California, Harvard University, and Harvard
indicators have been taken into account in this bibliometric study. Medical School being the top three.
• The most prolific journals are Frontiers in Microbiology,
Scientific Reports, PLoS One and Antibiotics.
4.2. Data analysis • The most contributing authors in this field are Hsin-Yao Wang,
Mahmoud Huleihel, and Ahmad Salman.
A total of 676 publications from 74 different countries and 1288 • The most commonly used algorithms in AMR prediction are
different institutions on the topic of AMR and machine learning have Bayesian hyperparameter optimization, quantitative structure–
been published in 310 different journals. These articles were activity relationships (QSAR), and support vector machines
contributed by 3913 authors, with a total of 10714 citations and 3061 (SVM) and logistic regression models.

Frontiers in Microbiology 12 frontiersin.org

Farhat et al. 10.3389/fmicb.2023.1179312

Data availability statement Conflict of interest

The original contributions presented in the study are included in The authors declare that the research was conducted in the
the article/supplementary material, further inquiries can be directed absence of any commercial or financial relationships that could
to the corresponding authors. be construed as a potential conflict of interest.

Author contributions Publisher’s note

FF: conceptualization, methodology, data curation, and writing— All claims expressed in this article are solely those of the authors
original draft preparation. FF and SS: formal analysis. DM: resources and do not necessarily represent those of their affiliated organizations,
and funding acquisition. MA, SA, DM, and SS: writing—review and or those of the publisher, the editors and the reviewers. Any product
editing. All authors contributed to the article and approved the that may be evaluated in this article, or claim that may be made by its
submitted version. manufacturer, is not guaranteed or endorsed by the publisher.

References
Arango-Argoty, G., Garner, E., Pruden, A., Heath, L. S., Vikesland, P., and Zhang, L. Marini, S., Oliva, M., Slizovskiy, I. B., Das, R. A., Noyes, N. R., Kahveci, T., et al. (2022).
(2018). Deep ARG: a deep learning approach for predicting antibiotic resistance genes AMR-meta: a k-mer and metafeature approach to classify antimicrobial resistance from
from metagenomic data. Microbiome 6:23. doi: 10.1186/S40168-018-0401-Z high-throughput short-read metagenomics data. GigaScience 11:giac029. doi: 10.1093/
gigascience/giac029
Aria, M., and Cuccurullo, C. (2017). Bibliometrix: an R-tool for comprehensive
science mapping analysis. J. Informet. 11, 959–975. doi: 10.1016/J.JOI.2017.08.007 Melo, M. C. R., Maasch, J. R. M. A., and de la Fuente-Nunez, C. (2021). Accelerating
antibiotic discovery through artificial intelligence. In communications biology.
Barlandas-Quintana, E. A., and Martinez-Ledesma, J. E. (2020). Detection of
Commun. Biol. 4:1050. doi: 10.1038/s42003-021-02586-0
carbapenems resistant k-mer sequences in bacteria of critical priority by the world
health organization (pseudomonas aeruginosa and acinetobacter baumannii). 2020 7th Moral-Muñoz, J. A., Herrera-Viedma, E., Santisteban-Espejo, A., Cobo, M. J.,
international conference on internet of things: systems, management and security. Daniel, P., and Aranda, S. (2019). Software tools for conducting bibliometric analysis in
IOTSMS 2020, 1–8. doi: 10.1109/IOTSMS52051.2020.9340214 science: an up-to-date review. Revista. Profesionaldelainformacio 29, 1699–2407. doi:
10.3145/epi.2020.ene.03
Berstock, J. R., Beswick, A. D., Lenguerrand, E., Whitehouse, M. R., and Blom, A. W.
(2014). Mortality after total hip replacement surgery: a systematic review. Bone Joint Res. Mufti, T., Gupta, B., Sohail, S. S., and Kumar, D. (2021). Contact tracing: a cloud based
3, 175–182. doi: 10.1302/2046-3758.36.2000239 architecture for safe Covid-19 mapping. 2021 international conference on computational
performance evaluation, ComPE 2021, 874–877.
Centers for Disease Control and Prevention. Antibiotic resistance threats in the
United States (2013). Available at: http://apps.who.int/iris/ Murray, C. J., Ikuta, K. S., Sharara, F., Swetschinski, L., Robles Aguilar, G., Gray, A.,
bitstream/10665/204419/1/9789241510134_eng.pdf?ua=1 (accessed May 1, 2023). et al. (2022). Global burden of bacterial antimicrobial resistance in 2019: a systematic
analysis. Lancet 399, 629–655. doi: 10.1016/S0140-6736(21)02724-0
Cherkasov, A., Hilpert, K., Jenssen, H., Fjell, C. D., Waldbrook, M., Mullaly, S. C., et al.
(2009). Use of artificial intelligence in the design of small peptide antibiotics effective Naaz, S., Khan, S. A., Siddiqui, F., Sohail, S. S., Madsen, D. Ø., and Ahmad, A. (2022,
against a broad spectrum of highly antibiotic-resistant superbugs. ACS Chem. Biol. 4, 2022). OdorTAM: technology acceptance model for biometric authentication system
65–74. doi: 10.1021/CB800240J/ASSET/IMAGES/MEDIUM/CB-2008-00240J_0001. using human body odor. Int. J. Environ. Res. Public Health 19:16777. doi: 10.3390/
JPG IJERPH192416777
Cockerill, F. R. (1999). Genetic methods for assessing antimicrobial resistance. Nguyen, M., Brettin, T., Long, S. W., Musser, J. M., Olsen, R. J., Olson, R., et al. (2018).
Antimicrob. Agents Chemother. 43, 199–212. doi: 10.1128/AAC.43.2.199 Developing an in silico minimum inhibitory concentration panel test for Klebsiella
pneumoniae. Sci. Rep. 8:421. doi: 10.1038/s41598-017-18972-w
Eck, N. V., and Waltman, L. (2010). Software survey: VOSviewer, a computer program
for bibliometric mapping. Scientometrics 84, 523–538. doi: 10.1007/s11192-009-0146-3 Patience, G. S., Patience, C. A., Blais, B., and Bertrand, F. (2017). Citation analysis of
scientific categories. Heliyon 3:e00300. doi: 10.1016/J.HELIYON.2017.E00300
Farhat, F., Sohail, S. S., Siddiqui, F., Irshad, R. R., and Madsen, D. Ø. (2023). Curcumin
in wound healing-a bibliometric analysis. Life 13:143. doi: 10.3390/LIFE13010143 Popa, S. L., Pop, C., Dita, M. O., Brata, V. D., Bolchis, R., Czako, Z., et al. (2022). Deep
learning and antibiotic resistance. Antibiotics 11:1674. doi: 10.3390/antibiotics11111674
Fatima, N., Agarwal, P., and Sohail, S. S. (2022). Security and privacy issues of
Blockchain Technology in Health Care—a Review. Lect. Notes Netw. Syst. 314, 193–201. Rashid Irshad, R., Alattab, A. A., Ali, O., Alsaiari, S., Saquib Sohail, S., Aziz, A., et al.
doi: 10.1007/978-981-16-5655-2_18/COVER (2023). An optimization-linked intelligent security algorithm for smart healthcare
organizations. Healthcare 11:580. doi: 10.3390/HEALTHCARE11040580
Fjell, C. D., Jenssen, H., Hilpert, K., Cheung, W. A., Panté, N., Hancock, R. E. W., et al.
(2009). Identification of novel antibacterial peptides by chemoinformatics and machine Schubert, A. M., Sinani, H., and Schloss, P. D. (2015). Antibiotic-induced alterations
learning. J. Med. Chem. 52, 2006–2015. doi: 10.1021/JM8015365/SUPPL_FILE/ of the murine gut microbiota and subsequent effects on colonization resistance against
JM8015365_SI_001.PDF Clostridium difficile. MBio 6:e00974. doi: 10.1128/MBIO.00974-15/
ASSET/562A2922-224B-4F61-A460-01169A8955C2/ASSETS/GRAPHIC/
Kim, J. I., Maguire, F., Tsang, K. K., Gouliouris, T., Peacock, S. J., McAllister, T. A., et al.
MBO0041523960006.JPEG
(2022). Machine learning for antimicrobial resistance prediction: current practice,
limitations, and clinical perspective. Clin. Microbiol. Rev. 35:e0017921. doi: 10.1128/ Sinha, M., Jupe, J., Mack, H., Coleman, T. P., Lawrence, S. M., and Fraley, S. I. (2018).
CMR.00179-21 Emerging technologies for molecular diagnosis of sepsis. Clin. Microbiol. Rev.
31:e00089-17. doi: 10.1128/CMR.00089-17
Li, X., Zhang, Z., Liang, B., Ye, F., and Antibiotics, W. G.-T. J. (2021). A review:
antimicrobial resistance data mining models and prediction methods study for Stokes, J., Yang, K., Swanson, K., Jin, W., Cubillos-Ruiz, A., Donghia, N. M., et al.
pathogenic bacteria. J. Antibiot. (Tokyo) 74, 838–849. doi: 10.1038/s41429-021-00471-w (2020). A deep learning approach to antibiotic discovery. Cells 180, 688–702.e13. doi:
10.1016/j.cell.2020.01.021
Liu, Z., Deng, D., Lu, H., Sun, J., Lv, L., Li, S., et al. (2020). Evaluation of machine
learning models for predicting antimicrobial resistance of Actinobacillus Trotta, F., Bracco, P., Zanetti, M., Chiari, W., Damayanti, R., Harapan, H., et al. (2022).
pleuropneumoniae from whole genome sequences. Front. Microbiol. 11:48. doi: 10.3389/ Trend of polymer research related to COVID-19 pandemic: bibliometric analysis.
FMICB.2020.00048/FULL Polymers (Basel) 14:3297. doi: 10.3390/polym14163297
Lv, J., Deng, S., and Health, L. Z.-B. (2021). A review of artificial intelligence Tyers, M., and Wright, G. D. (2019). Drug combinations: a strategy to extend the life
applications for antimicrobial resistance. Mednexus. Org. 3, 22–31. doi: 10.1016/j. of antibiotics in the 21st century. Nat. Rev. Microbiol. 17, 141–155. doi: 10.1038/
bsheal.2020.08.003 s41579-018-0141-x
Macesic, N., Bear Dont Walk, O. J., Peer, I., Tatonetti, N. P., Peleg, A. Y., and Valizadehaslani, T., Zhao, Z., Sokhansanj, B. A., and Rosen, G. L. (2020). Amino acid
Uhlemann, A.-C. (2020). Predicting phenotypic Polymyxin resistance in Klebsiella K-mer feature extraction for quantitative antimicrobial resistance (AMR) prediction by
pneumoniae through machine learning analysis of genomic data. mSystems 5:e00656-19. machine learning and model interpretation for biological insights. Biology 9, 1–92. doi:
doi: 10.1128/mSystems.00656-19 10.3390/biology9110365

Frontiers in Microbiology 13 frontiersin.org

Farhat et al. 10.3389/fmicb.2023.1179312

Veltri, D., Kamath, U., and Shehu, A. (2018). Deep learning improves antimicrobial Wilson, L. A., Rogers Van Katwyk, S., Fafard, P., Viens, A. M., and Hoffman, S. J.
peptide recognition. Bioinformatics 34, 2740–2747. doi: 10.1093/BIOINFORMATICS/ (2020). Lessons learned from COVID-19 for the post-antibiotic future. Glob. Health 16,
BTY179 1–3. doi: 10.1186/S12992-020-00623-X/METRICS
Wang, S., Zhao, C., Yin, Y., Chen, F., Chen, H., and Wang, H. (2022). A practical Yang, Y., Niehaus, K., Walker, T., Iqbal, Z., Walker, S., Wilson, D., et al. (2018).
approach for predicting antimicrobial phenotype resistance in Staphylococcus aureus Machine learning for classifying tuberculosis drug-resistance from DNA sequencing
through machine learning analysis of genome data. Front. Microbiol. 13:841289. doi: data. Bioinformatics 34, 1666–1671. doi: 10.1093/bioinformatics/btx801
10.3389/fmicb.2022.841289
Yasir, M., Karim, A. M., Malik, S. K., Bajaffer, A. A., Azhar, E. I., Yasir, M., et al.
Weis, C., Jutzeler, C., and Borgwardt, K. (2020). Machine learning for microbial (2022). Application of decision-tree-based machine learning algorithms for
identification and antimicrobial susceptibility testing on MALDI-TOF mass spectra: a prediction of antimicrobial resistance. Antibiotics 11:1593:1593. doi: 10.3390/
systematic review. Clin. Microbiol. Infect. 26, 1310–1317. doi: 10.1016/j.cmi.2020.03.014 ANTIBIOTICS11111593
Wiegand, I., Hilpert, K., and Protocols, R. H.-N.2008, undefined (2008). Agar and Yu, Y., Li, Y., Zhang, Z., Gu, Z., Zhong, H., Zha, Q., et al. (2020). A bibliometric
broth dilution methods to determine the minimal inhibitory concentration (MIC) of analysis using VOSviewer of publications on COVID-19. Ann. Transl. Med. 8:816. doi:
antimicrobial substances. Nat. Protoc. 3, 163–175. doi: 10.1038/nprot.2007.521 10.21037/atm-20-4235

Frontiers in Microbiology 14 frontiersin.org