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

Filariasis, a neglected tropical disease caused by roundworms, is a significant public health concern in many tropical countries. Microscopic examination of blood samples can detect and differentiate parasite species, but it is time consuming and requires expert microscopists, a resource that is not always available. In this context, artificial intelligence (AI) can assist in the diagnosis of this disease by automatically detecting and differentiating microfilariae. In line with the target product profile for lymphatic filariasis as defined by the World Health Organization, we developed an edge AI system running on a smartphone whose camera is aligned with the ocular of an optical microscope that detects and differentiates filarias species in real time without the internet connection. Our object detection algorithm that uses the Single-Shot Detection (SSD) MobileNet V2 detection model was developed with 115 cases, 85 cases with 1903 fields of view and 3342 labels for model training, and 30 cases with 484 fields of view and 873 labels for model validation before clinical validation, is able to detect microfilariae at 10x magnification and distinguishes four species of them at 40x magnification: Loa loa, Mansonella perstans, Wuchereria bancrofti, and Brugia malayi. We validated our augmented microscopy system in the clinical environment by replicating the diagnostic workflow encompassed examinations at 10x and 40x with the assistance of the AI models analyzing 18 samples with the AI running on a middle range smartphone. It achieved an overall precision of 94.14%, recall of 91.90% and F1 score of 93.01% for the screening algorithm and 95.46%, 97.81% and 96.62% for the species differentiation algorithm respectively. This innovative solution has the potential to support filariasis diagnosis and monitoring, particularly in resource-limited settings where access to expert technicians and laboratory equipment is scarce.

Digital Microscopy Augmented by Artificial Intelligence to Interpret Bone Marrow Samples for Hematological Diseases

Analysis of bone marrow aspirates (BMAs) is an essential step in the diagnosis of hematological disorders. This analysis is usually performed based on a visual examination of samples under a conventional optical microscope, which involves a labor-intensive process, limited by clinical experience and subject to high observer variability. In this work, we present a comprehensive digital microscopy system that enables BMA analysis for cell type counting and differentiation in an efficient and objective manner. This system not only provides an accessible and simple method to digitize, store, and analyze BMA samples remotely but is also supported by an Artificial Intelligence (AI) pipeline that accelerates the differential cell counting process and reduces interobserver variability. It has been designed to integrate AI algorithms with the daily clinical routine and can be used in any regular hospital workflow.

Unraveling response to temozolomide in preclinical GL261 glioblastoma with MRI/MRSI using radiomics and signal source extraction

Glioblastoma is the most frequent aggressive primary brain tumor amongst human adults. Its standard treatment involves chemotherapy, for which the drug temozolomide is a common choice. These are heterogeneous and variable tumors which might benefit from personalized, data-based therapy strategies, and for which there is room for improvement in therapy response follow-up, investigated with preclinical models. This study addresses a preclinical question that involves distinguishing between treated and control (untreated) mice bearing glioblastoma, using machine learning techniques, from magnetic resonance-based data in two modalities: MRI and MRSI. It aims to go beyond the comparison of methods for such discrimination to provide an analytical pipeline that could be used in subsequent human studies. This analytical pipeline is meant to be a usable and interpretable tool for the radiology expert in the hope that such interpretation helps revealing new insights about the problem itself. For that, we propose coupling source extraction-based and radiomics-based data transformations with feature selection. Special attention is paid to the generation of radiologist-friendly visual nosological representations of the analyzed tumors.

A SR-NET 3D-TO-2D ARCHITECTURE FOR PARASEPTAL EMPHYSEMA SEGMENTATION

Paraseptal emphysema (PSE) is a relatively unexplored emphysema subtype that is usually asymptomatic, but recently associated with interstitial lung abnormalities which are related with clinical outcomes, including mortality. Previous local-based methods for emphysema subtype quantification do not properly characterize PSE. This is in part for their inability to properly capture the global aspect of the disease, as some the PSE lesions can involved large regions along the chest wall. It is our assumption, that path-based approaches are not well-suited to identify this subtype and segmentation is a better paradigm. In this work we propose and introduce the Slice-Recovery network (SR-Net) that leverages 3D contextual information for 2D segmentation of PSE lesions in CT images. For that purpose, a novel convolutional network architecture is presented, which follows an encoding-decoding path that processes a 3D volume to generate a 2D segmentation map. The dataset used for training and testing the method comprised 664 images, coming from 111 CT scans. The results demonstrate the benefit of the proposed approach which incorporate 3D context information to the network and the ability of the proposed method to identify and segment PSE lesions with different sizes even in the presence of other emphysema subtypes in an advanced stage.

Collaborative intelligence and gamification for on-line malaria species differentiation

Background

Current World Health Organization recommendations for the management of malaria include the need for a parasitological confirmation prior to triggering appropriate treatment. The use of rapid diagnostic tests (RDTs) for malaria has contributed to a better infection recognition and a more targeted treatment. Nevertheless, low-density infections and parasites that fail to produce HRP2 can cause false-negative RDT results. Microscopy has traditionally been the methodology most commonly used to quantify malaria and characterize the infecting species, but the wider use of this technique remains challenging, as it requires trained personnel and processing capacity.

Objective

In this study, the feasibility of an on-line system for remote malaria species identification and differentiation has been investigated by crowdsourcing the analysis of digitalized infected thin blood smears by non-expert observers using a mobile app.

Methods

An on-line videogame in which players learned how to differentiate the young trophozoite stage of the five Plasmodium species has been designed. Images were digitalized with a smartphone camera adapted to the ocular of a conventional light microscope. Images from infected red blood cells were cropped and puzzled into an on-line game. During the game, players had to decide the malaria species (Plasmodium falciparum, Plasmodium malariae, Plasmodium vivax, Plasmodium ovale, Plasmodium knowlesi) of the infected cells that were shown in the screen. After 2 months, each player’s decisions were analysed individually and collectively.

Results

On-line volunteers playing the game made more than 500,000 assessments for species differentiation. Statistically, when the choice of several players was combined (n > 25), they were able to significantly discriminate Plasmodium species, reaching a level of accuracy of 99% for all species combinations, except for P. knowlesi (80%). Non-expert decisions on which Plasmodium species was shown in the screen were made in less than 3 s.

Conclusion

These findings show that it is possible to train malaria-naïve non-experts to identify and differentiate malaria species in digitalized thin blood samples. Although the accuracy of a single player is not perfect, the combination of the responses of multiple casual gamers can achieve an accuracy that is within the range of the diagnostic accuracy made by a trained microscopist.

Electronic supplementary material

The online version of this article (10.1186/s12936-019-2662-9) contains supplementary material, which is available to authorized users.

EMPHYSEMA CLASSIFICATION USING A MULTI-VIEW CONVOLUTIONAL NETWORK

In this article we propose and validate a fully automatic tool for emphysema classification in Computed Tomography (CT) images. We hypothesize that a relatively simple Convolutional Neural Network (CNN) architecture can learn even better discriminative features from the input data compared with more complex and deeper architectures. The proposed architecture is comprised of only 4 convolutional and 3 pooling layers, where the input corresponds to a 2.5D multiview representation of the pulmonary segment tissue to classify, corresponding to axial, sagittal and coronal views. The proposed architecture is compared to similar 2D CNN and 3D CNN, and to more complex architectures which involve a larger number of parameters (up to six times larger). This method has been evaluated in 1553 tissue samples, and achieves an overall sensitivity of 81.78 % and a specificity of 97.34%, and results show that the proposed method outperforms deeper state-of-the-art architectures particularly designed for lung pattern classification. The method shows satisfactory results in full-lung classification.

Image processing challenges in the creation of spatiotemporal gene expression atlases of developing embryos

To properly understand and model animal embryogenesis it is crucial to obtain detailed measurements, both in time and space, about their gene expression domains and cell dynamics. Such challenge has been confronted in recent years by a surge of atlases which integrate a statistically relevant number of different individuals to get robust, complete information about their spatiotemporal locations of gene patterns. This paper will discuss the fundamental image analysis strategies required to build such models and the most common problems found along the way. We also discuss the main challenges and future goals in the field.

Motion estimation of tagged cardiac magnetic resonance images using variational techniques

This work presents a new method for motion estimation of tagged cardiac magnetic resonance sequences based on variational techniques. The variational method has been improved by adding a new term in the optical flow equation that incorporates tracking points with high stability of phase. Results were obtained through simulated and real data, and were validated by manual tracking and with respect to a reference state-of-the-art method: harmonic phase imaging (HARP). The error, measured in pixels per frame, obtained with the proposed variational method is one order of magnitude smaller than the one achieved by the reference method, and it requires a lower computational cost.

An automatic quantification and registration strategy to create a gene expression atlas of zebrafish embryogenesis

In order to properly understand and model the gene regulatory networks in animals development, it is crucial to obtain detailed measurements, both in time and space, about their gene expression domains. In this paper, we propose a complete computational framework to fulfill this task and create a 3D Atlas of the early zebrafish embryogenesis annotated with both the cellular localizations and the level of expression of different genes at different developmental stages. The strategy to construct such an Atlas is described here with the expression pattern of 5 different genes at 6 hours of development post fertilization.

Cardiac motion analysis from ultrasound sequences using nonrigid registration: validation against Doppler tissue velocity

Early detection of cardiac motion abnormalities is one of the main goals of quantitative cardiac image processing. This article presents a new method to compute the 2-D myocardial motion parameters from gray-scale 2-D echocardiographic sequences, making special emphasis on the validation of the proposed technique in comparison with Doppler tissue imaging. Myocardial motion is computed using a frame-to-frame nonrigid registration technique on the whole sequence. The key feature of our method is the use of an analytical representation of the myocardial displacement based on a semilocal parametric model of the deformation using Bsplines. Myocardial motion analysis is performed to obtain displacement, velocity and strain parameters. Robustness and speed are achieved by introducing a multiresolution optimization strategy. To validate the method, velocity measurements in three different regions-of-interest in the septum have been compared with those obtained with Doppler tissue velocity in healthy and pathologic subjects. Regression and Bland-Altman analysis show very good agreement between the two different approaches, with the great advantage that the new method overcomes the angle-dependency limitations of the Doppler techniques, providing both longitudinal and radial measurements.