TYPE Perspective

PUBLISHED 08 March 2024

DOI 10.3389/fmed.2024.1338598

The potential role of artificial

OPEN ACCESS intelligence in the clinical
management of Hansen’s disease
EDITED BY
Shisan (Bob) Bao,
The University of Sydney, Australia

REVIEWED BY
Maria Pena,
(leprosy)
Health Resources and Services
Administration, United States
Patrícia D. Deps 1,2*, Rie Yotsu 3, Brunna C. R. S. Furriel 2,
*CORRESPONDENCE
Patrícia D. Deps
Bruno D. de Oliveira 2, Sergio L. de Lima 2 and Rafael M. Loureiro 2
[email protected] 1
Department of Social Medicine, Health Sciences Center, Federal University of Espirito Santo, Vitoria,
RECEIVED 14November 2023 Brazil, 2 Department of Radiology, Hospital Israelita Albert Einstein, São Paulo, Brazil, 3 Department of
ACCEPTED 20 February 2024 Tropical Medicine, Tulane University School of Public Health and Tropical Medicine, New Orleans, LA,
PUBLISHED 08 March 2024 United States

CITATION
Deps PD, Yotsu R, Furriel BCRS,
de Oliveira BD, de Lima SL and
Missed and delayed diagnoses of Hansen’s disease (HD) are making the battle against
Loureiro RM (2024) The potential role of it even more complex, increasing its transmission and significantly impacting those
artificial intelligence in the clinical affected and their families. This strains public health systems and raises the risk of
management of Hansen’s disease (leprosy).
Front. Med. 11:1338598.
lifelong impairments and disabilities. Worryingly, the three countries most affected
doi: 10.3389/fmed.2024.1338598 by HD witnessed a growth in new cases in 2022, jeopardizing the World Health
COPYRIGHT
Organization’s targets to interrupt transmission. Artificial intelligence (AI) can
© 2024 Deps, Yotsu, Furriel, de Oliveira, de help address these challenges by offering the potential for rapid case detection,
Lima and Loureiro. This is an open-access customized treatment, and solutions for accessibility challenges—especially in
article distributed under the terms of the
Creative Commons Attribution License regions with a shortage of trained healthcare professionals. This perspective
(CC BY). The use, distribution or reproduction article explores how AI can significantly impact the clinical management of HD,
in other forums is permitted, provided the focusing on therapeutic strategies. AI can help classify cases, ensure multidrug
original author(s) and the copyright owner(s)
are credited and that the original publication therapy compliance, monitor geographical treatment coverage, and detect
in this journal is cited, in accordance with adverse drug reactions and antimicrobial resistance. In addition, AI can assist in the
accepted academic practice. No use, early detection of nerve damage, which aids in disability prevention and planning
distribution or reproduction is permitted
which does not comply with these terms. rehabilitation. Incorporating AI into mental health counseling is also a promising
contribution to combating the stigma associated with HD. By revolutionizing
therapeutic approaches, AI offers a holistic solution to reduce the burden of HD
and improve patient outcomes.

KEYWORDS

artificial intelligence, leprosy, Hansen’s disease, digital health, skin neglected tropical
diseases

1 Introduction
Hansen’s disease (HD), commonly known as leprosy, is a contagious and chronic infectious
disease. It experienced a resurgence in new cases after a reduction during the COVID-19
pandemic in 2020 (1, 2). Reflecting this shift, the number of countries reporting HD
information increased from 143 in 2021 to 182 in 2022. Globally, 174,087 new cases were
reported in 2022, representing a 23.8% increase compared to the 140,594 cases reported in
2021. The rise in newly detected cases in 16 of the 23 global priority countries in 2022 was
attributed to improved leprosy services following the pandemic. While the overall increase in
all priority countries was 25% compared to 2021, the growth varied from 3.4% in Micronesia
to 107.4% in Myanmar (3).

Frontiers in Medicine 01 frontiersin.org

Deps et al. 10.3389/fmed.2024.1338598

HD is caused by Mycobacterium leprae; in some regions, diagnosed based on clinical signs and medical history. However,
Mycobacterium lepromatosis is also involved (4). The incubation untrained healthcare professionals, limited access to medical facilities
period is long (average of 5 or more years). The bacillus is likely (especially in low- and middle-income countries), and the stigma
transmitted through droplets from the nose and mouth during associated with HD often prevent timely diagnosis and treatment.
prolonged close contact with untreated patients (5). Therefore, this delay represents an obstacle to the effective
HD is considered a clinical and immunological spectral disorder management of HD (6).
that primarily affects the skin and peripheral nerves. Nerve damage is
not uncommon and affects sensory and motor functions in areas such
as the eyes, hands, and feet. This condition can be diagnosed by 3.2 Adverse drug reactions due to
identifying well-established clinical indicators and symptoms, multidrug therapy (MDT)
including a range of skin lesions, and detecting peripheral nerve
damage, such as sensory loss and (palpable) thickened peripheral MDT is the standard treatment for HD and typically consists
nerves. Diagnosis confirmation can be achieved by detecting acid-fast of a combination of drugs, including dapsone, rifampicin, and
bacilli in slit skin smears (SSS) or by specific findings in samples clofazimine. While MDT is highly effective in curing HD, it can
obtained from skin and nerve biopsies. Moreover, laboratory also be associated with adverse drug reactions. An analysis of 194
techniques such as molecular methods for detecting the M. leprae- patient records revealed that 88 patients (45%) had side effects
specific repetitive element RLEP in SSS and skin biopsy, along with linked to MDT, leading to the discontinuation of the suspected
anti-phenolic glycolipid I (PGL-I) serology titers, play a supporting drug in 47 of 88 cases (24% of patients) (9). These adverse drug
role in clarifying cases, monitoring patients and household contacts, reactions vary in severity and occurrence among individuals, and
and assessing subclinical infections within the community (5, 6). healthcare providers monitor patients closely during MDT to
This article discusses how artificial intelligence (AI) can manage any adverse effects (6).
be instrumental in managing HD. AI’s applications include classifying MDT has been a cornerstone in the global effort to combat
cases, selecting treatment regimens, assessing therapy compliance, and HD and reduce its prevalence. However, emerging concerns
monitoring local and national treatment coverage. It can also assist in about drug resistance highlight the importance of continued
detecting adverse drug reactions and antimicrobial resistance. research and personalized treatment approaches to ensure the
Moreover, AI may enable the early detection of nerve damage, thereby continued effectiveness of HD treatment. Considering the
preventing disabilities and aiding in planning rehabilitation strategies. patient’s specific Mycobacterium leprae strain and their clinical
AI can also contribute to mental health management by counseling and genetic factors, this approach is critical in ensuring that HD
affected individuals and their family members, thus combating stigma remains a curable and manageable disease despite evolving drug
directly and indirectly. resistance challenges (5).

2 World health organization’s 3.3 Underdiagnosis

initiatives
There is a suspicion of underdiagnosis of HD cases in endemic
The World Health Organization has set a target of eliminating (or countries. The discrepancy between expected and observed cases
interrupting) transmission in 120 countries by 2030 (7). The specific amounted to around 2.6 million from 2000 to 2012 and was projected
target for HD in 2030 is defined as “an epidemiological state in a to exceed 4 million by 2020 (10). The COVID-19 pandemic caused a
leprosy-endemic country or area where there is no more local 40% drop in new HD cases in Brazil in 2020 (1). This highlights a
transmission of M. leprae,” demonstrated by zero new autochthonous substantial pool of undiagnosed individuals, contributing to ongoing
cases in children under 15 years of age for at least 5 years (8). Of the 10 transmission and burden for affected individuals and public
challenges that need to be addressed within this timeframe to achieve health systems.
this goal, two are related to delays in diagnosis and the lack of HD
specialists. In addition, this disease is strongly associated with
stigmatization. Discrimination against people affected by HD is an 4 Potential role of AI in HD programs
obstacle to achieving the elimination target. Therefore, combating
stigma to ensure that human rights are respected is one of the four The rapid progress of digital technologies offers great potential
strategic pillars to achieve the goals of the “Zero Leprosy Roadmaps” for transforming the approach to HD programs. At the
in all endemic countries (7). International Federation of Anti-Leprosy Associations (ILEP)
conference in 2020, various participants explored the application
of new digital technologies to enhance HD services. ILEP members
3 Challenges in the diagnosis and are adopting state-of-the-art technologies in mapping, diagnostics,
treatment of Hansen’s disease health information systems, and patient management (11). This
underscores the importance of staying updated on emerging
3.1 Late diagnosis technological trends to improve HD programs and ultimately
enhance the well-being of individuals affected by the disease.
Delayed (or missed) diagnosis of HD can lead to serious health Figure 1 illustrates the integration of digital technologies and AI
problems, including disability and social exclusion. HD is usually in HD management.

Frontiers in Medicine 02 frontiersin.org

Deps et al. 10.3389/fmed.2024.1338598

5 Role of AI in diagnosis of HD atypical clinical manifestations, asymptomatic individuals, or

subclinical cases of HD, including SSS, skin or nerve biopsy, RLEP
The comprehensive strategy outlines decisive steps to combat HD quantitative polymerase chain reaction (qPCR), and anti-PGL-I
worldwide, especially in regions with a shortage of specialized serology (5).
physicians. First and foremost, AI could be key in screening and
diagnosing new HD cases, thus bridging this “expertise gap.” This
includes implementing initiatives to identify cases and contacts in 6 Treatment personalization with AI
regions with elevated HD detection rates, often referred to as HD
hotspots, with prior education campaigns at community level. In HD manifests differently among patients, requiring tailored
addition, the evaluation of diagnostic tools such as antibody detection treatment approaches for better outcomes and fewer complications.
and molecular techniques is particularly valuable in identifying AI-driven personalized HD treatment starts with the comprehensive
multibacillary cases, which account for approximately 69% of new HD collection and combination of various patient data. This includes
cases globally (12). medical history, genetic information, clinical evaluations, past
Machine learning, a branch of AI, enables computers to learn from treatments, and socioeconomic background. The remarkable ability
data and make decisions without explicit instructions. It is widely used of AI to analyze this abundance of information offers a comprehensive
in areas such as image recognition and speech processing. In healthcare, view of the patient’s situation, leading to more precise treatment
machine learning models can be trained to recognize skin lesions by decisions (16).
analyzing images to identify disease patterns. Machine learning AI can contribute significantly to personalized HD treatment
algorithms can also analyze patient data for diagnosis and risk through two key aspects: risk stratification and prediction of
prediction. These technologies, when validated and available in mobile individual response to therapy. By analyzing the collected data, AI
apps, could significantly improve timely HD diagnosis (11, 13–15). algorithms can classify patients into risk groups based on factors such
AI algorithms to diagnose HD will initially classify people as as HD type, bacterial load, and comorbidities. This risk stratification
suspected cases. However, a “human” doctor will make the final allows healthcare providers to allocate resources more effectively and
confirmation. AI can be a valuable tool for early detection and prioritize high-risk patients who require closer monitoring and
identification, helping medical experts make more accurate and efficient specialized care. In addition, based on a large dataset of treatment
diagnoses. Finally, the availability of algorithms capable of diagnosing outcomes, AI can identify patterns and factors that influence the
HD also signifies that an increasing number of new cases can be detected. effectiveness of specific treatments for individual patients. Equipped
AI effectively recognizes the typical clinical presentations of HD with this predictive capability, healthcare providers can make
(13–15), which experienced dermatologists or well-trained physicians informed decisions about drug selection and dosing, minimizing the
can readily identify. However, laboratory tests can be invaluable for risk of adverse effects, drug resistance, and treatment failure (17, 18).

FIGURE 1
Integration of digital technologies and artificial intelligence (AI) in Hansen’s disease (HD) management. The application of digital technologies involves
digital mapping, accurate diagnosis through imaging, efficient data management, and communication among healthcare providers. AI predicts
treatment outcomes, enhances cure rates, and ensures treatment adherence. It also detects drug resistance and adverse reactions early, supports early
diagnosis, offers mental health counseling, and tailors patient-specific treatment plans. AI-integrated mobile apps provide educational resources,
improving remote patient care and disease management. ADR, adverse drug reactions.

Frontiers in Medicine 03 frontiersin.org

Deps et al. 10.3389/fmed.2024.1338598

The role of AI can go beyond the prescription phase and actively administrative tasks, appointment scheduling, medical reminders,
participate in the patient’s treatment journey. AI-powered applications and language translation, improving communication between
can provide real-time reminders and support and ensure that patients patients and doctors (20). Additionally, AI can be used in patient
adhere to their treatment plans. This technology improves not only triage to prioritize critical cases for immediate assessment.
treatment outcomes but also the patient experience by providing Moreover, AI collects and analyzes data for medical research,
personalized guidance and encouragement. clinical trials, and telemedicine best practices (16).
This synergy of AI and telemedicine has great potential to make
healthcare accessible, efficient, and effective, regardless of
7 AI revolutionizing stigma reduction geographical restrictions.
and early diagnosis
Stigma often acts as a significant barrier to individuals seeking 9 Ethical considerations and data
timely medical care (19). privacy
AI-powered chatbots and symptom-checking applications offer
individuals a discreet way to assess their symptoms and provide The integration of AI into HD diagnosis and treatment offers
benefits such as anonymous assessment, education, and mental health immense potential but requires ethical diligence. The sensitive nature
support. These tools allow people to seek information without fear of of patient data requires reliable data protection and security. Patients
social stigma or judgment. They ensure confidentiality and must be able to trust that their data is protected from unauthorized
personalization of information for greater engagement and address access. Therefore, healthcare providers and AI developers have a
concerns in a non-judgmental and empathetic way (20). responsibility to implement robust safeguards to ensure data
AI-driven chatbots and educational platforms can be employed to privacy (29).
develop and disseminate educational materials and campaigns, dispel Informed consent is fundamental to ethical medical practice,
myths, and reduce fears. Mental health chatbots help those facing ensuring patients comprehensively understand the role of AI in their
emotional issues due to stigma by offering coping strategies, care, encompassing its benefits, limitations, and potential risks.
connecting people with support groups, or recommending Patients also have the legitimate right to comprehend how AI systems
professional counseling when needed (21). Also, these chatbots can arrive at diagnosis and treatment recommendations. Ethical AI
offer AI-driven translation services to bridge language barriers and operates with respect for patients’ autonomy, recognizing their
make accurate information accessible to individuals regardless of authority in shaping their treatment plans. Developers are responsible
their language. for prioritizing transparency and offering clear explanations and
AI tools can monitor social media and online platforms for justifications for AI-driven decisions. While AI provides valuable
discussions related to HD. Sentiment analysis can identify negative or recommendations, the ultimate authority in healthcare decisions rests
stigmatizing discussions and enable organizations to intervene with with the patient. AI should function as a tool that empowers patients
accurate information (22). AI algorithms can also analyze public and amplifies their choices rather than attempting to undermine
health data to predict areas with high stigma or low early diagnosis them (30).
rates. Health authorities can then conduct targeted awareness Another critical ethical dimension relates to bias and fairness. AI
campaigns and allocate resources more effectively (23). algorithms, especially machine learning models, learn from historical
By providing personal, confidential, and accessible support data. If the training data used to develop these models is biased, the
and information, AI empowers people to overcome their fears AI system may inherit and perpetuate this bias. In the context of HD,
and misconceptions, ultimately improving health outcomes and AI systems can also be influenced by biases originating from patient
reducing stigma. demographics, access to healthcare, or previous diagnostic decisions.
Biased AI can exacerbate inequalities in healthcare by potentially
misdiagnosing or undertreating certain groups. Several strategies are
8 Addressing accessibility challenges needed to reduce bias in AI diagnosis, such as diversifying training
data, deploying debiasing techniques, and conducting regular audits
AI and telemedicine enter into a symbiotic relationship. AI can to ensure fairness and accuracy (16, 30).
improve telemedicine through advanced diagnostics, real-time Ethical principles demand that AI not replace medical
monitoring, and personalized care. It automates administrative tasks, professionals but support them. It is crucial not only for patient safety
ensures data accuracy, and bridges language barriers, particularly but also for ensuring medical ethics that physicians retain control over
useful in regions with limited access to healthcare (20, 23–26). AI-driven decisions. Integrating AI into healthcare raises complex
In teleconsultations, AI can support healthcare providers with accountability issues, mainly when errors and adverse outcomes
cutting-edge diagnoses. For example, AI algorithms analyze occur. To address these concerns and protect patients’ rights and trust
medical images in real time (13–15). AI-driven devices in AI-enabled healthcare, it is essential to establish clear responsibilities
continuously monitor vital signs and health data, which are and liability issues between healthcare providers, AI developers, and
transmitted to healthcare professionals to track progress in real healthcare institutions (29).
time and adjust treatment plans (Figure 2) (24, 27, 28). Importantly, Emphasizing transparency, fairness, patient autonomy, and
AI is constantly learning and adapting as more patient data continuous improvement ensures the ethical use of AI technologies
becomes available, refining recommendations over time. Chatbots that benefit patients while respecting their rights and dignity
and virtual assistants are streamlining telemedicine by taking over throughout their healthcare journey.

Frontiers in Medicine 04 frontiersin.org

Deps et al. 10.3389/fmed.2024.1338598

FIGURE 2
Artificial intelligence (AI)-enhanced telemedicine for remote monitoring of Hansen’s disease (HD). It features an AI-powered mobile app that tracks
symptoms and provides diagnostic guidance, enhancing patient engagement and self-care. The e-health gateway securely integrates patient data into
healthcare networks via the internet, ensuring data privacy. Healthcare professionals use telemedicine platforms to access and manage this data,
supported by a secure cloud database for real-time analysis. This integrated system ensures personalized and timely medical services, improving care
continuity and health outcomes for HD patients, even in remote areas.

10 Necessary steps to build a While resource allocation for AI initially seemed simple, its
trustworthy AI-based approach for HD integration into current medical settings is complex. Financial
services issues are less of a concern now, as many healthcare facilities can
afford AI technology. However, two critical factors are proving to
Specific guidelines for validating AI algorithms are needed for the be the most significant obstacles: the availability of high-quality
diverse use of AI in healthcare. This ensures the rigorous testing and data and the adaptability of existing processes to AI models.
reliability of AI algorithms in HD management. To create a Questions arise about the seamless integration of data into AI
trustworthy and reliable AI-based approach to HD programs and models. Can data flow effectively into the models? Is there an
services, several key steps must be followed. These steps include data automatic mechanism for merging data from different sources? It
collection, pre-processing, feature selection, algorithm development, is also important to consider the location and usability of the AI
training, and validation of AI models, all guided by ethical models. Even advanced AI solutions might be underutilized in
considerations and collaboration with healthcare experts. The ultimate practice without usability, underscoring the need to address these
goal is to develop AI systems compliant with regulatory requirements questions thoroughly.
and enable more effective disease management and improved support
for people affected by HD (29).
Data availability statement
11 Conclusion The original contributions presented in the study are included in
the article/supplementary material, further inquiries can be directed
The future of AI in the management of neglected tropical diseases, to the corresponding author/s.
particularly HD, holds immense promise for equitable and accessible
healthcare. Pioneering initiatives have already made significant
progress in the application of AI in public health, providing insight Author contributions
into its transformative potential in the diagnosis and treatment of
HD. Collaboration between healthcare organizations, technology PD: Conceptualization, Data curation, Investigation, Project
companies, and research institutions will be critical to achieving administration, Writing – original draft, Writing – review & editing.
this potential. RY: Writing – review & editing. BF: Data curation, Investigation,

Frontiers in Medicine 05 frontiersin.org

Deps et al. 10.3389/fmed.2024.1338598

Writing – review & editing. BO: Data curation, Investigation, Writing as a preprint, accessible at https://doi.org/10.33774/
– review & editing. SL: Writing – review & editing. RL: Writing – coe-2023-45b3m.
review & editing, Conceptualization, Data curation, Investigation,
Writing – original draft.
Conflict of interest
Funding PD, BF, BO, SL, and RL are currently involved in a project financed
by the Program of Support for the Institutional Development of the
The author(s) declare financial support was received for the Unified Health System (PROADI-SUS) in collaboration with Hospital
research, authorship, and/or publication of this article. This Israelita Albert Einstein.
project was supported by the Program of Support for the The remaining author declares that the research was conducted in
Institutional Development of the Unified Health System the absence of any commercial or financial relationships that could be
(PROADI-SUS, 01/2020; NUP: 25000.161106/2020-61) and construed as a potential conflict of interest.
Hospital Israelita Albert Einstein.

Publisher’s note
Acknowledgments
All claims expressed in this article are solely those of the authors and
The authors would like to acknowledge the use of do not necessarily represent those of their affiliated organizations, or those
InstaText (https://instatext.io/) for the review and editing of the of the publisher, the editors and the reviewers. Any product that may be
English language in this manuscript. Additionally, a preliminary evaluated in this article, or claim that may be made by its manufacturer,
version of this manuscript has been previously published online is not guaranteed or endorsed by the publisher.

References
1. Deps P, Collin SM, de Andrade VLG. Hansen’s disease case detection in Brazil: a 14. Yotsu RR, Ding Z, Hamm J, Blanton RE. Deep learning for AI-based diagnosis of
backlog of undiagnosed cases due to COVID-19 pandemic. J Eur Acad Dermatol skin-related neglected tropical diseases: a pilot study. PLoS Negl Trop Dis. (2023)
Venereol. (2022) 36:e754–5. doi: 10.1111/jdv.18307 17:e0011230. doi: 10.1371/journal.pntd.0011230
2. World Health Organization, Organisation mondiale de la Santé. Relevé 15. Beesetty R, Reddy SA, Modali S. et al, Leprosy skin lesion detection: an AI
Épidémiologique Hebd. Wkly Epidemiol Rec. (2021) 96:421–44. approach using few shot learning in a small clinical dataset. Indian J Lepr. (2023)
95:89–102.
3. Weekly Epidemiological Record. (2023), [EN/FR] -World: ReliefWeb. Available at:
https://reliefweb.int/report/world/weekly-epidemiological-record-wer-8-september-2023- 16. Pai VV, Pai RB. Artificial intelligence in dermatology and healthcare: an overview.
vol-98-no-37-pp-409-430-enfr Indian J Dermatol Venereol Leprol. (2021) 87:457–67. doi: 10.25259/IJDVL_518_19
4. Deps P, Collin SM. Mycobacterium lepromatosis as a second agent of Hansen’s 17. de Andrade Rodrigues RS, Heise EFJ, Hartmann LF, Rocha GE, Olandoski M, de
disease. Front Microbiol. (2021) 12:698588. doi: 10.3389/fmicb.2021.698588 Araújo Stefani MM, et al. Prediction of the occurrence of leprosy reactions based on
Bayesian networks. Front Med. (2023) 10:1233220. doi: 10.3389/fmed.2023.1233220
5. White C, Franco-Paredes C. Leprosy in the 21st century. Clin Microbiol Rev. (2015)
28:80–94. doi: 10.1128/CMR.00079-13 18. Portelli S, Myung Y, Furnham N, Vedithi SC, Pires DEV, Ascher DB. Prediction of
rifampicin resistance beyond the RRDR using structure-based machine learning
6. Alemu Belachew W, Naafs B. Position statement: LEPROSY: diagnosis, treatment
approaches. Sci Rep. (2020) 10:18120. doi: 10.1038/s41598-020-74648-y
and follow-up. J Eur Acad Dermatol Venereol. (2019) 33:1205–13. doi: 10.1111/jdv.15569
19. Deps P, Delboni L, TIA O, Collin SM, Andrade MA, ELN M. Steps towards
7. Control of Neglected Tropical Diseases, SEARO Regional Office for the South East
eliminating Hansen’s disease stigma. Int Health. (2023) 15:iii7–9. doi: 10.1093/inthealth/
Asia (RGO) (2023). Towards zero leprosy: Global leprosy (Hansen’s Disease) strategy
ihad050
2021–2030. Available at: https://www.who.int/publications-detail-
redirect/9789290228509 20. Xu L, Sanders L, Li K, Chow JCL. Chatbot for health care and oncology
applications using artificial intelligence and machine learning: systematic review. JMIR
8. World Organization of Health. (2023). Control of Neglected Tropical Diseases,
Cancer. (2021) 7:e27850. doi: 10.2196/27850
SEARO Regional Office for the South East Asia (RGO). Interruption of
Transmission and Elimination of Leprosy Disease. World Health Organization. 21. van der Schyff EL, Ridout B, Amon KL, Forsyth R, Campbell AJ. Providing self-led
Regional Office for South-East Asia. Available at: https://www.who.int/ mental health support through an artificial intelligence-powered chat bot (Leora) to
publications-detail-redirect/9789290210467 meet the demand of mental health care. J Med Internet Res. (2023) 25:e46448. doi:
10.2196/46448
9. Deps PD, Nasser S, Guerra P, Simon M, Birshner RDC, Rodrigues LC. Adverse
effects from multi-drug therapy in leprosy: a Brazilian study. Lepr Rev. (2007) 78:216–22. 22. Babu NV, Kanaga EGM. Sentiment analysis in social media data for depression
doi: 10.47276/lr.78.3.216 detection using artificial intelligence: a review. SN Comput Sci. (2022) 3:74. doi: 10.1007/
s42979-021-00958-1
10. Smith WC, van Brakel W, Gillis T, Saunderson P, Richardus JH. The missing
millions: a threat to the elimination of leprosy. PLoS Negl Trop Dis. (2015) 9:e0003658. 23. De Souza MLM, Lopes GA, Branco AC, Fairley JK, Fraga LADO. Leprosy
doi: 10.1371/journal.pntd.0003658 screening based on artificial intelligence: development of a cross-platform app. JMIR
Mhealth Uhealth. (2021) 9:e23718. doi: 10.2196/23718
11. Warne G, Mukhier M. Application of digital technologies to leprosy programmes.
Lepr Rev. (2021) 92:182–5. doi: 10.47276/lr.92.2.182 24. Latif S, Qadir J, Farooq S, Imran M. How 5G wireless (and concomitant
technologies) will revolutionize healthcare? Future Internet. (2017) 9:93. doi: 10.3390/
12. Yang J, Li X, Sun Y, Zhang L, Jin G, Li G, et al. Global epidemiology of leprosy
fi9040093
from 2010 to 2020: a systematic review and meta-analysis of the proportion of sex,
type, grade 2 deformity and age. Pathog Glob Health. 116:467–76. doi: 25. Dahoklory DF, Haryanto J, Indarwati R. The application of digital health as a
10.1080/20477724.2022.2057722 nursing solution for leprosy patients during the COVID-19 pandemic: a systematic
review. JPMA J Pak Med Assoc. (2023) 73:S170. doi: 10.47391/JPMA.Ind-S2-38
13. Barbieri RR, Xu Y, Setian L, Souza-Santos PT, Trivedi A, Cristofono J, et al.
Reimagining leprosy elimination with AI analysis of a combination of skin lesion images 26. Saif-Ur-Rahman KM, Islam MS, Alaboson J, Ola O, Hasan I, Islam N, et al. Artificial
with demographic and clinical data. Lancet Reg Health Am. (2022) 9:100192. doi: intelligence and digital health in improving primary health care service delivery in LMICs: a
10.1016/j.lana.2022.100192 systematic review. J Evid-Based Med. (2023) 16:303–20. doi: 10.1111/jebm.12547

Frontiers in Medicine 06 frontiersin.org

Deps et al. 10.3389/fmed.2024.1338598

27. Dananjayan S, Raj GM. 5G in healthcare: how fast will be the transformation? Ir J 29. Geis JR, Brady AP, Wu CC, Spencer J, Ranschaert E, Jaremko JL, et al. Ethics of artificial
Med Sci. (2021) 190:497–501. doi: 10.1007/s11845-020-02329-w intelligence in radiology: summary of the joint European and north American multisociety
statement. J Am Coll Radiol. (2019) 16:1516–21. doi: 10.1016/j.jacr.2019.07.028
28. Devi DH, Duraisamy K, Armghan A, Alsharari M, Aliqab K, Sorathiya V, et al. 5G
Technology in healthcare and wearable devices: a review. Sensors. (2023) 23:2519. doi: 30. Keskinbora KH. Medical ethics considerations on artificial intelligence. J Clin
10.3390/s23052519 Neurosci. (2019) 64:277–82. doi: 10.1016/j.jocn.2019.03.001

Frontiers in Medicine 07 frontiersin.org