Evaluating Deep Learning Algorithms in Pulmonary Nodule Detection.

Lung cancer is considered the deadliest cancer worldwide. In order to detect it, radiologists need to inspect multiple Computed Tomography (CT) scans. This task is tedious and time consuming. In recent years, promising methods based on deep learning object detection algorithms were proposed for the automatic nodule detection and classification. With those techniques, Computed Aided Detection (CAD) software can be developed to alleviate radiologist's burden and help speed-up the screening process. However, among available object detection frameworks, there are just a limited number that have been used for this purpose. Moreover, it can be challenging to know which one to choose as a baseline for the development of a new application for this task. Hence, in this work we propose a benchmark of recent state-of-the-art deep learning detectors such as Faster-RCNN, YOLO, SSD, RetinaNet and EfficientDet in the challenging task of pulmonary nodule detection. Evaluation is done using automatically segmented 2D images extracted from volumetric chest CT scans.

Specific isotype immune response in the diagnosis of human schistosomiasis pathology?

Since the few indirect markers available for assessing the development and the stage of intestinal schistosomiasis morbidity are weakly specific, endoscopy is still the only method able to detect severe forms of pathology. Therefore, we evaluated the isotype antibody response to the current schistosome antigen preparation (soluble egg antigens [SEA]) in 142 Senegalese patients infected with Schistosoma mansoni. They were stratified into three different stages of pathology according to ultrasonographic, endoscopic, and clinical parameters (stage 1 = no detectable pathology; stage 2 = moderate morbidity; stage 3 = severe forms of pathology). Only median specific IgG4, IgE, and IgA responses changed according to the stage of pathology. The IgA level was significantly higher in stages 2 and 3 compared with stage 1, and the IgE level was higher in stage 3 compared with stage 1. A high specific IgG4 level was observed only in stage 3 and was significantly different compared with stage 2. We show an association between the variability of the specific response to SEA and the degree of morbidity, and demonstrate that IgA and IgG4 responses could be combined markers to easily discriminate the different stages of pathology due to infection with S. mansoni.

Increase of malaria attacks among children presenting concomitant infection by Schistosoma mansoni in Senegal.

Helminthic infections concomitant with malaria are common in inter-tropical areas. A recent study showed that mice co-infected with Schistosoma mansoni and Plasmodium chabaudi develop higher P. chabaudi parasitaemia and had a higher mortality rate. This important observation deserved to be further investigated among human populations.Malaria attacks were recorded in 512 children aged 6-15 years living in Richard Toll (Northern Senegal) among whom 336 were infected by S. mansoni, and 175 were not. The incidence rate of malaria attacks was significantly higher among S. mansoni-infected individuals, particularly those carrying the highest worm loads, as compared to uninfected subjects (26.6% versus 16,4 %). In contrast, the rate of malaria attacks was lower, without reaching significance, in medium grade S. mansoni infections. Thus, infection by S. mansoni affects susceptibility to malaria, but this can vary according to the intensity of parasite load. The immunological mechanisms underlying this dual effect need to be further explored.