Comparative Performance Evaluation of Routine Malaria Diagnosis at Ho Municipal Hospital

Autokeras Approach: A Robust Automated Deep Learning Network for Diagnosis Disease Cases in Medical Images

Automated deep learning is promising in artificial intelligence (AI). However, a few applications of automated deep learning networks have been made in the clinical medical fields. Therefore, we studied the application of an open-source automated deep learning framework, Autokeras, for detecting smear blood images infected with malaria parasites. Autokeras is able to identify the optimal neural network to perform the classification task. Hence, the robustness of the adopted model is due to it not needing any prior knowledge from deep learning. In contrast, the traditional deep neural network methods still require more construction to identify the best convolutional neural network (CNN). The dataset used in this study consisted of 27,558 blood smear images. A comparative process proved the superiority of our proposed approach over other traditional neural networks. The evaluation results of our proposed model achieved high efficiency with impressive accuracy, reaching 95.6% when compared with previous competitive models.

Accuracy of diagnosis among clinical malaria patients: comparing microscopy, RDT and a highly sensitive quantitative PCR looking at the implications for submicroscopic infections

BACKGROUND

The World Health Organization recommends parasitological confirmation of all suspected malaria cases by microscopy or rapid diagnostic tests (RDTs) before treatment. These conventional tools are widely used for point-of-care diagnosis in spite of their poor sensitivity at low parasite density. Previous studies in Ghana have compared microscopy and RDT using standard 18S rRNA PCR as reference with varying outcomes. However, how these conventional tools compare with ultrasensitive varATS qPCR has not been studied. This study, therefore, sought to investigate the clinical performance of microscopy and RDT assuming highly sensitive varATS qPCR as gold standard.

METHODS

1040 suspected malaria patients were recruited from two primary health care centers in the Ashanti Region of Ghana and tested for malaria by microscopy, RDT, and varATS qPCR. The sensitivity, specificity, and predictive values were assessed using varATS qPCR as gold standard.

RESULTS

Parasite prevalence was 17.5%, 24.5%, and 42.1% by microscopy, RDT, and varATS qPCR respectively. Using varATS qPCR as the standard, RDT was more sensitive (55.7% vs 39.3%), equally specific (98.2% vs 98.3%), and reported higher positive (95.7% vs 94.5%) and negative predictive values (75.3% vs 69.0%) than microscopy. Consequently, RDT recorded better diagnostic agreement (kappa = 0.571) with varATS qPCR than microscopy (kappa = 0.409) for clinical detection of malaria.

CONCLUSIONS

RDT outperformed microscopy for the diagnosis of Plasmodium falciparum malaria in the study. However, both tests missed over 40% of infections that were detected by varATS qPCR. Novel tools are needed to ensure prompt diagnosis of all clinical malaria cases.

Performance of an HRP-2 based (First Response®) and p-LDH-based (Optimal®) rapid diagnostic tests for diagnosis of malaria in paediatric sickle cell disease patients

BACKGROUND

Rapid diagnostic tests (RDTs) have been extensively evaluated and play an important role in malaria diagnosis. However, the accuracy of RDT for malaria diagnosis in patients with sickle cell disease (SCD) is unknown.

METHODS

We compared the performance of a histidine rich protein 2 (HRP-2) - based RDT (First Response®), and a lactate dehydrogenase (LDH) - based RDT (Optimal®) with routine microscopy as reference standard in 445 SCD children with an acute febrile illness in Accra, Ghana.

RESULTS

The overall sensitivity, specificity, positive and negative predictive values of the HRP-2 based RDT were, 100%, 95.7%, 73.8% and 100%, respectively. Comparable values for the LDH based RDT were, 91.7%, 99.5%, 95.7% and 99.0% respectively. A total of 423 results were true in both tests, one result was false in both tests, 16 results were false in the HRP-2 test only and 5 were false in the LDH test only (McNemar's test, p=0.03). At follow up, 73.7 % (28/38), 52.6 % (20/38), 48.6 % (17/35), and 13.2 % (5/38) of study participants were HRP-2 positive on days 14, 28, 35 and 42, respectively, compared to 0, 2.6 % (1/38), 2.9 % (1/35), and 2.6 % (1/38) for LDH.

CONCLUSION

The HRP2 based RDT fulfilled WHO criteria for malaria diagnosis in SCD patients and may provide diagnostic evidence for treatment to begin in cases where treatment would otherwise have begun presumptively based on symptoms, while LDH based RDT may be more suitable as a confirmatory test in low-parasitaemic sub-groups, such as SCD patients.

Diagnostic Accuracy of CareStart™ Malaria HRP2 and SD Bioline Pf/PAN for Malaria in Febrile Outpatients in Varying Malaria Transmission Settings in Cameroon

Background: There was an increase in the number of malaria cases in Cameroon in 2018 that could reflect changes in provider practice, despite effective interventions. In this study, we assessed the diagnostic performance of two malaria rapid diagnostic tests (mRDTs) for diagnostic confirmation of suspected cases of malaria in public and private health facilities in two malaria transmission settings in Cameroon. Methods: We evaluated the diagnostic performance of CareStart pf and SD Bioline Pf/PAN mRDT and compared these parameters by RDT type and transmission setting. Nested PCR and blood film microscopy were used as references. The chi square test was used for independent sample comparisons, while the McNemar’s test was used to test for the dependence of categorical data in paired sample testing. A p < 0.05 was considered significant in all comparisons. The R (v.4.0.2) software was used for analyses. Results: A total of 1126 participants consented for the study in the four sites. The diagnostic accuracy of the CareStart Pf mRDT was 0.93.6% (0.911–0.961) in Yaoundé, 0.930% (0.90–0.960) in Ngounso, 0.84% (0.794–0.891) in St Vincent Catholic Hospital Dschang and 0.407 (0.345–0.468) in Dschang district hospital. For SD Bioline Pf/PAN the accuracy was 0.759 (0.738–0.846) for St Vincent Catholic Hospital Dschang and 0.426 (0.372–0.496) for the Dschang district hospital. The accuracy was slightly lower in each case but not statistically different when PCR was considered as the reference. The likelihood ratios of the positive and negative tests were high in the high transmission settings of Yaoundé (10.99 (6.24–19.35)) and Ngounso (14.40 (7.89–26.28)) compared to the low transmission settings of Dschang (0.71 (0.37–1.37)) and St Vincent Catholic hospital (7.37 (4.32−12.59)). There was a high degree of agreement between the tests in Yaoundé (Cohen’s Kappa: 0.85 ± 0.05 (0.7–0.95)) and Ngounso (Cohen’s Kappa: 0.86 ± 0.05 (0.74, 0.97)) and moderate agreement in St Vincent hospital Dschang (k: 0.58 ± 0.06 (0.44–0.71)) and poor agreement in the District Hospital Dschang (Cohen’s Kappa: −0.11 ± 0.05 (−0.21–0.01)). The diagnostic indicators of the SD Bioline Pf/PAN were slightly better than for CareStart Pf mRDT in St Vincent Catholic hospital Dschang, irrespective of the reference test. Conclusions: Publicly procured malaria rapid diagnostic tests in Cameroon have maintained high accuracy (91–94%) in the clinical diagnosis of malaria in high malaria transmission regions of Cameroon, although they failed to reach WHO standards. We observed an exception in the low transmission region of Dschang, West region, where the accuracy tended to be lower and variable between facilities located in this town. These results underscore the importance of the routine monitoring of the quality and performance of malaria RDTs in diverse settings in malaria endemic areas.

Diagnostic performance of rapid diagnostic test, light microscopy and polymerase chain reaction during mass survey conducted in low and high malaria-endemic areas from two North-Eastern states of India

An early and accurate diagnosis followed by prompt treatment is pre-requisite for the management of any disease. Malaria diagnosis is routinely performed by microscopy and rapid diagnostic tests (RDTs) in the field settings; however, their performance may vary across regions, age and asymptomatic status. Owing to this, we assessed the diagnostic performance of conventional and advanced molecular tools for malaria detection in low and high malaria-endemic settings. We performed mass blood surveys in low and high endemic regions of two North-Eastern districts from the states of Assam and Meghalaya. A total of 3322 individuals were screened for malaria using RDT, microscopy and PCR and measures of diagnostic accuracy were estimated. Out of 3322 individuals, 649 (19.5%) were detected with malaria parasite. Asymptomatic were 86.4% (2872/3322), of which 19.4% (557/2872) had Plasmodium infection. The sensitivity and specificity of microscopy were 42.7% and 99.3%, and RDT showed 49.9% and 90.4%, respectively, considering PCR as standard. RDT (AUC: 0.65 vs 0.74; p = 0.001) and microscopy (AUC: 0.64 vs 0.76; p < 0.0001) performances were significantly lower in low compared to high endemic areas. True positive rate was lower in asymptomatics but true negative rate was found similar to symptomatic individuals. The conventional diagnostic tools (RDT and microscopy) had detected malaria in children with nearly twofold greater sensitivity than in the adults (p < 0.05). To conclude, asymptomatics, adults and low malaria-endemic regions require major attention due to mediocre performance of conventional diagnostic tools in malaria detection.

Overdiagnosis of Malaria Illness in an Endemic Setting: A Facility-Based Surveillance Study in Malawi

In endemic settings where asymptomatic malaria infections are common, malaria infection can complicate fever diagnosis. Factors influencing fever misdiagnosis, including accuracy of malaria rapid diagnostic tests (mRDTs) and the malaria-attributable fraction of fevers (MAF), require further investigation. We conducted facility-based surveillance in Malawi, from January 2012 through December 2013 in settings of high perennial (Chikhwawa), high seasonal (Thoylo), and moderate seasonal (Ndirande) malaria transmission. Consecutive patients presenting to outpatient departments were screened; those with suspected malaria illness were tested by mRDT or routine thick-smear microscopy. Test positivity rates (TPRs), positive predictive value (PPVs) of mRDTs, and MAFs were calculated by site, age, and season. Of 41,471 patients, 10,052 (24.2%) tested positive for malaria. The TPR was significantly greater in Chikhwawa (29.9%; 95% CI, 28.6-30.0) compared with Thyolo (13.2%; 95% CI, 12.5-13.7) and Ndirande (13.1%; 95% CI, 12.2-14.4). The overall PPV was 77.8% (95% CI, 76.8-78.7); it was lowest among infants (69.9%; 95% CI, 65.5-74.2) and highest among school-age children (81.9%; 95% CI, 80.3-83.4). Malaria infection accounted for about 50% of fevers in children younger than 5 years old with microscopy-confirmed Plasmodium falciparum infection, and less than 20% of such fevers in school-age children. Outpatient settings in Malawi had a high burden of malaria illness, but also possible overdiagnosis of malaria illness. Interventions to reduce malaria transmission and rapid testing for other common febrile illness may improve diagnostic clarity among outpatients in malaria endemic settings.

Inter-rater Variability in Malaria Microscopy at the LEKMA Hospital, Ghana

BACKGROUND

Malaria remains a major cause of morbidity and mortality worldwide and particularly in sub-Saharan Africa where it is endemic. As such, it is important that a proper diagnosis is made before treatment is initiated. Malaria parasite count plays a key role in the diagnosis and management of malaria. Variations in ratings by laboratory personnel can impact negatively on the treatment regimen for malaria-infected patients. The study is thus aimed at evaluating and comparing the proficiency and parasitaemia counts by two different categories of laboratory staff at the LEKMA Hospital, Ghana.

MATERIALS AND METHODS

A total of 200 confirmed malaria-positive samples were used in the study. Six thick and thin films were prepared from each sample and uniquely labelled. Two of the six slides were given to two WHO-accredited malaria microscopists to examine and report their respective parasite count/μl (parasite count/WBC × 8000). These were used as the reference for the two categories of laboratory staffs: rater A being diploma holders (Technical Officers referred to as untrained rater) and rater B being degree holders (Medical Laboratory Scientist referred to as trained rater) at the LEKMA Hospital.

RESULTS

In comparison to the expected outcome, the parasite count by the rater group A (190 (151-239)]) and the rater group B (177 (140-224)) demonstrated significant positive correlation (r = 0.995, p < 0.0001 vs. r = 0.995, p < 0.0001, respectively) with the expected outcome in the cases of heavy parasitaemia. A statistically significant difference (p < 0.05) between counts by the different raters in low parasitemia was observed in this study. A persistent nosedive inter-rater agreement from k = 0.82 to k = 0.40 with increasing density cutoff was observed in this study.

CONCLUSION

The study observed that the degree of inter-rater agreement of parasite density count by various categories of laboratory personnel is almost perfect. However, the parasite count between raters varied significantly with very low levels of parasitemia but better correlated with heavy parasitemia.

Prevalence of Asymptomatic Malaria among Children in the Tamale Metropolis: How Does the PfHRP2 CareStart™ RDT Perform against Microscopy?

Background

Asymptomatic carriage of the malaria parasites, likewise its misdiagnosis, especially false negatives, due to the use of substandard rapid diagnosis tests (RDTs) has been shown to hinder the progress of the fight against malaria.

Method

The study assessed the prevalence of asymptomatic malaria as well as the performance of Plasmodium falciparum-specific protein and histidine-rich protein 2 (PfHRP2) CareStart™ RDT against standard microscopy in the detection of malaria infection among 345 children (1–15 yrs) from two (2) basic schools in Tamale Metropolis.

Results

From the microscopy (considered as gold standard), prevalence of malaria among the asymptomatic children was found to be 2.6%, with sensitivity and specificity of CareStart™ RDT in detecting P. falciparum infections found to be 55.6% and 93.8%, respectively. The positive predictive value (PPV) and negative predictive value (NPV) of CareStart™ RDT were found to be 19.23% and 98.45%, respectively. There was an evidence showing a significant relation between CareStart™ RDT and microscopy in determining malaria infection (χ² = 30.579, p < 0.001).

Conclusion

Prevalence of asymptomatic malaria among children was found to be 2.6%. The study reported low sensitivity and PPV for PfHRP2 CareStart™ RDT in an asymptomatic population at instances of low parasitaemia.

A scoping review on the field validation and implementation of rapid diagnostic tests for vector-borne and other infectious diseases of poverty in urban areas

BACKGROUND

Health personnel face challenges in diagnosing vector-borne and other diseases of poverty in urban settings. There is a need to know what rapid diagnostic technologies are available, have been properly assessed, and are being implemented to improve control of these diseases in the urban context. This paper characterizes evidence on the field validation and implementation in urban areas of rapid diagnostics for vector-borne diseases and other diseases of poverty.

MAIN BODY

A scoping review was conducted. Peer-reviewed and grey literature were searched using terms describing the targeted infectious diseases, diagnostics evaluations, rapid tests, and urban setting. The review was limited to studies published between 2000 and 2016 in English, Spanish, French, and Portuguese. Inclusion and exclusion criteria were refined post hoc to identify relevant literature regardless of study design and geography. A total of 179 documents of the 7806 initially screened were included in the analysis. Malaria (n = 100) and tuberculosis (n = 47) accounted for the majority of studies that reported diagnostics performance, impact, and implementation outcomes. Fewer studies, assessing mainly performance, were identified for visceral leishmaniasis (n = 9), filariasis and leptospirosis (each n = 5), enteric fever and schistosomiasis (each n = 3), dengue and leprosy (each n = 2), and Chagas disease, human African trypanosomiasis, and cholera (each n = 1). Reported sensitivity of rapid tests was variable depending on several factors. Overall, specificities were high (> 80%), except for schistosomiasis and cholera. Impact and implementation outcomes, mainly acceptability and cost, followed by adoption, feasibility, and sustainability of rapid tests are being evaluated in the field. Challenges to implementing rapid tests range from cultural to technical and administrative issues.

CONCLUSIONS

Rapid diagnostic tests for vector-borne and other diseases of poverty are being used in the urban context with demonstrated impact on case detection. However, most evidence comes from malaria rapid diagnostics, with variable results. While rapid tests for tuberculosis and visceral leishmaniasis require further implementation studies, more evidence on performance of current tests or development of new alternatives is needed for dengue, Chagas disease, filariasis, leptospirosis, enteric fever, human African trypanosomiasis, schistosomiasis and cholera.

Evaluation of the Clinical Proficiency of RDTs, Microscopy and Nested PCR in the Diagnosis of Symptomatic Malaria in Ilorin, North-Central, Nigeria

BACKGROUND:

Accurate laboratory diagnosis of suspected malaria is the hallmark to the control of the disease.

AIM:

The clinical proficiency of commercial Rapid Diagnostic test kits (RDTs) using nested PCR as quality control was evaluated among patients attending two public healthcare providing institutions in Ilorin, Kwara state, North-Central, Nigeria.

METHOD:

A cross-sectional evaluation of finger prick blood samples of volunteer patients were accessed for malaria parasites with pLDH, HRP2, Pf, Pf/PAN and nested PCR molecular assays. The data derived were analysed using standard formulae for diagnostic accuracy, and the obtained predictive values were subjected to a comparison with one-way analysis of variance (ANOVA).

RESULT:

Three hundred and sixty-eight (368) patients comprising 203 (55%) females and 165 (45%) males participated in this study. Routine microscopy revealed that 54 (32.7%) males and 80 (39.4%) was infected with Plasmodium falciparum. SD Bioline (pLDH) 47.4%; Carestart Malaria (HRP2) 49.8% recorded low sensitivities. Micropoint (pfPAN) 82.8% and Micropoint (Mal. Pf) 64.4% recorded a high sensitivity. SD Bioline (pLDH) 67.4%; Carestart Malaria (HRP2) 85.9%; Micropoint (PfPAN) 62.2% and Micropoint (Mal. Pf) 86.7% had high specificities. The positive predictive value (PPV) ranged from 67.7% to 85.94%, while the negative predictive values (NPV) of 64.4% for SD Bioline (pLDH); 86.7% for Carestart Malaria (HRP2); 89.3% for Micropoint (pfPAN) and 58.5% for Micropoint (Mal. Pf). Agarose gel analysis of P. falciparumssrRNA gene (206 bp) for 28 specimens containing 10% concordant and discordant samples showed that all 12 negative specimens for RDTs and routine microscopy were truly negative for nPCR. However, the remaining 16 specimens were positive for nPCR and showed discrepancies with routine microscopy and RDTs. Cohen’s interrater diagnostic measure analysis revealed that the weighted kappa for the RDTs was moderate 0.417 (p=0.027), 95%CI (0.756, 0.078) and good for nPCR 0.720 (p < 0.001), 95%CI (0.963, 0.477). The area under the curve (AUC) specify that nPCR has been more effective than the RDTs (nPCRAUC = 0.875; p < 0.001 and RDTsAUC = 0.708; p = 0.063).

CONCLUSION:

A thorough large-scale quality control is advocated on all commercial RDTs being used in most sub-Saharan African countries. This is to avoid double jeopardy consequent upon misdiagnosis on unidentified positive cases serving as pool reservoir for the insect vector and cyclical infection and re-infection of the populace.