Revista Médica Vozandes
Volumen 34, Número 1, 2023
Articial intelligence (AI) has shown signicant promise as a prognostic tool
in various gastrointestinal tract pathologies, including colorectal cancer,
esophageal disorders, inammatory bowel disease (IBD), liver diseases,
and pancreatic disorders. AI algorithms analyze patient data to provide
insights into disease progression, treatment response, and prognosis. In
gastroenterology, AI has excelled in upper gastrointestinal endoscopy,
surpassing human performance in detecting esophageal and gastric
cancer. It offers benets like cancer screening and automated report
generation. In IBD, AI predicts disease are-ups, assesses therapy response,
and tailors treatments for individual patients. For liver diseases, AI identies
subtle radiographic features, monitors brosis progression, and evaluates
treatment response patterns. In pancreatic disorders, AI models predict
outcomes, optimize surgeries, and enable targeted therapies. Integrating
AI as a prognostic tool brings advantages like processing vast data
quickly, enhancing diagnostic accuracy, and aiding in risk stratication
and treatment planning. Challenges include ethical considerations and
the need for validation through larger studies. Overall, AI has the potential
to revolutionize managing gastrointestinal tract pathologies, improving
patient outcomes and quality of life.
A inteligência articial (IA) tem mostrado um grande potencial como
ferramenta prognóstica em diversas patologias do trato gastrointestinal,
incluindo câncer colorretal, distúrbios esofágicos, doenças inamatórias
intestinais (DII), doenças hepáticas e distúrbios pancreáticos. Algoritmos de
IA analisam dados do paciente para fornecer insights sobre a progressão
da doença, resposta ao tratamento e prognóstico. Na gastroenterologia,
a IA tem se destacado na endoscopia do trato gastrointestinal superior,
superando o desempenho humano na detecção de câncer esofágico
e gástrico. Ela oferece benefícios como triagem de câncer e geração
automatizada de relatórios. Nas DII, a IA prevê crises da doença, avalia a
resposta à terapia e personaliza tratamentos para pacientes individuais.
Para doenças hepáticas, a IA identica características radiográcas
sutis, monitora a progressão da brose e avalia padrões de resposta
ao tratamento. Em distúrbios pancreáticos, modelos de IA preveem
resultados, otimizam cirurgias e permitem terapias direcionadas. A
integração da IA como ferramenta prognóstica traz vantagens como
processamento rápido de grandes volumes de dados, aumento da
precisão diagnóstica e auxílio na estraticação de riscos e planejamento
de tratamento. Desaos incluem considerações éticas e a necessidade
de validação por meio de estudos maiores. No geral, a IA tem o potencial
de revolucionar o manejo de patologias do trato gastrointestinal,
melhorando os resultados e a qualidade de vida dos pacientes.
1 University of South Wales, Program of Acute Medici-
ne and Gastroenterology, United Kingdom.
2 Postgraduate Program in Medicine: Pathology, School
of Medicine, Universidade Federal de Ciências da
Saúde de Porto Alegre, Brasil
Soldera Jonathan:
*Corresponding author: Soldera Jonathan
Soldera Jonathan 1,2*
Este artículo está bajo una
licencia de Creative Com-
mons de tipo Reconocimien-
to – No comercial – Sin obras
derivadas 4.0 International.
Soldera J. Artificial Intelligence as a
Prognostic Tool for Gastrointestinal
Tract Pathologies. Med Vozandes. 2023;
34 (1): 9 - 14
Received: 07 – Apr – 2023
Accepted: 04 – Jun – 2023
Publish: 01 – Jul – 2023
Article history
Palavras-chave: Inteligência articial; Doenças
gastrointestinais; Prognóstico; Endoscopia; Tratamento
Keywords: Articial intelligence; Gastrointesti-
nal diseases; Prognosis; Endoscopy; Personalized
DOI: 10.48018/rmv.v34.i1.e
Soldera J
Revista Médica Vozandes
Volumen 34, Número 1, 2023
for gastrointestinal tract pathologies, such as
inammatory bowel disease, colorectal cancer, and
gastroesophageal reux disease. By analyzing extensive
patient data, including medical records, imaging
studies, and laboratory results, AI algorithms can provide
valuable insights into disease progression, treatment
response, and overall prognosis [4].
The application of AI in upper gastrointestinal
endoscopy has gained signicant attention in
recent years. AI systems have been developed to
assist in the assessment of (pre-)cancerous lesions
of the gastrointestinal tract. These systems have
shown promising results in several areas, including
the detection, characterization, and delineation of
esophageal and gastric cancer, prediction of tumor
invasion, and detection of Helicobacter pylori [5]. AI
algorithms have demonstrated high accuracy rates of
up to 99% in detecting supercial and advanced upper
GI cancers, surpassing the performance of trainee
and experienced endoscopists [5]. Additionally, AI has
outperformed mid-level and trainee endoscopists,
although not expert endoscopists, in the detection
of esophageal lesions and atrophic gastritis [5]. The
integration of AI in upper gastrointestinal endoscopy holds
the potential to improve early diagnosis of esophageal
and gastric cancer and enhance the identication of
patients suitable for endoscopic resection [5].
The eld of gastrointestinal endoscopy has witnessed
signicant advancements in the application of AI. AI-
assisted endoscopy has the potential to revolutionize
the practice of gastroenterologists, offering benets
such as cancer screening and automated report
generation [6]. Over the years, AI has made remarkable
progress in endoscopy, with the development of
sophisticated models and potential applications that
encompass various aspects of modern endoscopic
practice [6]. These AI models have the ability to enhance
diagnostic accuracy, risk stratication, and pathologic
identication, paving the way for improved patient
care in gastroenterology [6]. However, it is important to
acknowledge the limitations of AI and understand its
role as a tool to enhance human decision-making rather
than replace healthcare professionals [6].
Colorectal Cancer
Colorectal cancer (CRC) is a signicant contributor to
morbidity and mortality on a global scale. The application
of AI algorithms has demonstrated remarkable success
in analyzing medical images, including colonoscopy
and histopathological slides. These AI systems assist
in the early detection of precancerous lesions and
accurate staging of tumors, enabling clinicians to make
informed decisions regarding treatment strategies,
personalized therapies, and prognostic predictions.
The advancements in AI technology, such as machine
learning and deep learning, have revolutionized
the eld of medicine, offering promising prospects
for various applications, including medical image
The advent of Articial Intelligence (AI) has brought
about remarkable transformations across various
industries, including the eld of medicine. AI has the
potential to revolutionize healthcare in gastroenterology
by serving as a prognostic tool for gastrointestinal tract
pathologies. The history of AI in medicine dates back to
1950, with early models facing limitations that hindered
their widespread acceptance and application in
medicine [1]. However, advancements in deep learning
algorithms in the early 2000s overcame many of these
limitations, enabling AI systems to analyze complex
algorithms and self-learn [1]. This marked a new era in
medicine where AI can be applied to clinical practice,
improving diagnostic accuracy, workow efciency,
and risk assessment models [1].
The evolution of AI in medicine has been signicant in
recent years, with major applications in gastroenterology
and endoscopy [1]. AI-powered medical technologies
are rapidly evolving and becoming applicable solutions
for clinical practice, leveraging deep learning algorithms
to handle large amounts of data from wearables,
smartphones, and other mobile monitoring sensors [2].
The application of AI in specic clinical settings, such
as the detection of atrial brillation, epilepsy seizures,
and hypoglycemia, as well as the diagnosis of diseases
through histopathological examination or medical
imaging, has shown promising results [2]. However, the
implementation of augmented medicine and the
integration of AI in clinical practice face challenges,
including resistance from physicians and the need for
validation through traditional clinical trials [2]. Despite
these challenges, the benets of AI applications in
clinical practice are becoming increasingly apparent,
leading to discussions about its future opportunities
and risks on physicians, healthcare institutions, medical
education, and bioethics [2].
As laboratory medicine continues to evolve from a mainly
manual profession to a highly automated discipline,
generating vast amounts of diagnostic data, AI algorithms
are poised to play a crucial role in structuring and making
sense of this data, merging the worlds of the laboratory
and clinics [3]. By providing valuable suggestions in
diagnosis, prognosis, and therapeutic options, AI can
enhance the efciency and accuracy of medical
decision-making, freeing physicians’ time to focus more
on patient care [3]. However, the implementation of AI
algorithms in healthcare also poses challenges and
requires careful consideration of the ethical implications
and potential impact on healthcare workers [3].
This editorial delves into the role of AI in prognostication
for these conditions, highlighting its benets, challenges,
and future prospects.
AI in Gastrointestinal Tract Pathologies
In the domain of gastroenterology, AI algorithms
have shown great potential as prognostic tools
Revista Médica Vozandes
Volumen 34, Número 1, 2023
recognition, biotechnology, auxiliary diagnosis, drug research
and development, and nutrition [7-9].
CRC is a common gastrointestinal malignancy that poses a
serious threat to human health, with high mortality rates. Many
CRC cases arise from the malignant transformation of colorectal
polyps. Therefore, early diagnosis and treatment play a crucial
role in determining the prognosis of CRC. Diagnostic methods
for CRC include imaging diagnosis, endoscopy, and pathology
diagnosis, while treatment options encompass endoscopic
treatment, surgical treatment, and drug treatment [7].
AI technology, although currently in its nascent stage,
predominantly focuses on image recognition and auxiliary
analysis, lacking in-depth communication capabilities with
patients. Nonetheless, its potential to revolutionize CRC
diagnosis, treatment, and prognosis is signicant. By leveraging
AI, clinicians can identify high-risk patients, devise precise and
personalized treatment plans, and predict prognoses with
greater accuracy [8].
The integration of AI technology in the diagnosis, treatment,
and prognosis of CRC holds immense promise for improving
patient outcomes. Further research and development in this
eld are needed to harness the full potential of AI in CRC
management [7-9]. One practical application of AI in CRC
prognostication is the development of predictive models
that utilize machine learning algorithms to analyze clinical
and molecular data. These models can help clinicians assess
the likelihood of disease progression, recurrence, and patient
survival based on individual patient characteristics and tumor
features. By integrating diverse data sources and considering
multiple variables, AI-driven prognostic models have the
potential to enhance treatment decision-making and improve
patient outcomes in CRC. [8]
Esophageal Disorders
Esophageal disorders encompass a wide range of conditions,
including gastroesophageal reux disease (GERD) and
esophageal cancer. The integration of articial intelligence
(AI) models has shown signicant potential in predicting
disease progression, identifying high-risk individuals, and
guiding therapeutic interventions for patients with esophageal
disorders. These AI models analyze clinical data, such as patient
history, endoscopic ndings, and imaging studies, to optimize
treatment plans and improve long-term outcomes [10-13].
AI has shown promise in the prognostication of esophageal
disorders by utilizing advanced predictive modeling techniques.
By analyzing a combination of clinical data, including patient
demographics, medical history, and diagnostic test results, AI
algorithms can generate personalized prognostic assessments.
These assessments provide valuable insights into disease
progression, treatment response, and long-term outcomes,
enabling clinicians to make informed decisions and optimize
patient care strategies [10-13].
Several studies have demonstrated the effectiveness of AI
techniques, particularly deep learning and convolutional
neural networks, in analyzing endoscopic images and videos
for the early detection and characterization
of esophageal neoplasia [10,13]. Furthermore,
AI-based approaches have the potential to
facilitate pathological diagnosis, gene diagnosis,
and risk stratication models, although further
clinical research is needed for validation [12,13].
The use of AI-powered decision support systems
empowers clinicians to enhance their treatment
strategies and achieve better results for patients
with esophageal disorders [11].
Inammatory Bowel Disease
Inammatory Bowel Disease (IBD), encompassing
Crohn’s disease and ulcerative colitis, is
characterized by chronic inammation of the
gastrointestinal tract. The management of IBD
can be challenging due to its multifactorial
nature, involving factors such as host genetics, the
immune system, environmental inuences, and
the gut microbiome. To address these challenges,
AI-based algorithms have been developed
to integrate multiple data sources, including
patient-reported outcomes, genetic proles, and
biomarkers. These algorithms have shown promise
in predicting disease are-ups, assessing therapy
response, and tailoring treatment strategies for
individual patients [14,15].
The application of AI in IBD research has been
facilitated by technological advancements
such as next-generation sequencing, high-
throughput omics data generation, and
molecular networks [15]. Machine learning and
systems biology, which are subsets of AI, enable
the efcient integration and interpretation of
large datasets, leading to the discovery of
clinically relevant knowledge. For example,
machine learning approaches can facilitate
patient stratication, prediction of disease
progression, and therapy responses, ultimately
allowing for personalized treatment options that
improve patient outcomes and reduce costs [15,16].
Studies have demonstrated the potential of AI
in various aspects of IBD care. For instance, AI
has been applied in the diagnosis, follow-up,
treatment selection, and prognosis of IBD. It has
also been utilized in the analysis of histopathology,
endoscopy, and imaging, highlighting its role
in improving disease assessment [17,18]. AI-driven
predictive models have shown promise in
identifying features and patterns across diverse
datasets, leading to insights that can enhance
disease management. These models have the
potential to transform clinical practice by providing
personalized metrics for disease outcomes [19].
Despite the potential benets, there are
challenges associated with the implementation
of AI in clinical IBD research. Technical barriers,
Soldera J
Revista Médica Vozandes
Volumen 34, Número 1, 2023
targeted therapies[25-28]. For instance, AI can be utilized
in endoscopic ultrasound (EUS) for pancreatic disorders,
where it aids in computer-aided diagnosis (CAD) by
extracting and selecting features from imaging data
and utilizing deep learning-based algorithms[25]. By
leveraging AI’s predictive capabilities, clinicians can
intervene at an earlier stage, potentially improving
survival rates and patient outcomes[25-28]. Additionally, AI
can contribute to early detection of pancreatic cancer
through the use of medical images, pathological
examination, biomarkers, and other aspects, leading to
the early screening of high-risk groups and lesions[28]. It
can also predict prognosis, recurrence risk, metastasis,
and therapy response, inuencing patient outcomes[28].
Furthermore, AI nds applications in pancreatic cancer
health records, estimating medical imaging parameters,
and developing computer-aided diagnosis systems[28].
The advancement of AI applications in pancreatic
disorders requires collaboration among clinicians,
researchers, and engineers, as well as continuous efforts
to overcome limitations and harness the power of
computing in the ght against pancreatic cancer[27,28].
By incorporating AI into the management of pancreatic
disorders, healthcare professionals can potentially
improve the detection, treatment, and overall prognosis
of patients.
Benets of AI in Prognostication
The integration of AI as a prognostic tool for
gastrointestinal tract pathologies provides several
signicant advantages. AI enables the processing of
vast amounts of data in a shorter timeframe, leading
to timely and accurate prognostic information for
clinicians. AI algorithms have the capability to detect
subtle patterns and associations that may go unnoticed
by humans, thereby enhancing diagnostic accuracy
and prognostic precision. This can aid in risk stratication,
treatment planning, and optimizing patient outcomes [29].
Recent advancements in computing technology and
the application of AI in various elds, including medical
practice, have increased the potential for improved
outcomes [30]. AI algorithms can process complex
mathematical data, allowing for the consideration
of multiple parameters and sophisticated formulas
to determine conclusions that would be impractical
or impossible for humans alone [30]. This individualized
approach could lead to more tailored treatments for
each patient. However, it is important to note that most
studies conducted so far are retrospective and further
evaluation through large-scale prospective studies is
needed [30]. In addition to medical challenges, the
implementation of AI on a large scale raises ethical and
nancial considerations [30].
AI, particularly deep learning, has the potential to
enhance GI endoscopy in various areas, including
lesion detection, classication, quality metrics, and
documentation [31]. Computer vision in endoscopy
powered by advanced machine learning algorithms
can improve prediction and treatment outcomes for
bias within datasets, and the need for validation in
larger cohorts are among the obstacles that need to be
addressed. Ethical considerations related to the use of
AI in healthcare and data interpretation/validation are
also important aspects to be taken into account [18,20].
AI-based approaches have the potential to revolutionize
the management of IBD by integrating diverse data
sources and providing personalized insights for patients.
By leveraging machine learning and systems biology, AI
algorithms can enhance disease assessment, prediction,
and treatment selection, ultimately improving patient
outcomes and quality of life.
Liver Diseases
Liver diseases, including non-alcoholic fatty liver
disease (NAFLD) and viral hepatitis, have a signicant
impact on global healthcare systems. Detecting
and managing these diseases is crucial for patient
outcomes. AI algorithms trained on extensive datasets
can help identify subtle radiographic features, monitor
brosis progression, and assess treatment response
patterns. By leveraging machine learning models,
healthcare professionals can identify patients at risk,
intervene in a timely manner, and make accurate
prognostic assessments to guide clinical management
decisions, such as liver transplantation. For example,
machine learning models have been developed to
predict major adverse cardiovascular events after
orthotopic liver transplantation, providing valuable
insights for recipient selection and identifying
individuals at elevated risk for post-transplantation
MACE [21]. Another study utilized a machine learning
algorithm to predict 30-day and 365-day mortality
after liver transplantation, achieving high accuracy
in predicting patient outcomes [22]. Furthermore, these
tools have demonstrated effectiveness in challenging
scenarios, such as predicting post-LT major adverse
cardiac events [23]. Additionally, machine learning
algorithms have been applied to predict rebleeding
and mortality for oesophageal variceal bleeding in
cirrhotic patients, outperforming other assessment
tools like CLIF-SOFA and MELD score [24]. Such tool has
been also prospectively validated [25]. Deep learning
techniques have also been employed to improve
the imaging diagnosis of hepatocellular carcinoma,
aiding in the detection and diagnosis of this type of
liver cancer [26]. This demonstrates the potential of AI
and machine learning in liver disease management and
highlight the value of integrating these technologies into
clinical practice.
Pancreatic Disorders
Pancreatic disorders, such as pancreatitis and
pancreatic cancer, often manifest with late-stage
symptoms and poor prognoses. The integration of AI
models can signicantly impact the eld of pancreatic
disease by analyzing clinical records, genetic
information, and medical imaging to predict disease
outcomes, optimize surgical interventions, and enable
Revista Médica Vozandes
Volumen 34, Número 1, 2023
patients with GI disorders and cancer [31]. The availability of
large libraries of endoscopic images, such as “EndoNet,” can
facilitate the development and application of AI systems [31]. The
regulatory environment for AI implementation is evolving, with
colon polyp detection highlighted as a potential clinical trial
endpoint [31]. Ongoing collaboration among gastroenterologists,
industry experts, and regulatory agencies is crucial to ensure
rapid progress and meaningful clinical impact [31].
AI and machine learning also have potential applications in
the eld of liver disease, particularly in acute on chronic liver
failure (ACLF) [32]. AI methods, including predictive, prognostic,
probabilistic, and simulation modeling, can minimize cognitive
load and impact short-term and long-term patient outcomes [32].
However, ethical considerations and a lack of proven benets
temper the enthusiasm for AI implementation [32]. AI models
can contribute to the understanding of morbidity and mortality
mechanisms in ACLF [32]. The impact of AI on patient-centered
outcomes and other aspects of patient care is still uncertain [32].
In the eld of pancreatic cystic lesions (PCLs), radiomics
combined with machine learning and AI methods has the
potential to differentiate between benign and malignant
lesions, leading to improved clinical decision-making and
resource utilization [33]. Radiomics involves quantitative image
analysis to extract features and develop imaging biomarkers for
predicting high-risk PCLs [33]. However, further studies are needed
to validate the clinical application of radiomics [33].
AI and big data analysis have signicantly inuenced clinical
oncology and research [34]. Next-generation sequencing (NGS)
platforms, coupled with AI and machine learning, enable the
identication of novel biomarkers, therapeutic targets, and
accurate prognosis in cancer [34]. However, challenges and
limitations, such as data analysis and validation, remain [34].
While AI and ML promise transformative changes in healthcare,
there are provisos that need to be addressed. Reliability of
input data, interpretation of output data, data privacy, liability
issues, decreased human interaction, patient satisfaction,
affordability, and skepticism regarding cost-benet are
important considerations [34].
Challenges and Limitations
Despite the promising potential of AI in prognostication,
there are several challenges and limitations that need to be
addressed. The availability of high-quality data is crucial for
effectively training AI algorithms[35]. However, concerns related
to data privacy and ethical considerations surrounding the
use of patient information must be carefully addressed to
ensure patient condentiality and establish trust[35]. Improving
the interpretability and transparency of AI algorithms is also
necessary to enable clinicians to understand and validate the
reasoning behind the generated prognostic predictions[35].
In a study on AI in primary health care, experts reported
that AI has the potential to improve managerial and clinical
decisions and processes[35]. Common data standards would
facilitate the realization of this potential[35]. However, there
was no consensus among the experts regarding whether AI
applications should learn and adapt to clinician
preferences or behavior, and they did not agree
on the extent of AI’s potential harm to patients[35].
Assessing the impact of AI-based applications
on continuity and coordination of care was
found to be more challenging[35].
In the eld of surgery, AI has shown promise but
also comes with certain limitations[36]. Machine
learning, articial neural networks, natural
language processing, and computer vision
are the main subelds of AI with applications in
surgery[36]. These applications include big data
analytics and clinical decision support systems[36].
Surgeons have the opportunity to integrate
AI into modern practice, partnering with data
scientists to capture data and provide clinical
context[36]. This collaboration has the potential to
revolutionize surgery and improve patient care.
In drug discovery, AI has gained traction and
is increasingly being used to accelerate the
process[36,37]. AI techniques, such as machine
learning and deep learning, have been
applied to various aspects of drug discovery,
including quantitative structure-activity/
property relationship modeling, de novo
molecular design, and chemical synthesis
prediction[37]. These AI-driven approaches have
the potential to address some of the challenges
in CNS drug discovery[38]. They can aid in target
identication, compound screening, hit/lead
generation and optimization, drug response
and synergy prediction, de novo drug design,
and drug repurposing[38]. However, there are still
limitations and challenges to be overcome, such
as the need for large amounts of annotated
data and issues with interpretability[37,38].
In the eld of nuclear medicine, AI has the potential
to impact various aspects of the profession[39].
It can be applied to different stages of the
imaging workow, including planning, scanning,
interpretation, and reporting[39]. However, current
AI techniques have limitations, such as the need
for interpretability and the requirement for large
amounts of annotated data[39].
Future Prospects
The future of AI in prognostication for
gastrointestinal tract pathologies holds immense
potential. As AI algorithms continue to evolve and
improve, their integration with electronic health
records and imaging systems can enhance real-
time decision-making and facilitate personalized
treatment approaches[35,39]. Collaboration
between clinicians, data scientists, and AI experts
is essential to develop robust and reliable AI
models specically tailored to gastrointestinal
diseases [35,39]. For example, in the eld of peptic
Soldera J
Revista Médica Vozandes
Volumen 34, Número 1, 2023
Transparency in AI algorithms and effective
communication of their limitations to patients and
healthcare providers are essential[35]. Respecting patient
autonomy, privacy, and informed consent are crucial
during the development and implementation of AI-
based prognostic tools[35]. It is important to ensure that
AI complements clinical decision-making and does not
replace the human touch and personalized care that
patients deserve[35].
The integration of Articial Intelligence as a prognostic tool
for gastrointestinal tract pathologies holds tremendous
promise in revolutionizing disease management and
treatment strategies. AI has the potential to improve
diagnostic accuracy, enhance prognostic precision,
and ultimately optimize patient outcomes. However,
addressing challenges such as data quality, ethical
concerns, and algorithm transparency is vital for the
successful implementation of AI in routine clinical
practice. Through collaborative efforts and ongoing
research, AI can become an invaluable asset in the ght
against gastrointestinal diseases, equipping clinicians
with powerful tools for prognostication and patient-
centered care.
ulcers, AI has shown promising applications in pathogenic
factor identication, diagnosis, and management[35].
AI can aid in the identication of Helicobacter pylori
(Hp) infection, differential diagnosis, and management
of complications such as bleeding, obstruction,
perforation, and canceration[35]. These AI-based tools
have the potential to improve the management of
peptic ulcer patients[35]. Similarly, in colonoscopy, AI
has the potential to enhance the early identication,
resection, and treatment of precancerous adenoma
and early-stage cancer, leading to a reduction in
colorectal cancer prevalence and mortality[40]. AI
technologies, such as computer-aided detection and
diagnosis (CAD), can provide decision support during
colonoscopy, increasing the adenoma detection rate
and improving the effectiveness of screening[41]. The
future prospects of AI in colonoscopy include improving
diagnostic performance, reducing costs, optimizing
endoscopic schedules, and addressing limitations and
challenges[41]. AI also holds potential in the management
of inammatory bowel disease (IBD)[42]. AI applications in
IBD range from genomics to endoscopic applications,
enabling disease classication, stratication, self-
monitoring, and personalized management[42]. The
practical applications of AI in IBD are already being
used, and the future holds further potential for AI to
enhance patient care[42].
As AI becomes increasingly embedded in healthcare,
ethical considerations must remain at the forefront[35].
1. Kaul V, Enslin S, Gross SA. History of articial
intelligence in medicine. Gastrointest En-
dosc. 2020 Oct;92(4):807-812. doi: 10.1016/j.
gie.2020.06.040. PMID: 32565184.
2. Briganti G, Le Moine O. Articial Intelligence in
Medicine: Today and Tomorrow. Front Med
(Lausanne). 2020 Feb 5;7:27. doi: 10.3389/
fmed.2020.00027. PMID: 32118012; PMCID:
3. Cadamuro J. Rise of the Machines: The In-
evitable Evolution of Medicine and Medi-
cal Laboratories Intertwining with Articial
Intelligence-A Narrative Review. Diagnostics
(Basel). 2021 Aug 2;11(8):1399. doi: 10.3390/
diagnostics11081399. PMID: 34441333; PM-
CID: PMC8392825.
4. Kaul V, Enslin S, Gross SA. History of articial
intelligence in medicine. Gastrointest En-
dosc. 2020 Oct;92(4):807-812. doi: 10.1016/j.
gie.2020.06.040. PMID: 32565184.
5. Tokat M, van Tilburg L, Koch AD, Spaander
MCW. Articial Intelligence in Upper Gas-
trointestinal Endoscopy. Gastrointest En-
dosc. 2021;93(5):1204-1206. doi: 10.1016/j.
gie.2020.12.008. PMID: 34348267.
6. Abadir AP, Ali MF, Karnes W, Samarasena JB. Ar-
ticial Intelligence in Gastrointestinal Endos-
copy. Clin Endosc. 2020;53(2):132-141. doi:
10.5946/ce.2020.038. PMID: 32252506; PMCID:
7. Liang F, Wang S, Zhang K, Liu TJ, Li JN. Deve-
lopment of articial intelligence technology
in diagnosis, treatment, and prognosis of co-
lorectal cancer. World J Gastrointest Oncol.
2022 Jan 15;14(1):124-152. doi: 10.4251/wjgo.
8. Qiu H, Ding S, Liu J, Wang L, Wang X. Applications
of Articial Intelligence in Screening, Diagno-
sis, Treatment, and Prognosis of Colorectal
Cancer. Curr Oncol. 2022 Mar 7;29(3):1773-
1795. doi: 10.3390/curroncol29030146.
9. Wang Y, He X, Nie H, Zhou J, Cao P, Ou C. Appli-
cation of articial intelligence to the diag-
nosis and therapy of colorectal cancer. Am
J Cancer Res. 2020 Nov 1;10(11):3575-3598.
PMID: 33294256.
10. Huang LM, Yang WJ, Huang ZY, Tang CW, Li J.
Articial intelligence technique in detection
of early esophageal cancer. World J Gas-
troenterol. 2020 Oct 21;26(39):5959-5969. doi:
11. Hussein M, Everson M, Haidry R. Esopha-
geal squamous dysplasia and cancer: Is
articial intelligence our best weapon?
Best Pract Res Clin Gastroenterol. 2021
Jun-Aug;52-53:101723. doi: 10.1016/j.
12. Yang H, Hu B. Recent advances in early
esophageal cancer: diagnosis and
treatment based on endoscopy. Post-
grad Med. 2021 Aug;133(6):665-673. doi:
13. Ma H, Wang L, Chen Y, Tian L. Convolutional
neural network-based articial intelligen-
ce for the diagnosis of early esophageal
cancer based on endoscopic images: A
meta-analysis. Saudi J Gastroenterol. 2022
Sep-Oct;28(5):332-340. doi: 10.4103/sjg.
14. Sadat Seyed Tabib N, Madgwick M, Sudhakar
P, Verstockt B, Korcsmaros T, Vermeire S. Big
data in IBD: big progress for clinical practice.
Gut. 2020 Aug;69(8):1520-1532. doi: 10.1136/
gutjnl-2019-320065. PMID: 32111636
15. Da Rio L, Spadaccini M, Parigi TL, Gabbiadini
R, Dal Buono A, Busacca A, et al. Articial in-
telligence and inammatory bowel disease:
Where are we going? World J Gastroenterol.
2023 Jan 21;29(3):508-520. doi: 10.3748/wjg.
v29.i3.508. PMID: 36688019
16. Cohen-Mekelburg S, Berry S, Stidham RW, Zhu
J, Waljee AK. Clinical applications of articial
intelligence and machine learning-based
methods in inammatory bowel disease. J
Gastroenterol Hepatol. 2021 Feb;36(2):279-
285. doi: 10.1111/jgh.15405. PMID: 33624888
17. Javaid A, Shahab O, Adorno W, Fernandes
P, May E, Syed S. Machine Learning Predic-
tive Outcomes Modeling in Inammatory
Bowel Diseases. Inamm Bowel Dis. 2022 Jun
3;28(6):819-829. doi: 10.1093/ibd/izab187.
PMID: 34417815
18. Peng J, Jury EC, Dönnes P, Ciurtin C. Machi-
ne Learning Techniques for Personalised
Medicine Approaches in Immune-Mediated
Chronic Inammatory Diseases: Applications
and Challenges. Front Pharmacol. 2021 Sep
30;12:720694. doi: 10.3389/fphar.2021.720694.
PMID: 34658859
19. Iacucci M, Parigi TL, Del Amor R, Meseguer
P, Mandelli G, Bozzola A, et al. Articial Inte-
Revista Médica Vozandes
Volumen 34, Número 1, 2023
lligence Enabled Histological Prediction of
Remission or Activity and Clinical Outcomes
in Ulcerative Colitis. Gastroenterology. 2023
Jun;164(7):1180-1188.e2. doi: 10.1053/j.gas-
tro.2023.02.031. PMID: 36871598
20. Chen G, Shen J. Articial Intelligence Enhances
Studies on Inammatory Bowel Disease. Front
Bioeng Biotechnol. 2021 Jul 8;9:635764. doi:
10.3389/fbioe.2021.635764. PMID: 34307315
21. Jain V, Bansal A, Radakovich N, Sharma V,
Khan MZ, Harris K, et al. Machine Learning
Models to Predict Major Adverse Cardio-
vascular Events After Orthotopic Liver Trans-
plantation: A Cohort Study. J Cardiothorac
Vasc Anesth. 2021 Jul;35(7):2063-2069. doi:
10.1053/j.jvca.2021.02.006. PMID: 33750661.
22. Soldera J, Tomé F, Corso LL, Ballotin VR, Bigarella
LG, Balbinot RS, et al. Predicting 30 and 365-
Day Mortality After Liver Transplantation Using
a Machine Learning Algorithm. Gastroentero-
logy. 2021 May;160(6 Suppl):S-789-S-790. doi:
23. Soldera J, Tomé F, Corso LL, Machado Rech M,
Ferrazza AD, Terres AZ, et al. Use of a Machi-
ne Learning Algorithm to Predict Rebleeding
and Mortality for Oesophageal Variceal Blee-
ding in Cirrhotic Patients. EMJ Gastroenterol.
24. Ballotin VR, Bigarella LG, Soldera J, Soldera J.
Deep learning applied to the imaging diag-
nosis of hepatocellular carcinoma. Artif Intell
Gastrointest Endosc. 2021 Aug 28;2(4):127-
135. doi: 10.37126/aige.v2.i4.127.
25. Tonozuka R, Mukai S, Itoi T. The Role of Arti-
cial Intelligence in Endoscopic Ultrasound
for Pancreatic Disorders. Diagnostics (Basel).
2020 Dec 24;11(1):18. doi: 10.3390/diag-
nostics11010018. PMID: 33374181; PMCID:
26. Barat M, Chassagnon G, Dohan A, Gaujoux S,
Coriat R, Hoeffel C, et al. Articial intelligen-
ce: a critical review of current applications
in pancreatic imaging. Jpn J Radiol. 2021
Jun;39(6):514-523. doi: 10.1007/s11604-021-
01098-5. Epub 2021 Feb 6. PMID: 33550513.
27. Huang B, Huang H, Zhang S, Zhang D, Shi Q,
Liu J, et al. Articial intelligence in pancreatic
cancer. Theranostics. 2022 Oct 3;12(16):6931-
6954. doi: 10.7150/thno.77949. PMID:
36276650; PMCID: PMC9576619.
28. Kenner B, Chari ST, Kelsen D, Klimstra DS,
Pandol SJ, Rosenthal M, et al. Articial Inte-
lligence and Early Detection of Pancreatic
Cancer: 2020 Summative Review. Pancreas.
2021 Mar 1;50(3):251-279. doi: 10.1097/
MPA.0000000000001762. PMID: 33835956
29. Moldoveanu AC, Fierbinteanu-Braticevici C.
A Primer into the Current State of Articial
Intelligence in Gastroenterology. J Gastroin-
testin Liver Dis. 2022 Jun 12;31(2):244-253. doi:
10.15403/jgld-4180. PMID: 35694986.
30. Parasa S, Wallace M, Bagci U, Antonino M,
Berzin T, Byrne M, et al. Proceedings from the
First Global Articial Intelligence in Gastroen-
terology and Endoscopy Summit. Gastroin-
test Endosc. 2020 Oct;92(4):938-945.e1. doi:
10.1016/j.gie.2020.04.044. Epub 2020 Apr 25.
PMID: 32343978.
31. Gary PJ, Lal A, Simonetto DA, Gajic O, Ga-
llo de Moraes A. Acute on chronic liver
failure: prognostic models and articial
intelligence applications. Hepatol Com-
mun. 2023 Mar 24;7(4):e0095. doi: 10.1097/
HC9.0000000000000095. PMID: 36972378.
32. Dalal V, Carmicheal J, Dhaliwal A, Jain M,
Kaur S, Batra SK. Radiomics in stratication of
pancreatic cystic lesions: Machine learning in
action. Cancer Lett. 2020 Jan 28;469:228-237.
doi: 10.1016/j.canlet.2019.10.023. Epub 2019
Oct 17. PMID: 31629933.
33. Dlamini Z, Francies FZ, Hull R, Marima R. Arti-
cial intelligence (AI) and big data in can-
cer and precision oncology. Comput Struct
Biotechnol J. 2020 Aug 28;18:2300-2311. doi:
10.1016/j.csbj.2020.08.019. PMID: 32994889.
34. Bhardwaj A. Promise and Provisos of Arti-
cial Intelligence and Machine Learning in
Healthcare. J Healthc Leadersh. 2022 Jul
20;14:113-118. doi: 10.2147/JHL.S369498.
PMID: 35898671.
35. Liyanage H, Liaw ST, Jonnagaddala J, Schrei-
ber R, Kuziemsky C, Terry AL, et al. Articial
Intelligence in Primary Health Care: Percep-
tions, Issues, and Challenges. Yearb Med
Inform. 2019 Aug;28(1):41-46. doi: 10.1055/s-
0039-1677901. PMID: 31022751
36. Hashimoto DA, Rosman G, Rus D, Meireles OR.
Articial Intelligence in Surgery: Promises and
Perils. Ann Surg. 2018 Jul;268(1):70-76. doi:
37. Jiménez-Luna J, Grisoni F, Weskamp N, Schnei-
der G. Articial intelligence in drug discovery:
recent advances and future perspectives.
Expert Opin Drug Discov. 2021 Sep;16(9):949-
959. doi: 10.1080/17460441.2021.1909567.
38. Vatansever S, Schlessinger A, Wacker D, Kanis-
kan HÜ, Jin J, Zhou MM, et al. Articial inte-
lligence and machine learning-aided drug
discovery in central nervous system disea-
ses: State-of-the-arts and future directions.
Med Res Rev. 2021 May;41(3):1427-1473. doi:
10.1002/med.21764. PMID: 33295676
39. Nensa F, Demircioglu A, Rischpler C. Articial
Intelligence in Nuclear Medicine. J Nucl Med.
2019 Sep;60(Suppl 2):29S-37S. doi: 10.2967/
40. Zhao PY, Han K, Yao RQ, Ren C, Du XH. Appli-
cation Status and Prospects of Articial Inte-
lligence in Peptic Ulcers. Front Surg. 2022 Jun
16;9:894775. doi: 10.3389/fsurg.2022.894775.
41. Kamitani Y, Nonaka K, Isomoto H. Current Sta-
tus and Future Perspectives of Articial Intelli-
gence in Colonoscopy. J Clin Med. 2022 May
22;11(10):2923. doi: 10.3390/jcm11102923.
42. Brooks-Warburton J, Ashton J, Dhar A, Tham T,
Allen PB, Hoque S, et al. Articial intelligence
and inammatory bowel disease: practicali-
ties and future prospects. Frontline Gastroen-
terol. 2021 Dec 10;13(4):325-331. doi: 10.1136/
gastro-2021-102003. PMID: 35722596