Reimagining leprosy elimination with AI analysis of a combination of skin lesion images with demographic and clinical data

Barbieri, Raquel Rodrigues; Xu, Yixi; Setian, Lucy; Souza-Santos, Paulo Thiago de; Trivedi, Anusua; Cristofono, Jim; Bhering, Ricardo; White, Kevin; Sales, Anna Maria; Miller, Geralyn; Nery, José A. C.; Sharman, Michael; Bumann, Richard; Zhang, Shun; Goldust, Mohamad; Sarno, Euzenir Nunes; Mirza, Fareed; Cavaliero, Arielle; Timmer, Sander W.; Bonfiglioli, Elena; Smith, Cairns; Scollard, David M.; Navarini, Alexander; Aerts, Ann; Ferres, Juan Lavista; Moraes, Milton Ozório

doi:10.1016/j.lana.2022.100192

Cited by 15 publications

(25 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, we believe if there were enough images, stratifying may increase the accuracy of the predicted diagnosis. Moreover, as prior study on AIbased diagnosis for leprosy showed, clinical data other than images, most importantly loss of sensation for leprosy, are essential to be combined in the deep learning dataset for better model development (17). The percentages of correct diagnosis increased with PCR-positive cases of Buruli ulcer.…”

Section: Discussionmentioning

confidence: 99%

Deep learning for AI-based diagnosis of skin-related neglected tropical diseases: a pilot study

Yotsu

Ding

Hamm

et al. 2023

Preprint

View full text Add to dashboard Cite

Background Deep learning, which is a part of a broader concept of artificial intelligence (AI) and/or machine learning has achieved remarkable success in vision tasks. While there is growing interest in the use of this technology in diagnostic support for skin-related neglected tropical diseases (skin NTDs), there have been limited studies in this area and fewer focused on dark skin. In this study, we aimed to develop deep learning based AI models with clinical images we collected for five skin NTDs, namely, Buruli ulcer, leprosy, mycetoma, scabies, and yaws, to understand how diagnostic accuracy can or cannot be improved using different models and training patterns. Methodology This study used photographs collected prospectively in Côte d'Ivoire and Ghana through our ongoing studies with use of digital health tools for clinical data documentation and for teledermatology. Our dataset included a total of 1,709 images from 506 patients. Two convolutional neural networks, ResNet-50 and VGG-16 models were adopted to examine the performance of different deep learning architectures and validate their feasibility in diagnosis of the targeted skin NTDs. Principal findings The two models were able to correctly predict over 70% of the diagnoses, and there was a consistent performance improvement with more training samples. The ResNet-50 model performed better than the VGG-16 model. Model trained with PCR confirmed cases of Buruli ulcer yielded 1-3% increase in prediction accuracy over training sets including unconfirmed cases. Conclusions Our approach was to have the deep learning model distinguish between multiple pathologies simultaneously – which is close to real-world practice. The more images used for training, the more accurate the diagnosis became. The percentages of correct diagnosis increased with PCR-positive cases of Buruli ulcer. This demonstrated that it may be better to input images from the more accurately diagnosed cases in the training models also for achieving better accuracy in the generated AI models. However, the increase was marginal which may be an indication that the accuracy of clinical diagnosis alone is reliable to an extent for Buruli ulcer. Diagnostic tests also have its flaws, and they are not always reliable. One hope for AI is that it will objectively resolve this gap between diagnostic tests and clinical diagnoses with addition of another tool. While there are still challenges to be overcome, there is a potential for AI to address the unmet needs where access to medical care is limited, like for those affected by skin NTDs.

show abstract

Section: Discussionmentioning

confidence: 99%

Deep learning for AI-based diagnosis of skin-related neglected tropical diseases: a pilot study

Yotsu

Ding

Hamm

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Machine learning algorithms can also analyze patient data for diagnosis and risk prediction. These technologies, when validated and available in mobile apps, could significantly improve timely HD diagnosis (11,(13)(14)(15).…”

Section: Potential Role Of Ai In Hd Programsmentioning

confidence: 99%

The potential role of artificial intelligence in the clinical management of Hansen’s disease (leprosy)

Deps,

Yotsu,

Furriel

et al. 2024

Front. Med.

View full text Add to dashboard Cite

Missed and delayed diagnoses of Hansen’s disease (HD) are making the battle against it even more complex, increasing its transmission and significantly impacting those affected and their families. This strains public health systems and raises the risk of lifelong impairments and disabilities. Worryingly, the three countries most affected by HD witnessed a growth in new cases in 2022, jeopardizing the World Health Organization’s targets to interrupt transmission. Artificial intelligence (AI) can help address these challenges by offering the potential for rapid case detection, customized treatment, and solutions for accessibility challenges—especially in regions with a shortage of trained healthcare professionals. This perspective article explores how AI can significantly impact the clinical management of HD, focusing on therapeutic strategies. AI can help classify cases, ensure multidrug therapy compliance, monitor geographical treatment coverage, and detect adverse drug reactions and antimicrobial resistance. In addition, AI can assist in the early detection of nerve damage, which aids in disability prevention and planning 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.

show abstract

“…These technologies, if validated and available in mobile apps, could significantly improve timely HD diagnosis. [18][19][20][21] https://doi.org/10.33774/coe-2023-45b3m ORCID: https://orcid.org/0000-0002-1635-2225 Content not peer-reviewed by Cambridge University Press. License: CC BY-NC-SA 4.0…”

Section: The Role Of Ai In Leprosy Diagnosismentioning

confidence: 99%

“…For example, AI algorithms analyze medical images in real time. 19,21,34 AIdriven devices continuously monitor vital signs and health data, transmitted to healthcare professionals for real-time tracking and adjustments to treatment plans (Figure 2). 22,26,35 Importantly, AI continually learns and adapts as more patient data becomes available, refining recommendations over time.…”

Section: Addressing Accessibility Challengesmentioning

confidence: 99%

The Potential Role of AI in Addressing Challenges in the Clinical Management of Hansen's Disease

Deps,

Lima,

Loureiro

2023

Preprint

View full text Add to dashboard Cite

Missed and delayed diagnoses exacerbate the burden of Hansen's disease (HD), also known as leprosy, contributing to the ongoing transmission of the causative agents and having a significant impact on the lives of affected individuals and their families. These delays also strain public health systems due to the heightened risk of lifelong impairments and disabilities. A concerning factor is that, in 2022, the top three HD-affected countries saw an increase in new cases compared to the previous year, threatening WHO targets of interrupting transmission. Artificial intelligence (AI) can serve as a strong ally for endemic countries, aiding in the rapid detection of new HD cases, personalizing HD treatment, and addressing accessibility issues, especially in regions facing a shortage of trained healthcare professionals. This article reviews the main published studies and discusses the potential role of AI in the clinical management of HD.

show abstract

Reimagining leprosy elimination with AI analysis of a combination of skin lesion images with demographic and clinical data

Cited by 15 publications

References 9 publications

Deep learning for AI-based diagnosis of skin-related neglected tropical diseases: a pilot study

Deep learning for AI-based diagnosis of skin-related neglected tropical diseases: a pilot study

The potential role of artificial intelligence in the clinical management of Hansen’s disease (leprosy)

The Potential Role of AI in Addressing Challenges in the Clinical Management of Hansen's Disease

Contact Info

Product

Resources

About