scholarly journals Diagnosing Malaria Patients with Plasmodium falciparum and vivax Using Deep Learning for Thick Smear Images

Diagnostics ◽  
2021 ◽  
Vol 11 (11) ◽  
pp. 1994
Author(s):  
Yasmin M. Kassim ◽  
Feng Yang ◽  
Hang Yu ◽  
Richard J. Maude ◽  
Stefan Jaeger

We propose a new framework, PlasmodiumVF-Net, to analyze thick smear microscopy images for a malaria diagnosis on both image and patient-level. Our framework detects whether a patient is infected, and in case of a malarial infection, reports whether the patient is infected by Plasmodium falciparum or Plasmodium vivax. PlasmodiumVF-Net first detects candidates for Plasmodium parasites using a Mask Regional-Convolutional Neural Network (Mask R-CNN), filters out false positives using a ResNet50 classifier, and then follows a new approach to recognize parasite species based on a score obtained from the number of detected patches and their aggregated probabilities for all of the patient images. Reporting a patient-level decision is highly challenging, and therefore reported less often in the literature, due to the small size of detected parasites, the similarity to staining artifacts, the similarity of species in different development stages, and illumination or color variations on patient-level. We use a manually annotated dataset consisting of 350 patients, with about 6000 images, which we make publicly available together with this manuscript. Our framework achieves an overall accuracy above 90% on image and patient-level.

2021 ◽  
Vol 20 (1) ◽  
Author(s):  
Mandella King ◽  
Alexander E. George ◽  
Pau Cisteró ◽  
Christine K. Tarr-Attia ◽  
Beatriz Arregui ◽  
...  

Abstract Background Malaria diagnosis in many malaria-endemic countries relies mainly on the use of rapid diagnostic tests (RDTs). The majority of commercial RDTs used in Africa detect the Plasmodium falciparum histidine-rich protein 2 (PfHRP2). pfhrp2/3 gene deletions can therefore lead to false-negative RDT results. This study aimed to evaluate the frequency of PCR-confirmed, false-negative P. falciparum RDT results in Monrovia, Liberia. Methods PfHRP2-based RDT (Paracheck Pf®) and microscopy results from 1038 individuals with fever or history of fever (n = 951) and pregnant women at first antenatal care (ANC) visit (n = 87) enrolled in the Saint Joseph’s Catholic Hospital (Monrovia) from March to July 2019 were used to assess the frequency of false-negative RDT results. True–false negatives were confirmed by detecting the presence of P. falciparum DNA by quantitative PCR in samples from individuals with discrepant RDT and microscopy results. Samples that were positive by 18S rRNA qPCR but negative by PfHRP2-RDT were subjected to multiplex qPCR assay for detection of pfhrp2 and pfhrp3. Results One-hundred and eighty-six (19.6%) and 200 (21.0%) of the 951 febrile participants had a P. falciparum-positive result by RDT and microscopy, respectively. Positivity rate increased with age and the reporting of joint pain, chills and shivers, vomiting and weakness, and decreased with the presence of coughs and nausea. The positivity rate at first ANC visit was 5.7% (n = 5) and 8% (n = 7) by RDT and microscopy, respectively. Out of 207 Plasmodium infections detected by microscopy, 22 (11%) were negative by RDT. qPCR confirmed absence of P. falciparum DNA in the 16 RDT-negative but microscopy-positive samples which were available for molecular testing. Among the 14 samples that were positive by qPCR but negative by RDT and microscopy, 3 only amplified pfldh, and among these 3 all were positive for pfhrp2 and pfhrp3. Conclusion There is no qPCR-confirmed evidence of false-negative RDT results due to pfhrp2/pfhrp3 deletions in this study conducted in Monrovia (Liberia). This indicates that these deletions are not expected to affect the performance of PfHRP2-based RDTs for the diagnosis of malaria in Liberia. Nevertheless, active surveillance for the emergence of PfHRP2 deletions is required.


2021 ◽  
pp. 113472
Author(s):  
Young Lo ◽  
Yee-Wai Cheung ◽  
Lin Wang ◽  
Megan Lee ◽  
Gabriela Figueroa-Miranda ◽  
...  

2006 ◽  
Vol 50 (4) ◽  
pp. 1586-1589 ◽  
Author(s):  
Audrey Gego ◽  
Olivier Silvie ◽  
Jean-François Franetich ◽  
Khemaïs Farhati ◽  
Laurent Hannoun ◽  
...  

ABSTRACT Plasmodium liver stages represent potential targets for antimalarial prophylactic drugs. Nevertheless, there is a lack of molecules active on these stages. We have now developed a new approach for the high-throughput screening of drug activity on Plasmodium liver stages in vitro, based on an infrared fluorescence scanning system. This method allowed us to count automatically and rapidly Plasmodium-infected hepatocytes, using different hepatic cells and different Plasmodium species, including Plasmodium falciparum. This new technique is well adapted for high-throughput drug screening and should facilitate the identification of new antimalarial compounds active on Plasmodium liver stages.


2021 ◽  
Vol 5 (3) ◽  
pp. e202101237
Author(s):  
Kutub Ashraf ◽  
Shahin Tajeri ◽  
Christophe-Sébastien Arnold ◽  
Nadia Amanzougaghene ◽  
Jean-François Franetich ◽  
...  

Artemisinin-based combination therapies (ACT) are the frontline treatments against malaria worldwide. Recently the use of traditional infusions from Artemisia annua (from which artemisinin is obtained) or Artemisia afra (lacking artemisinin) has been controversially advocated. Such unregulated plant-based remedies are strongly discouraged as they might constitute sub-optimal therapies and promote drug resistance. Here, we conducted the first comparative study of the anti-malarial effects of both plant infusions in vitro against the asexual erythrocytic stages of Plasmodium falciparum and the pre-erythrocytic (i.e., liver) stages of various Plasmodium species. Low concentrations of either infusion accounted for significant inhibitory activities across every parasite species and stage studied. We show that these antiplasmodial effects were essentially artemisinin-independent and were additionally monitored by observations of the parasite apicoplast and mitochondrion. In particular, the infusions significantly incapacitated sporozoites, and for Plasmodium vivax and P. cynomolgi, disrupted the hypnozoites. This provides the first indication that compounds other than 8-aminoquinolines could be effective antimalarials against relapsing parasites. These observations advocate for further screening to uncover urgently needed novel antimalarial lead compounds.


2010 ◽  
Vol 278 (1712) ◽  
pp. 1705-1712 ◽  
Author(s):  
Tiago Antao ◽  
Ian M. Hastings

Plasmodium falciparum malaria is subject to artificial selection from antimalarial drugs that select for drug-resistant parasites. We describe and apply a flexible new approach to investigate how epistasis, inbreeding, selection heterogeneity and multiple simultaneous drug deployments interact to influence the spread of drug-resistant malaria. This framework recognizes that different human ‘environments’ within which treatment may occur (such as semi- and non-immune humans taking full or partial drug courses) influence the genetic interactions between parasite loci involved in resistance. Our model provides an explanation for how the rate of spread varies according to different malaria transmission intensities, why resistance might stabilize at intermediate frequencies and also identifies several factors that influence the decline of resistance after a drug is removed. Results suggest that studies based on clinical outcomes might overestimate the spread of resistant parasites, especially in high-transmission areas. We show that when transmission decreases, prevalence might decrease without a corresponding change in frequency of resistance and that this relationship is heavily influenced by the extent of linkage disequilibrium between loci. This has important consequences on the interpretation of data from areas where control is being successful and suggests that reducing transmission might have less impact on the spread of resistance than previously expected.


2019 ◽  
pp. 1633-1655
Author(s):  
Catalina Soriana Sitnikov ◽  
Claudiu Bocean ◽  
Sorin Tudor

Currently, the adoption of a specific approach to business activities that highlights the strategic importance of corporate social responsibility hereafter CSR is the most important element influencing the existence and continuity of an organization. Thus, there is not a surprise that universities shall identify, in terms of own activities, the possibility to lead their orientation beyond teaching-learning process, towards the operations and institutional activities. At the same time, recent decades have experienced the failure of CSR as a way of doing business, govern or provide solutions and evaluate ethical issues and, thus, of the need to apply and implement a new approach - CSR 2.0. The transition from the current CSR, or 1.0, to CSR 2.0 requires the adoption of five new principles—creativity, scalability, responsiveness, glocality, and circularity—and embedding them within organizations management and culture. The paper will unfold towards two steps: the first, dedicated to the correlation between education (Blessinger's models and frameworks elements) with business (based on higher education business models), and the second, represented by integrating the new built model with the concepts and principles of CSR 2.0 developed by Visser. The new framework can be used to manage the context and processes of a socially responsible university as part of a world influenced by CSR 2.0 principles.


1993 ◽  
Vol 35 (6) ◽  
pp. 495-502 ◽  
Author(s):  
Maria Carmen Arroyo Sanchez ◽  
Sandra do Lago Moraes de Avila ◽  
Vera de Paula Quartier-Oliveira ◽  
Antonio Walter Ferreira

The study evaluated six Plasmodium falciparum antigen extracts to be used in the IgG and IgM enzyme-linked immunosorbent assays (ELISA), for malaria diagnosis and epidemiological studies. Results obtained with eighteen positive and nine negative control sera indicated that there were statistically significant differences among these antigen extracts (Multifactor ANOVA, p< 0.0001). Urea, sodium deoxycholate and Zwittergent antigen extracts performed better than did the three others, their features being very similar for the detection of IgG antibodies. Urea, alkaline and sodium deoxycholate antigen extracts proved to be better than the others for the detection of IgM antibodies. A straight line relationship was found between the optical densities (or their respective log 10) and the log 10 of antibody dilutions, with a very constant slope. Thus serum titers could be determined by direct titration and by two different equations, needing only one serum dilution. For IgM antibody detections, log 10 expression gave results that better correlated with direct titration (95% Bonferroni). For IgG antibody detections, the titer differences were not significant. The reproducibility of antibody titers and antigen batches was also evaluated, giving satisfactory results.


2019 ◽  
Vol 19 (1) ◽  
Author(s):  
Mesele Damte Argaw ◽  
Thandisizwe Redford Mavundla ◽  
Kassa Daka Gidebo

Abstract Background Malaria is one of the leading public health problems in sub-Saharan Africa that contributes to significant patient morbidity and mortality. The aim of the study was to investigate adherence to malaria diagnosis and treatment guidelines by private health sector providers and compare their performance against the public private partnership (PPP) status. Methods A facility-based retrospective clinical audit was conducted between October 2016 and January 2017 in 11 medium clinics in the West Gojjam zone of the Amhara Region, North-west Ethiopia. Data was extracted from patient medical records using pretested data abstraction forms. Descriptive statistics were employed to present the findings and adherence of health workers against the national and international standards were classified as ideal, acceptable, minor error and major error for both malaria diagnosis and treatment. A chi-square (X2) test was used to test for a statistically significant relationship after the data had been categorized using public private partnership status at P < 0.05. Results One thousand six hundred fifty clinical files were audited. All malaria suspected patients were investigated either with microscopy or rapid diagnostics test (RDT) for parasitological confirmation. The proportion of malaria treated cases was 23.7% (391/1650). Of which 16.6% (274/1650) were uncomplicated, 3.69% (61 /1650) were severe and complicated and the rest 3.39% (56/1650) were clinical diagnosed malaria cases. And the malaria parasite positivity rate was 20.30% (335/1650). All malaria suspected patients were not investigated with ideal malaria diagnosis recommendations; only 19.4% (320/1650) were investigated with acceptable malaria diagnosis (public private partnership (PPP) 19.4%; 176/907; and non-public private partnership (NPPP) 19.38%; 144/743, X2 (1) = 0.0With regards to treatments of malaria cases, the majority 82.9% of Plasmodium vivax cases were managed with ideal recommended treatment (X2 (1) = 0.35, P = 0.55); among Plasmodium falciparum, mixed (Plasmodium falciparum and Plasmodium vivax). Conclusion The clinical audit revealed that the majority of malaria patients had received minor error malaria diagnostic services. In addition, only one fifth of malaria patients had received ideal malaria treatment services. To understand the reasons for the low levels of malaria diagnosis and treatment adherence with national guidelines, a qualitative exploratory descriptive study is recommended.


2014 ◽  
Vol 13 (1) ◽  
pp. 110 ◽  
Author(s):  
Marie Varela ◽  
Romy Razakandrainibe ◽  
Delphine Aldebert ◽  
Jean Barale ◽  
Ronan Jambou

Sign in / Sign up

Export Citation Format

Share Document