Amoxicillin/clavulanate in combination with rifampicin/clarithromycin is bactericidal against Mycobacterium ulcerans.

BACKGROUND: Buruli ulcer (BU) is a skin neglected tropical disease (NTD) caused by Mycobacterium ulcerans. WHO-recommended treatment requires 8-weeks of daily rifampicin (RIF) and clarithromycin (CLA) with wound care. Treatment compliance may be challenging due to socioeconomic determinants. Previous minimum Inhibitory Concentration and checkerboard assays showed that amoxicillin/clavulanate (AMX/CLV) combined with RIF+CLA were synergistic against M. ulcerans. However, in vitro time kill assays (TKA) are a better approach to understand the antimicrobial activity of a drug over time. Colony forming units (CFU) enumeration is the in vitro reference method to measure bacterial load, although this is a time-consuming method due to the slow growth of M. ulcerans. The aim of this study was to assess the in vitro activity of RIF, CLA and AMX/CLV combinations against M. ulcerans clinical isolates by TKA, while comparing four methodologies: CFU enumeration, luminescence by relative light unit (RLU) and optical density (at 600 nm) measurements, and 16S rRNA/IS2404 genes quantification. METHODOLOGY/PRINCIPAL FINDINGS: TKA of RIF, CLA and AMX/CLV alone and in combination were performed against different M. ulcerans clinical isolates. Bacterial loads were quantified with different methodologies after 1, 3, 7, 10, 14, 21 and 28 days of treatment. RIF+AMX/CLV and the triple RIF+CLA+AMX/CLV combinations were bactericidal and more effective in vitro than the currently used RIF+CLA combination to treat BU. All methodologies except IS2404 quantitative PCR provided similar results with a good correlation with CFU enumeration. Measuring luminescence (RLU) was the most cost-effective methodology to quantify M. ulcerans bacterial loads in in vitro TKA. CONCLUSIONS/SIGNIFICANCE: Our study suggests that alternative and faster TKA methodologies can be used in BU research instead of the cumbersome CFU quantification method. These results provide an in vitro microbiological support to of the BLMs4BU clinical trial (NCT05169554, PACTR202209521256638) to shorten BU treatment.

Emergence and spread of Mycobacterium ulcerans at different geographic scales.

The classical lineage of Mycobacterium ulcerans is the most prevalent clonal group associated with Buruli ulcer in humans. Its reservoir is strongly associated with the environment. We analyzed together 1,045 isolates collected from 13 countries on two continents to define the evolutionary history and population dynamics of this lineage. We confirm that this lineage spread over 7,000 years from Australia to Africa with the emergence of outbreaks in distinct waves in the 18th and 19th centuries. In sharp contrast with its global spread over the last century, transmission chains are now mostly local, with little or no dissemination between endemic areas. This study provides new insights into the phylogeography and population dynamics of M. ulcerans, highlighting the importance of comparative genomic analyses to improve our understanding of pathogen transmission. IMPORTANCE: Mycobacterium ulcerans is an environmental mycobacterial pathogen that can cause Buruli ulcer, a severe cutaneous infection, mostly spread in Africa and Australia. We conducted a large genomic study of M. ulcerans, combining genomic and evolutionary approaches to decipher its evolutionary history and pattern of spread at different geographic scales. At the scale of villages in an endemic area of Benin, the circulating genotypes have been introduced in recent decades and are not randomly distributed along the river. On a global scale, M. ulcerans has been spreading for much longer, resulting in distinct and compartmentalized endemic foci across Africa and Australia.

A lesion in two: Buruli ulcer and squamous cell carcinoma coexistence.

The concurrent diagnoses of Buruli ulcer (BU) and cutaneous squamous cell carcinoma (SCC) is a phenomenon not previously described, despite the fact that both conditions are highly prevalent in Australia. This report presents an intriguing case of concurrent diagnoses, with clues alluding to more than one skin condition being present. The case involves a 73-year-old man with BU diagnosed on the scalp, an atypical location, which led to the consideration of malignancy, ultimately revealing concurrent SCC. This case highlights the importance of considering both conditions in patients with epidemiological risk factors, necessitating multiple lines of investigation for accurate diagnosis. Medical practitioners must remain vigilant and incorporate this possibility into their diagnostic algorithms for suspicious skin lesions to optimize treatment and outcomes. This is the first recorded instance of simultaneous diagnosis, underlining the need for enhanced awareness and attention to these unique cases.

Unlocking Opportunities for Mycobacterium leprae and Mycobacterium ulcerans.

In the recent decade, scientific communities have toiled to tackle the emerging burden of drug-resistant tuberculosis (DR-TB) and rapidly growing opportunistic nontuberculous mycobacteria (NTM). Among these, two neglected mycobacteria species of the Acinetobacter family, Mycobacterium leprae and Mycobacterium ulcerans, are the etiological agents of leprosy and Buruli ulcer infections, respectively, and fall under the broad umbrella of neglected tropical diseases (NTDs). Unfortunately, lackluster drug discovery efforts have been made against these pathogenic bacteria in the recent decade, resulting in the discovery of only a few countable hits and majorly repurposing anti-TB drug candidates such as telacebec (Q203), P218, and TB47 for current therapeutic interventions. Major ignorance in drug candidate identification might aggravate the dramatic consequences of rapidly spreading mycobacterial NTDs in the coming days. Therefore, this Review focuses on an up-to-date account of drug discovery efforts targeting selected druggable targets from both bacilli, including the accompanying challenges that have been identified and are responsible for the slow drug discovery. Furthermore, a succinct discussion of the all-new possibilities that could be alternative solutions to mitigate the neglected mycobacterial NTD burden and subsequently accelerate the drug discovery effort is also included. We anticipate that the state-of-the-art strategies discussed here may attract major attention from the scientific community to navigate and expand the roadmap for the discovery of next-generation therapeutics against these NTDs.

Mosquitoes provide a transmission route between possums and humans for Buruli ulcer in southeastern Australia.

Buruli ulcer, a chronic subcutaneous infection caused by Mycobacterium ulcerans, is increasing in prevalence in southeastern Australia. Possums are a local wildlife reservoir for M. ulcerans and, although mosquitoes have been implicated in transmission, it remains unclear how humans acquire infection. We conducted extensive field survey analyses of M. ulcerans prevalence among mosquitoes in the Mornington Peninsula region of southeastern Australia. PCR screening of trapped mosquitoes revealed a significant association between M. ulcerans and Aedes notoscriptus. Spatial scanning statistics revealed overlap between clusters of M. ulcerans-positive Ae. notoscriptus, M. ulcerans-positive possum excreta and Buruli ulcer cases, and metabarcoding analyses showed individual mosquitoes had fed on humans and possums. Bacterial genomic analysis confirmed shared single-nucleotide-polymorphism profiles for M. ulcerans detected in mosquitoes, possum excreta and humans. These findings indicate Ae. notoscriptus probably transmit M. ulcerans in southeastern Australia and highlight mosquito control as a Buruli ulcer prevention measure.

Pseudogenomic insights into the evolution of Mycobacterium ulcerans.

BACKGROUND: Buruli ulcer (BU) disease, caused by Mycobacterium ulcerans (MU), and characterized by necrotic ulcers is still a health problem in Africa and Australia. The genome of the bacterium has several pseudogenes due to recent evolutionary events and environmental pressures. Pseudogenes are genetic elements regarded as nonessential in bacteria, however, they are less studied due to limited available tools to provide understanding of their evolution and roles in MU pathogenicity. RESULTS: This study developed a bioinformatic pipeline to profile the pseudogenomes of sequenced MU clinical isolates from different countries. One hundred and seventy-two MU genomes analyzed revealed that pseudogenomes of African strains corresponded to the two African lineages 1 and 2. Pseudogenomes were lineage and location specific and African lineage 1 was further divided into A and B. Lineage 2 had less relaxation in positive selection than lineage 1 which may signify different evolutionary points. Based on the Gil-Latorre model, African MU strains may be in the latter stages of evolutionary adaption and are adapting to an environment rich in metabolic resources with a lower temperature and decreased UV radiation. The environment fosters oxidative metabolism and MU may be less reliant on some secondary metabolites. In-house pseudogenomes from Ghana and Cote d'Ivoire were different from other African strains, however, they were identified as African strains. CONCLUSION: Our bioinformatic pipeline provides pseudogenomic insights to complement other whole genome analyses, providing a better view of the evolution of the genome of MU and suggest an adaptation model which is important in understanding transmission. MU pseudogene profiles vary based on lineage and country, and an apparent reduction in insertion sequences used for the detection of MU which may adversely affect the sensitivity of diagnosis.

SIGNIFICANCE: Prevention and treatment of Buruli ulcer is still a problem but large whole genome datasets on M. ulcerans are readily available. However, genomic studies fail to thoroughly investigate pseudogenes to probe evolutionary changes in the bacteria, and this can be attributed to the lack of bioinformatic tools. This work studied pseudogenes in Mycobacterium ulcerans (MU) to understand its adapted niche and evolutionary differences across African strains. Our results posit an MU niche-adapted model important in understanding transmission. Also, MU pseudogene profiles vary based on lineage and country, suggesting their influence on pseudogenization patterns in the genome. We further identify a reduction in insertion sequences that are used for the detection of the bacteria which may affect the sensitivity of diagnosis.

Mycobacterium ulcerans-Bordetella trematum chronic tropical cutaneous ulcer: A four-case series, Côte d'Ivoire.

BACKGROUND: Chronic tropical cutaneous ulcers remain a neglected medical condition in West Africa, particularly Buruli ulcer, which is caused by mycolactone cytotoxin-secreting Mycobacterium ulcerans (M. ulcerans). Medical management of this highly debilitating and necrotising skin infection may be modified by colonisation and co-infection of the ulcer by opportunistic and pathogenic microorganisms, which considerably delays and increases the cost of treatment. METHODOLOGY/PRINCIPAL FINDING: We diagnosed chronic tropical cutaneous ulcers in nine patients in Côte d'Ivoire using M. ulcerans-specific PCRs and culturomics. This revealed M. ulcerans in 7/9 ulcer swabs and 5/9 control swabs as well as an additional 122 bacterial species, 32 of which were specific to ulcers, 61 specifics to the controls, and 29 which were shared, adding 40 bacterial species to those previously reported. Whole genome sequencing of four Bordetella trematum (B. trematum) isolates in four Buruli ulcer swabs and no controls indicated cytolethal distending toxins, as confirmed by cytotoxic assay. CONCLUSIONS/SIGNIFICANCE: In four cases of Buruli ulcer in Côte d'Ivoire, B. trematum was a co-pathogen which was resistant to rifampicin and clarithromycin, unmatching M. ulcerans antibiotic susceptibility profile and counteracting the current treatment of Buruli ulcer in West Africa and Australia. Thus, we report here chronic mixed M. ulcerans-B. trematum chronic tropical ulcer as a specific form of Buruli ulcer in West Africa.

Repurposing of Tuberculosis Drug Candidates for the Treatment of Mycobacterium ulcerans Disease.

Buruli ulcer (BU) is a chronic necrotizing skin disease caused by Mycobacterium ulcerans. Historically, the disease was treated by surgical excision of the skin lesions, until an 8-week combination therapy of rifampicin and streptomycin was introduced in 2004. This treatment modality was effective and reduced recurrence rates. Rifampicin is the most efficacious antibiotic for the treatment of BU and, should rifampicin-resistant M. ulcerans strains emerge, there is currently no replacement for it. As for mycobacterial diseases in general, there is a pressing need for the development of novel, fast-acting drugs. Under market economy conditions, repurposing of new tuberculosis drug candidates is the most promising avenue for alternative BU treatments. Our drug repurposing activities have led to the identification of several actives against M. ulcerans. In particular, the cytochrome bc1 complex inhibitor telacebec (Q203) is a promising drug candidate for the treatment of BU in Africa and Australia. While an active cytochrome-bd oxidase bypass limits the potency of the cytochrome-bc1-specific inhibitor telacebec against M. tuberculosis, classical lineage M. ulcerans strains rely exclusively on cytochrome-bc1 to respire. Hence, telacebec is effective at nanomolar concentration against M. ulcerans, and a high treatment efficacy in an experimental mouse infection model indicates that treatment of BU could be substantially shortened and simplified by telacebec.

Key Contributions by the Swiss Tropical and Public Health Institute Towards New and Better Drugs for Tropical Diseases.

Thanks to its expertise in clinical research, epidemiology, infectious diseases, microbiology, parasitology, public health, translational research and tropical medicine, coupled with deeply rooted partnerships with institutions in low- and middle-income countries (LMICs), the Swiss Tropical and Public Health Institute (Swiss TPH) has been a key contributor in many drug research and development consortia involving academia, pharma and product development partnerships. Our know-how of the maintenance of parasites and their life-cycles in the laboratory, plus our strong ties to research centres and disease control programme managers in LMICs with access to field sites and laboratories, have enabled systems for drug efficacy testing in vitro and in vivo, clinical research, and modelling to support the experimental approaches. Thus, Swiss TPH has made fundamental contributions towards the development of new drugs - and the better use of old drugs - for neglected tropical diseases and infectious diseases of poverty, such as Buruli ulcer, Chagas disease, food-borne trematodiasis (e.g. clonorchiasis, fascioliasis and opisthorchiasis), human African trypanosomiasis, leishmaniasis, malaria, schistosomiasis, soil-transmitted helminthiasis and tuberculosis. In this article, we show case the success stories of molecules to which Swiss TPH has made a substantial contribution regarding their use as anti-infective compounds with the ultimate aim to improve people's health and well-being.

Inflammasome-triggered IL-18 controls skin inflammation in the progression of Buruli ulcer.

Buruli ulcer is an emerging chronic infectious skin disease caused by Mycobacterium ulcerans. Mycolactone, an exotoxin produced by the bacterium, is the only identified virulence factor so far, but the functions of this toxin and the mechanisms of disease progression remain unclear. By interfering Sec61 translocon, mycolactone inhibits the Sec61-dependent co-translational translocation of newly synthesized proteins, such as induced cytokines and immune cell receptors, into the endoplasmic reticulum. However, in regard to IL-1ß, which is secreted by a Sec61-independent mechanism, mycolactone has been shown to induce IL-1ß secretion via activation of inflammasomes. In this study, we clarified that cytokine induction, including that of IL-1ß, in infected macrophages was suppressed by mycolactone produced by M. ulcerans subsp. shinshuense, despite the activation of caspase-1 through the inflammasome activation triggered in a manner independent of mycolactone. Intriguingly, mycolactone suppressed the expression of proIL-1ß as well as TNF-α at the transcriptional level, suggesting that mycolactone of M. ulcerans subsp. shinshuense may exert additional inhibitory effect on proIL-1ß expression. Remarkably, constitutively produced IL-18 was cleaved and mature IL-18 was actually released from macrophages infected with the causative mycobacterium. IL-18-deficient mice infected subcutaneously with M. ulcerans exhibited exacerbated skin inflammation during the course of disease progression. On the other hand, IL-1ß controls bacterial multiplication in skin tissues. These results provide information regarding the mechanisms and functions of the induced cytokines in the pathology of Buruli ulcer.

Multi-epitope vaccine candidates based on mycobacterial membrane protein large (MmpL) proteins against Mycobacterium ulcerans.

Buruli ulcer (BU) is a neglected tropical disease. It is caused by the bacterium Mycobacterium ulcerans and is characterized by skin lesions. Several studies were performed testing the Bacillus Calmette-Guérin (BCG) vaccine in human and animal models and M. ulcerans-specific vaccines in animal models. However, there are currently no clinically accepted vaccines to prevent M. ulcerans infection. The aim of this study was to identify T-cell and B-cell epitopes from the mycobacterial membrane protein large (MmpL) proteins of M. ulcerans. These epitopes were analysed for properties including antigenicity, immunogenicity, non-allergenicity, non-toxicity, population coverage and the potential to induce cytokines. The final 8 CD8+, 12 CD4+ T-cell and 5 B-cell epitopes were antigenic, non-allergenic and non-toxic. The estimated global population coverage of the CD8+ and CD4+ epitopes was 97.71%. These epitopes were used to construct five multi-epitope vaccine constructs with different adjuvants and linker combinations. The constructs underwent further structural analyses and refinement. The constructs were then docked with Toll-like receptors. Three of the successfully docked complexes were structurally analysed. Two of the docked complexes successfully underwent molecular dynamics simulations (MDS) and post-MDS analysis. The complexes generated were found to be stable. However, experimental validation of the complexes is required.

Accelerating the healing of hard-to-heal wounds with food supplements: nutritional analysis in the Côte d'Ivoire.

OBJECTIVE: Hard-to-heal wounds are an important, yet often neglected, public health issue in low- and middle-income countries (LMICs). Malnutrition has been identified as a risk factor for prolonged healing times. However, nutritional supplements are not routinely provided for patients with hard-to-heal wounds, and so this study aimed to investigate their benefits. METHOD: This 9-month study was conducted in the Taabo Health and Demographic Surveillance System in the south-central part of Côte d'Ivoire. Patients with wounds (≥30mm2) were recruited. Treatment was standardised for inpatients (72%) and outpatients (28%). There were three intervention groups: supplemented with soy; orange flesh sweet potato (OFSP); or both. Another group was included without supplement, serving as control. General linear models were employed to assess the effects of log initial wound size, type of wound, food treatment group, haemoglobin, sex, age, place of treatment and body mass index on the rate of wound closure. RESULTS: The cohort consisted of 56 patients, 41 of whom were placed in intervention groups, and the remainder as controls. Within the cohort, 37 (66%) patients suffered from Buruli ulcer, 15 (27%) from traumatic wounds and four (7%) from erysipelas. We found a significant effect (p=0.004) of diet supplemented with OFSP on the wound healing rate. CONCLUSION: OFSP is a nutritional rehabilitation supplement, characterised by a high content of beta-carotene and carbohydrates. It is associated with shortened wound healing times, reduced discomfort and reduced cost of wound care. Further research should investigate the effect of a diet rich in beta-carotene, in combination with standard medical care, on hard-to-heal wound healing in LMICs.

Mycobacterium ulcerans not detected by PCR on human skin in Buruli ulcer endemic areas of south eastern Australia.

INTRODUCTION: Mycobacterium ulcerans (MU) causes Buruli ulcer (Buruli), a geographically restricted infection that can result in skin loss, contracture and permanent scarring. Lesion-location maps compiled from more than 640 cases in south eastern Australia suggest biting insects are likely involved in transmission, but it is unclear whether MU is brought by insects to humans or if MU is already on the skin and inoculation is an opportunistic event that need not be insect dependent. METHODS: We validated a PCR swab detection assay and defined its dynamic range using laboratory cultured M. ulcerans and fresh pigskin. We invited volunteers in Buruli-endemic and non-endemic areas to sample their skin surfaces with self-collected skin swabs tested by IS2404 quantitative PCR. RESULTS: Pigskin validation experiments established a limit-of-detection of 0.06 CFU/cm2 at a qPCR cycle threshold (Ct) of 35. Fifty-seven volunteers returned their self-collected kits of 4 swabs (bilateral ankles, calves, wrists, forearms), 10 from control areas and 47 from endemic areas. Collection was timed to coincide with the known peak-transmission period of Buruli. All swabs from human volunteers tested negative (Ct ≥35). CONCLUSIONS: M. ulcerans was not detected on the skin of humans from highly Buruli endemic areas.

Implementing patient-public engagement for improved health: Lessons from three Ghanaian community-based programmes.

BACKGROUND: Community-based health interventions have been implemented as a key strategy for achieving improved health outcomes in Ghana. Effectiveness, however, largely depends on the successful implementation of patient-public engagement (PPE). Although several PPE studies have been conducted in Ghana, little research has been done to understand the specific role of PPE in the context of implementing community-based health programmes. This paper, therefore, examines the extent of PPE implementation in three selected community-based health programmes (Community-based Health Planning and Service [CHPS], Community-based Maternal and Child Health and Buruli Ulcer) to understand their specific effects on health outcomes. METHODS: Three focus groups, involving 26 participants, were held in three districts of the Ashanti region of Ghana. Participants were mainly health service users involving community health committee members/volunteers, residents and health professionals. They were invited to participate based on their roles in the design and implementation of the programmes. Participants focused on each of Rifkin's spider-gram components. Data were transcribed and analysed descriptively using NVIVO 12 Plus. RESULTS: PPE implementation was found to be extensive across the three programmes in specific areas such as organisation and resource mobilisation. PPE was more restricted in relation to community needs assessment, leadership and management, particularly for the CHPS and Buruli Ulcer programmes. CONCLUSION: Findings suggest that benefits from community-based health interventions are likely to be greater if PPE can be widely implemented across all dimensions of the spider-gram framework.

Comprehensive Case-Control Study of Protective and Risk Factors for Buruli Ulcer, Southeastern Australia.

To examine protective and risk factors for Buruli ulcer (BU), we conducted a case-control study of 245 adult BU cases and 481 postcode-matched controls across BU-endemic areas of Victoria, Australia. We calculated age- and sex-adjusted odds ratios for socio-environmental, host, and behavioral factors associated with BU by using conditional logistic regression. Odds of BU were >2-fold for persons with diabetes mellitus and persons working outdoors who had soil contact in BU-endemic areas (compared with indoor work) but were lower among persons who had bacillus Calmette-Guérin vaccinations. BU was associated with increasing numbers of possums and with ponds and bore water use at residences. Using insect repellent, covering arms and legs outdoors, and immediately washing wounds were protective; undertaking multiple protective behaviors was associated with the lowest odds of BU. Skin hygiene/protection behaviors and previous bacillus Calmette-Guérin vaccination might provide protection against BU in BU-endemic areas.

Mycolactone A vs. B: Multiscale Simulations Reveal the Roles of Localization and Association in Isomer-Specific Toxicity.

Mycolactone is an exotoxin produced by Mycobacterium ulcerans that causes the neglected tropical skin disease Buruli ulcer. This toxin inhibits the Sec61 translocon in the endoplasmic reticulum (ER), preventing the host cell from producing several secretory and transmembrane proteins, resulting in cytotoxic and immunomodulatory effects. Interestingly, only one of the two dominant isoforms of mycolactone is cytotoxic. Here, we investigate the origin of this specificity by performing extensive molecular dynamics (MD) simulations with enhanced free energy sampling to query the association trends of the two isoforms with both the Sec61 translocon, using two distinct cryo-electron microscopy (cryo-EM) models as references, and the ER membrane, which serves as a toxin reservoir prior to association. Our results suggest that mycolactone B (the cytotoxic isoform) has a stronger association with the ER membrane than mycolactone A due to more favorable interactions with membrane lipids and water molecules. This could increase the reservoir of toxin proximal to the Sec61 translocon. In one model of Sec61 inhibited by mycolactone, we find that isomer B interacts more closely with residues thought to play a key role in signal peptide recognition and, thus, are essential for subsequent protein translocation. In the other model, we find that isomer B interacts more closely with the lumenal and lateral gates of the translocon, the dynamics of which are essential for protein translocation. These interactions induce a more closed conformation, which has been suggested to block signal peptide insertion and subsequent protein translocation. Collectively, these findings suggest that isomer B's unique cytotoxicity is a consequence of both increased localization to the ER membrane and channel-locking association with the Sec61 translocon, facets that could be targeted in the development of Buruli Ulcer diagnostics and Sec61-targeted therapeutics.

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

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. A model trained with PCR confirmed cases of Buruli ulcer yielded 1-3% increase in prediction accuracy across all diseases, except, for mycetoma, over a model which training sets included 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 their 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 the 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.