Detection of Human Visceral Leishmaniasis Parasites in Microscopy Images from Bone Marrow Parasitological Examination

Abstract

Visceral Leishmaniasis (VL) is a neglected disease that affects between 50,000 and 90,000 new cases annually worldwide. In Brazil, VL causes about 3500 cases/per year. This chronic disease can lead to death in 90% of untreated cases. Thus, it is necessary to study safe technologies for diagnosing, treating, and controlling VL. Specialized laboratories carry out the VL diagnosis, and this step has a significant automation power through methods based on computational tools. The gold standard for detecting VL is the microscopy of material aspirated from the bone marrow to search for amastigotes. This work aims to assist in detecting amastigotes from microscopy images using deep learning techniques. The proposed methodology consists of segmenting the Leishmania parasites in the images, precisely indicating the location of the amastigotes in the image. In the detection of VL parasites, in this methodology, a Dice of 80.4% was obtained, Intersection over Union (IoU) of 75.2%, Accuracy of 99.1%, Precision of 81.5%, Sensitivity of 72.2%, Specificity of 99.6%, and Area under the Receiver Operating Characteristics Curve (AUC) of 86.5%. The results are promising and demonstrate that deep learning models trained with images of microscopy slides of biological material can precisely help the specialist detect VL in humans.

Edge Artificial Intelligence (AI) for real-time automatic quantification of filariasis in mobile microscopy

“…Microscopy is a widely used technique for filariasis diagnosis, as it can distinguish parasite species, but it requires expert microscopists, and is time-consuming. Numerous studies have incorporated AI to aid in the diagnosis of microscopic images, targeting mostly malaria image analysis [43,44]-a recent review has identified 95 publications for malaria [27] -, and more recently also appeared works that deals with STH and schistosomiasis [24,32,[45][46][47][48], leishmaniasis [49], Chagas diseases [50,51], etc. The number of studies in this topic is very limited, but have yielded promising outcomes, aiming to facilitate the diagnosis of these diseases in LMICs.…”

“…Gonc ¸alves et al proposed the use of a u-net to segment human visceral leishmaniasis parasites on bone marrow samples. Developed with 150 images with 559 parasites (70% for training, 10% for validation and 20% for testing), the model achieved a Dice coefficient of 80.4%[49].Morais et al proposed the detection of chagas parasites using graph based segmentation algorithm and random forest. Developed with 33 mice samples with 1314 parasites (80% for training and 20% for testing)…”

How many labels do I need? Self-supervised learning strategies for multiple blood parasites classification in microscopy images

“…Despite the considerable efforts in this field, all previously proposed AI-based systems for parasite detection in microscopy blood images focus on specific parasites without considering the potential coexistence of various parasites within the same sample. The majority of methods exhibit this targeted approach, particularly in the context of malaria [3]- [6], while few methods address diseases like Chagas or leishmania [7]- [9], and none for filariasis. The absence of a universal method to detect any blood parasite in a given sample is a significant limitation, particularly in areas with coinfection [10], [11].…”