Automatic Evaluation of Histological Prognostic Factors Using Two Consecutive Convolutional Neural Networks on Kidney Samples.

BACKGROUND AND OBJECTIVES: The prognosis of patients undergoing kidney tumor resection or kidney donation is linked to many histologic criteria. These criteria notably include glomerular density, glomerular volume, vascular luminal stenosis, and severity of interstitial fibrosis/tubular atrophy. Automated measurements through a deep-learning approach could save time and provide more precise data. This work aimed to develop a free tool to automatically obtain kidney histologic prognostic features. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: In total, 241 samples of healthy kidney tissue were split into three independent cohorts. The "Training" cohort (n=65) was used to train two convolutional neural networks: one to detect the cortex and a second to segment the kidney structures. The "Test" cohort (n=50) assessed their performance by comparing manually outlined regions of interest to predicted ones. The "Application" cohort (n=126) compared prognostic histologic data obtained manually or through the algorithm on the basis of the combination of the two convolutional neural networks. RESULTS: In the Test cohort, the networks isolated the cortex and segmented the elements of interest with good performances (>90% of the cortex, healthy tubules, glomeruli, and even globally sclerotic glomeruli were detected). In the Application cohort, the expected and predicted prognostic data were significantly correlated. The correlation coefficients r were 0.85 for glomerular volume, 0.51 for glomerular density, 0.75 for interstitial fibrosis, 0.71 for tubular atrophy, and 0.73 for vascular intimal thickness, respectively. The algorithm had a good ability to predict significant (>25%) tubular atrophy and interstitial fibrosis level (receiver operator characteristic curve with an area under the curve, 0.92 and 0.91, respectively) or a significant vascular luminal stenosis (>50%) (area under the curve, 0.85). CONCLUSION: This freely available tool enables the automated segmentation of kidney tissue to obtain prognostic histologic data in a fast, objective, reliable, and reproducible way.

Fusobacterium nucleatum bacteremia with pneumopathy in a patient receiving eculizumab: A case report.

Fusobacterium nucleatum is a common oral commensal bacterium capable of severe invasive infections. We report a case of a diffuse bilateral pneumopathy with F. nucleatum-positive blood culture successfully treated by common antibiotics in a patient receiving eculizumab for a drug-induced thrombotic microangiopathy (TMA). It is the first described case of a severe F. nucleatum-associated infection in a patient undergoing terminal complement inhibitor therapy. We suggest providing preventive dental care before eculizumab initiation.

Prevalence and incidence of kidney diseases leading to hospital admission in people living with HIV in France: an observational nationwide study.

OBJECTIVES: To describe hospitalisations for kidney disease (KD) among people living with HIV (PLHIV) in France and to identify the factors associated with such hospitalisations since data on the epidemiology of KD leading to hospitalisation are globally scarce. DESIGN: Observational nationwide study using the French Programme de Médicalisation des Systèmes d'Information database. SETTING: France 2008-2013. PARTICIPANTS: Around 10 862 PLHIV out of a mean of 5 210 856 patients hospitalised each year. All hospital admissions with a main diagnosis code indicating KD (International Classification of Diseases, 10th revision codes, N00 to -N39) were collected. MAIN OUTCOME MEASURES: The prevalence and incidence of KD leading to hospital admission in PLHIV and the associated risk factors. RESULTS: The prevalence of patients hospitalised for KD was 1.5 higher in PLHIV than in the general population, and increased significantly from 3.0% in 2008 to 3.7% in 2013 (p<0.01). The main cause of hospitalisation for KD was acute renal failure (ARF, 25.4%). Glomerular diseases remained stable (6.4%) throughout the study period, focal segmental glomerulosclerosis being the main diagnosis (37.6%). Only 41.3% of patients hospitalised for glomerular disease were biopsied. The other common motives for admission were nephrolithiasis (22.1%) and pyelonephritis (22.6%).The 5-year cumulative incidence of KD requiring hospitalisation was 5.9% in HIV patients newly diagnosed for HIV in 2009. Factors associated with a higher risk of incident KD requiring hospitalisation were cardiovascular disease (HR 3.30, 95% CI 1.46 to 7.49), and, for female patients, AIDS (HR 2.45, 95% CI 1.07 to 5.58). Two-thirds of hospitalisations for incident ARF occurred in the first 2 years of follow-up. CONCLUSIONS: Hospital admission for KD is more frequent in PLHIV than in the general population and increases over time. ARF remains the leading cause. Glomerular diseases are infrequently documented by renal biopsies. Older patients and those with cardiovascular disease are particularly concerned.