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).

High nigral iron deposition in LRRK2 and Parkin mutation carriers using R2* relaxometry.

OBJECTIVES: The goal of this work was to investigate iron deposition in the basal ganglia and thalamus in symptomatic and asymptomatic leucine-rich repeat kinase 2 (LRRK2) and Parkin-associated Parkinson's disease (PD), using R2* relaxometry rate. METHODS: Twenty subjects with genetic PD (four symptomatic and two asymptomatic Parkin subjects, nine symptomatic and five asymptomatic LRRK2 subjects) were compared with 20 patients with idiopathic PD (IPD) and 20 healthy subjects. Images were obtained at 3 teslas, using multi-echo T2 and T2* sequences. R2 and R2* values were calculated in the substantia nigra (SN), the striatum, the globus pallidus, and the thalamus. RESULTS: The R2* values in the SN were increased in IPD and mutation-carrying patients as compared with controls and in mutation-carrying patients as compared with IPD. Asymptomatic mutation carriers showed higher R2* values than controls and did not differ from IPD patients. No changes were seen in the other structures or in R2 values. CONCLUSION: These results are consistent with increased iron load in LRRK2- and Parkin-mutation carriers. The increased R2* in asymptomatic PD-mutation carriers suggests that iron deposition occurs early during the preclinical phase of the disease. R2* measurements may be used as markers for investigating nigrostriatal damage in preclinical mutation-carrying patients.

Diagnostic accuracy of SeptiFast multi-pathogen real-time PCR in the setting of suspected healthcare-associated bloodstream infection.

PURPOSE: SeptiFast is a real-time PCR assay which targets ribosomal DNA sequences of bacteria and fungi, enabling detection and identification of the commonest pathogens in blood within a few hours, including those acquired in healthcare settings. We report here the first detailed assessment of SeptiFast that focuses on healthcare-associated bloodstream infections which develop during routine critical care. METHODS: This was a prospective multicentre study designed to compare the clinical diagnostic accuracy of SeptiFast versus microbiological culture and independent clinical adjudication. This Phase III diagnostic study was performed in an adequately sized cohort of adult patients who developed new signs of suspected bloodstream infection while receiving routine critical care. RESULTS: Of 1,006 new episodes of suspected bloodstream infection in 853 patients, 922 (92 %) of these episodes in 795 patients met the inclusion criteria of the study. Patients had been exposed to a median of 8 days (interquartile range 4-16) of hospital care and had received high levels of organ support and recent antibiotic exposure. The SeptiFast test, when compared with bloodstream infection at the species/genus level, had a greater specificity [0.86, 95 % confidence interval (CI) 0.83-0.88] than sensitivity (0.50, 95 % CI 0.39-0.61). There was a low prevalence of blood culture-proven pathogens (9.2 %, 95 % CI 7.4-11.2 %), and the post-test probabilities of both a positive (26.3 %, 95 % CI 19.8-33.7 %) and a negative SeptiFast test (5.6 %, 95 % CI 4.1-7.4 %) indicated potential limitations of this technology in diagnosing bloodstream infection. CONCLUSION: When compared with blood culture, SeptiFast is likely to have limited utility for the diagnosis of healthcare-associated bloodstream infection in critical care patients despite its potential to deliver results more rapidly.

Parkinson's disease patients show reduced cortical-subcortical sensorimotor connectivity.

Reduced dopamine input to cortical and subcortical brain structures, particularly those in the sensorimotor network, is a hallmark of Parkinson's disease (PD). The extent to which dopamine dysfunction affects connectivity within this and other brain networks remains to be investigated. The purpose of this study was to measure anatomical and functional connectivity in groups of PD patients and controls to determine whether connectivity deficits within the cortico-basal ganglia thalamocortical system could be attributed to PD, particularly in sensorimotor connections. A neuroimaging paradigm involving diffusion-weighted magnetic resonance imaging (MRI) and resting-state functional MRI was implemented in a large cohort of PD patients and control subjects. Probabilistic tractography and functional correlation analyses were performed to map connections between brain structures and to derive indices of connectivity that were then used to compare groups. Anatomical connectivity deficits were demonstrated in PD patients, specifically for sensorimotor connections. Functional deficits were also found in some of the same connections. In addition, functional connectivity was found to increase in associative and limbic connections in PD patients compared with controls. This study lends support to findings regarding the dysfunction of the sensorimotor circuit in PD. As deficits in anatomical and functional connectivity within this circuit were in some cases concordant in PD patients, a possible link between brain structure and function is suggested. Increases in functional connectivity in other cortico-basal ganglia thalamocortical circuits may be indicative of compensatory effects in response to system deficits elsewhere.

The structural correlates of functional deficits in early huntington's disease.

Neuropathological studies in Huntington disease (HD) have demonstrated neuronal loss in the striatum, as well as in other brain regions including the cortex. With diffusion tensor MRI we evaluated the hypothesis that the clinical dysfunction in HD is related to regionally specific lesions of circuit-specific cortico-basal ganglia networks rather than to the striatum only. We included 27 HD and 24 controls from the TRACK-HD Paris cohort. The following assessments were used: self-paced tapping tasks, trail B making test (TMT), University of Pennsylvania smell identification test (UPSIT), and apathy scores from the problem behaviors assessment. Group comparisons of fractional anisotropy and mean diffusivity and correlations were performed using voxel-based analysis. In the cortex, HD patients showed significant correlations between: (i) self paced tapping and mean diffusivity in the parietal lobe at 1.8 Hz and prefrontal areas at 3 Hz, (ii) UPSIT and mean diffusivity in the parietal, and median temporal lobes, the cingulum and the insula, and fractional anisotropy in the insula and the external capsule, (iii) TMT B and mean diffusivity in the white matter of the superior frontal, orbital, temporal, superior parietal and post central areas, and (iv) apathy and fractional anisotropy in the white matter of the rectus gyrus. In the basal ganglia, we found correlations between the self paced tapping, UPSIT, TMT tests, and mean diffusivity in the anterior part of the putamen and the caudate nucleus. In conclusion, disruption of motor, associative and limbic cortico-striatal circuits differentially contribute to the clinical signs of the disease.

Magnetic resonance imaging of the substantia nigra in Parkinson's disease.

Until recently, conventional magnetic resonance imaging (MRI) was most often negative in Parkinson's disease or showed nonspecific findings. Recent developments in structural MRI, including relaxometry, magnetization transfer, and neuromelanin imaging, have demonstrated improved contrast and enabled more accurate visualization of deep brain nuclei, in particular, the substantia nigra. Meanwhile, diffusion imaging has provided useful biomarkers of substantia nigra degeneration, showing reduced anisotropy and anatomical connectivity with the striatum and thalamus. These advances in structural imaging are complemented by findings of magnetic resonance spectroscopy on brain metabolism and resting-state functional MRI on functional connectivity. This article presents an overview of these new structural, metabolic, and resting-state functional MRI techniques and their implications for Parkinson's disease. The techniques are reviewed in the context of their potential for better understanding the disease in terms of diagnosis and pathophysiology and as biomarkers of its progression.

Brain networks in posterior cortical atrophy: a single case tractography study and literature review.

Posterior cortical atrophy (PCA) is rare neurodegenerative dementia, clinically characterized by a progressive decline in higher-visual object and space processing. After a brief review of the literature on the neuroimaging in PCA, here we present a study of the brain structural connectivity in a patient with PCA and progressive isolated visual and visuo-motor signs. Clinical and cognitive data were acquired in a 58-years-old patient (woman, right-handed, disease duration 18 months). Brain structural and diffusion tensor (DT) magnetic resonance imaging (MRI) were obtained. A voxel-based morphometry (VBM) study was performed to explore the pattern of gray matter (GM) atrophy, and a fully automatic segmentation was assessed to obtain the hippocampal volumes. DT MRI-based tractography was used to assess the integrity of long-range white matter (WM) pathways in the patient and in six sex- and age-matched healthy subjects. This PCA patient had a clinical syndrome characterized by left visual neglect, optic ataxia, and left limb apraxia, as well as mild visuo-spatial episodic memory impairment. VBM study showed bilateral posterior GM atrophy with right predominance; DT MRI tractography demonstrated WM damage to the right hemisphere only, including the superior and inferior longitudinal fasciculi and the inferior fronto-occipital fasciculus, as compared to age-matched controls. The homologous left-hemisphere tracts were spared. No difference was found between left and right hippocampal volumes. These data suggest that selective visuo-spatial deficits typical of PCA might not result from cortical damage alone, but by a right-lateralized network-level dysfunction including WM damage along the major visual pathways.

Impact of outliers on diffusion tensor and Q-ball imaging: clinical implications and correction strategies.

PURPOSE: To measure the impact of corrupted images often found to occur in diffusion-weighted magnetic resonance imaging (DW-MRI). To propose a robust method for the correction of outliers, applicable to diffusion tensor imaging (DTI) and q-ball imaging (QBI). MATERIALS AND METHODS: Monte Carlo simulations were carried out to measure the impact of outliers on DTI and QBI reconstruction in a single voxel. Methods to correct outliers based on q-space interpolation and direction removal were then implemented and validated in real image data. RESULTS: Corruption in a single voxel led to clear variations in DTI and QBI metrics. In real data, the method of q-space interpolation was successful in identifying corrupted voxels and restoring them to values consistent with those of uncorrupted images. CONCLUSION: For images containing few gradient directions, where outlier removal was either impossible due to limited volumes or resulted in large changes in DTI/QBI metrics, q-space interpolation proved to be the method of choice for image restoration. A simple decision support system is proposed to assist clinicians in the correction of their corrupted DW data.