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

Background: Leprosy is an infectious disease that mostly affects underserved populations. Although it has been largely eliminated, still about 200'000 new patients are diagnosed annually. In the absence of a diagnostic test, clinical diagnosis is often delayed, potentially leading to irreversible neurological damage and its resulting stigma, as well as continued transmission. Accelerating diagnosis could significantly contribute to advancing global leprosy elimination. Digital and Artificial Intelligence (AI) driven technology has shown potential to augment health workers abilities in making faster and more accurate diagnosis, especially when using images such as in the fields of dermatology or ophthalmology. That made us start the quest for an AI-driven diagnosis assistant for leprosy, based on skin images. Methods: Here we describe the accuracy of an AI-enabled image-based diagnosis assistant for leprosy, called AI4Leprosy, based on a combination of skin images and clinical data, collected following a standardized process. In a Brazilian leprosy national referral center, 222 patients with leprosy or other dermatological conditions were included, and the 1229 collected skin images and 585 sets of metadata are stored in an open-source dataset for other researchers to exploit. Findings: We used this dataset to test whether a CNN-based AI algorithm could contribute to leprosy diagnosis and employed three AI models, testing images and metadata both independently and in combination. AI modeling indicated that the most important clinical signs are thermal sensitivity loss, nodules and papules, feet paresthesia, number of lesions and gender, but also scaling surface and pruritus that were negatively associated with leprosy. Using elastic-net logistic regression provided a high classification accuracy (90%) and an area under curve (AUC) of 96.46% for leprosy diagnosis. Interpretation: Future validation of these models is underway, gathering larger datasets from populations of different skin types and collecting images with smartphone cameras to mimic real world settings. We hope that the results of our research will lead to clinical solutions that help accelerate global leprosy elimination. Funding: This study was partially funded by Novartis Foundation and Microsoft (in-kind contribution).

Quantitative PCR for leprosy diagnosis and monitoring in household contacts: A follow-up study, 2011-2018.

Household contacts (HHC) of leprosy patients exhibit high-risk of developing leprosy and contact tracing is helpful for early diagnosis. From 2011 to 2018,2,437 HHC were examined in a clinic in Rio de Janeiro, Brazil and 16S qPCR was used for diagnosis and monitoring of contacts. Fifty-four HHCs were clinically diagnosed with leprosy at intake. Another 25 exhibited leprosy-like skin lesions at intake, 8 of which were confirmed as having leprosy (50% of which were qPCR positive) and 17 of which were diagnosed with other skin diseases (6% qPCR positive). In skin biopsies, qPCR presented a sensitivity of 0.50 and specificity of 0.94. Furthermore, 955 healthy HHCs were followed-up for at least 3 years and skin scrapings were collected from earlobes for qPCR detection. Positive qPCR indicated a non-significant relative risk of 2.52 of developing the disease. During follow-up, those who progressed towards leprosy exhibited 20% qPCR positivity, compared to 9% of those who remained healthy. Disease-free survival rates indicated that age had a significant impact on disease progression, where patients over 60 had a greater chance of developing leprosy [HR = 32.4 (3.6-290.3)]. Contact tracing combined with qPCR may assist in early diagnosis and age is a risk factor for leprosy progression.

Quantitative polymerase chain reaction in paucibacillary leprosy diagnosis: A follow-up study.

OBJECTIVE: The diagnosis of paucibacillary (PB) leprosy cases remains a challenge because of the absence of a confirmatory laboratory method. While quantitative polymerase chain reaction (qPCR) has been shown to provide reliable sensitivity and specificity in PB diagnoses, a thorough investigation of its efficacy in clinical practice has not yet been published. The present study evaluated patients with suspected leprosy skin lesions by using qPCR to identify PB individuals in the Leprosy Outpatient clinic at the Oswaldo Cruz Foundation in Rio de Janeiro, Brazil. METHODS: One hundred seventy-two suspected PB cases were included in the study. The patients were evaluated by a dermatologist at three different times. The clinical dermato-neurological examination and collected samples were performed on the first visit. On the second visit, the results of the histopathological analysis and PCR assay (DNA-based Mycobacterium leprae qPCR-targeting 16S gene) results were analyzed, and a decision regarding multi-drug therapy was made. A year later, the patients were re-examined, and the consensus diagnosis was established. RESULTS: In 58% (100/172) of cases, a conclusive diagnosis via histopathological analysis was not possible; however, 30% (30/100) of these cases had a positive PCR. One hundred ten patients (110/172) attended the third visit. The analysis showed that while the sensitivity of the histopathological test was very low (35%), a qPCR alone was more effective for identifying leprosy, with 57% sensitivity. CONCLUSION: The use of qPCR in suspected PB cases with an inconclusive histology improved the sensitivity of leprosy diagnoses.