Immune-mediated thrombocytopenia and IL-6-mediated thrombocytosis observed in idiopathic multicentric Castleman disease.

Idiopathic multicentric Castleman disease (iMCD) is a rare haematological disorder characterized by generalized lymphadenopathy with atypical histopathological features and systemic inflammation caused by a cytokine storm involving interleukin-6 (IL-6). Three clinical subtypes are recognized: thrombocytopenia, anasarca, fever, renal dysfunction, organomegaly (iMCD-TAFRO); idiopathic plasmacytic lymphadenopathy (iMCD-IPL), involving thrombocytosis and hypergammaglobulinaemia; and iMCD-not otherwise specified (iMCD-NOS), which includes patients who do not meet criteria for the other subtypes. Disease pathogenesis is poorly understood, with potential involvement of infectious, clonal and/or autoimmune mechanisms. To better characterize iMCD clinicopathology and gain mechanistic insights into iMCD, we analysed complete blood counts, other clinical laboratory values and blood smear morphology among 63 iMCD patients grouped by clinical subtype. Patients with iMCD-TAFRO had large platelets, clinical severity associated with lower platelet counts and transfusion-resistant thrombocytopenia, similar to what is observed with immune-mediated destruction of platelets in immune thrombocytopenic purpura. Conversely, elevated platelet counts in iMCD-IPL were associated with elevated IL-6 and declined following anti-IL-6 therapy. Our data suggest that autoimmune mechanisms contribute to the thrombocytopenia in at least a portion of iMCD-TAFRO patients whereas IL-6 drives thrombocytosis in iMCD-IPL, and these mechanisms likely contribute to disease pathogenesis.

Homozygous SPTA1-associated hereditary pyropoikilocytosis presenting as hydrops fetalis.

INTRODUCTION: Hereditary pyropoikilocytosis (HPP) is a heterogeneous inherited disorder of red blood cell (RBC) membrane and cytoskeletal proteins that leads to hemolytic anemia. HPP is characterized by marked poikilocytosis, microspherocytes, RBC fragmentation, and elliptocytes on peripheral blood smear. Mutations in SPTA1 can cause HPP due to a quantitative defect in α-spectrin and can lead to profound fetal anemia and nonimmune hydrops fetalis, which can be managed with intrauterine transfusion. CASE PRESENTATION: We present a case of a 26-year-old G4P2102 woman of Amish-Mennonite ancestry with a pregnancy complicated by fetal homozygosity for an SPTA1 gene variant (SPTA1c.6154delG) as well as severe fetal anemia and hydrops fetalis, which was managed with four intrauterine transfusions between 26 and 30 weeks gestation. Pre-transfusion peripheral smears from fetal blood samples showed RBC morphology consistent with HPP. The neonate had severe hyperbilirubinemia at birth, which has resolved, but remains transfusion-dependent at 6 months of life. DISCUSSION/CONCLUSION: To our knowledge, this is the first report that correlates homozygosity of the SPTA1c.6154delG gene variant with RBC dysmorphology and establishes the diagnosis of HPP.

Classification of peripheral blood neutrophils using deep learning.

Deep learning has been used to classify the while blood cells in peripheral blood smears. However, the classification of developing neutrophils is rarely studied. Moreover, it is still unknown whether deep learning can work well on the data coming from different sources. In this study, we therefore investigate the classification performance of deep learning for immature and mature neutrophils. In particular, we used three open-access datasets obtained from different imaging systems: CellaVision DM 96, CellaVision DM 100, and iCELL ME-150. A total of 26,050 images identified by one laboratory technologist were randomly split into training, validation, and testing datasets. A total of 10 convolutional neural networks were trained to classify six blood cell types: myeloblast, promyelocyte, myelocyte, metamyelocyte, banded neutrophil, and segmented neutrophil. The experimental results showed that compared to any single model, the average ensemble model could achieve a better classification performance and provide a testing accuracy of 90.1%. The sensitivity and specificity of the average ensemble model for the six blood cell types were above 83.5% and 96.9%, respectively. Our results suggest that deep learning is a promising tool for the classification of developing neutrophils, but further improvement is required.

Fulminant Streptococcus suis infection detected on peripheral blood smear: A case report.

Streptococcus suis, a gram-positive coccus, is recognized as an emerging zoonotic pathogen that causes serious infections in humans, such as bacterial meningitis and sepsis, with poor outcomes. The pathogen is known to be transmitted through the consumption of raw pork or occupational exposure to pigs. A previously healthy 38-year-old woman with occupational exposure to raw pork was presented to our emergency department with a clinical diagnosis of rapidly progressive septic shock. Peripheral blood smears detected chains of cocci inside granulocytes, which led to the early recognition of gram-positive cocci in short chains before the blood culture test results. Blood cultures later tested positive for S. suis serotype 2. The patient's condition deteriorated despite aggressive resuscitative measures including antibiotics, vasopressors, multiple blood transfusions, mechanical ventilation, and renal replacement therapy. Initiation of veno-arterial extracorporeal membrane oxygenation was ineffective, and the patient died 16 h after admission. The identification of bacteria in the peripheral blood smear indicated an overwhelming infection and led to the rapid recognition of bacteremia. Our report aims to raise awareness about fatal zoonotic pathogens and to promote the unique role of peripheral blood smears that could provide preliminary diagnostic information before blood culture results.

Artificial intelligence of digital morphology analyzers improves the efficiency of manual leukocyte differentiation of peripheral blood.

BACKGROUND AND OBJECTIVE: Morphological identification of peripheral leukocytes is a complex and time-consuming task, having especially high requirements for personnel expertise. This study is to investigate the role of artificial intelligence (AI) in assisting the manual leukocyte differentiation of peripheral blood. METHODS: A total of 102 blood samples that triggered the review rules of hematology analyzers were enrolled. The peripheral blood smears were prepared and analyzed by Mindray MC-100i digital morphology analyzers. Two hundreds leukocytes were located and their cell images were collected. Two senior technologists labeled all cells to form standard answers. Afterward, the digital morphology analyzer unitized AI to pre-classify all cells. Ten junior and intermediate technologists were selected to review the cells with the AI pre-classification, yielding the AI-assisted classifications. Then the cell images were shuffled and re-classified without AI. The accuracy, sensitivity and specificity of the leukocyte differentiation with or without AI assistance were analyzed and compared. The time required for classification by each person was recorded. RESULTS: For junior technologists, the accuracy of normal and abnormal leukocyte differentiation increased by 4.79% and 15.16% with the assistance of AI. And for intermediate technologists, the accuracy increased by 7.40% and 14.54% for normal and abnormal leukocyte differentiation, respectively. The sensitivity and specificity also significantly increased with the help of AI. In addition, the average time for each individual to classify each blood smear was shortened by 215 s with AI. CONCLUSION: AI can assist laboratory technologists in the morphological differentiation of leukocytes. In particular, it can improve the sensitivity of abnormal leukocyte differentiation and lower the risk of missing detection of abnormal WBCs.

CLINICAL AND EPIDEMIOLOGICAL FEATURES OF MENINGOCOCCAL INFECTION AND ITS EARLY DIAGNOSIS IN RESIDENTS OF THE TRANSCARPATHIAN REGION.

OBJECTIVE: The aim: To identify clinical and epidemiological features of meningococcal infection on the initial day of a patient's medical consultation, as well as the efficacy of laboratory examinations. PATIENTS AND METHODS: Materials and methods: A retrospective analysis of 76 patients' histories diagnosed with meningococcal disease was carried out. CONCLUSION: Conclusions: Patients in the Transcarpathian region mainly develop an atypical form of meningococcal disease. Only half of all patients diagnosed with meningococcemia had a classical hemorrhagic rash. Generalized forms of meningococcal disease may proceed with normal or subfebrile temperature and without severe leukocytosis. We doubt the use of bacteriological methods of laboratory diagnosis due to their low effectiveness. The most sensitive method of laboratory diagnosis is a microscopic examination of blood smear, and cerebrospinal fluid.

Comparison of hematologic abnormalities between hospitalized coronavirus disease 2019 positive and negative patients with correlation to disease severity and outcome.

Peripheral blood smear (PBS) changes in coronavirus disease 2019 (COVID-19) are diverse and have been reported in the literature in the form of case series with relatively smaller sample sizes and with a handful of studies showing the association between PBS and clinical severity. This study aims to highlight the numerical and morphological changes in peripheral blood of COVID-19 patients and to compare the same in intensive care unit (ICU) and non-ICU settings as well as with disease severity and outcome. The study included 80 COVID-19 positive (41 ICU and 39 non-ICU) patients and 32 COVID-19 negative ICU patients. Complete blood counts (CBCs) and PBS findings were studied and scored by two pathologists blindfolded. Absolute lymphocyte count (ALC) and absolute eosinophil count (AEC) were significantly lower in COVID-19 positive cases as compared to the COVID-19 negative group (p = 0.001 and p = 0.001). COVID-19 positive group showed significant left myeloid shift (p = 0.021), Dohle bodies (p = 0.025) with significant prominence of acquired pseudo-Pelger-Huët anomaly, ring-shaped neutrophils, monolobate neutrophils, and plasmacytoid lymphocytes as compared to control group (p = 0.000, p = 0.009, p = 0.046, and p = 0.011, respectively). The overall mean white blood cell (WBC) counts were higher in COVID-19 positive ICU patients as compared to non-ICU COVID patients with significant shift to left, presence of ring-shaped neutrophils, monocyte vacuolation, and large granular lymphocytes (p = 0.017, p = 0.007, p = 0.008, and p = 0.004, respectively). Deceased group showed significantly higher WBC count (p = 0.018) with marked neutrophilia (p = 0.024) and toxic granulation (p = 0.01) with prominence of monocyte vacuolization, ring-shaped neutrophils, large granular lymphocytes, and reactive lymphocytes. Parameters like myeloid left shift, ring-shaped neutrophils, monocyte vacuolation, and large granular lymphocytes emerged as highly sensitive markers of disease severity. Therefore, serial CBC with comprehensive PBS analysis should be done in every newly diagnosed hospitalized COVID-19 patient which potentially predicts the course of the disease.

Asymptomatic malaria infection, associated factors and accuracy of diagnostic tests in a historically high transmission setting in Northern Uganda.

BACKGROUND: Asymptomatic malaria infections are important parasite reservoirs and could sustain transmission in the population, but they are often unreported. A community-based survey was conducted to investigate the prevalence and factors associated with asymptomatic malaria infections in a historically high transmission setting in northern Uganda. METHODS: Using a cross-sectional design, 288 children aged 2-15 years were enrolled and tested for the presence of malaria parasites using rapid diagnostic tests (RDTs) and blood smear microscopy between January to May 2022. Statistical analysis was performed using the exact binomial and Fisher's exact test with p ≤ 0.05 indicating significance. The logistic regression was used to explore factors associated with asymptomatic malaria infections. RESULTS: Overall, the prevalence of asymptomatic infection was 34.7% (95% CI 29.2-40.5) with the highest observed in children 5-10 years 45.9% (95% CI 35.0-57.0). Gweri village accounted for 39.1% (95% CI 27.6-51.6) of malaria infections. Median parasite density was 1500 parasites/µl of blood. Plasmodium falciparum was the dominant species (86%) followed by Plasmodium malariae (5%). Factors associated with asymptomatic malaria infection were sleeping under mosquito net (Adjusted Odds Ratio (aOR) 0.27; 95% CI 0.13-0.56), p = 0.001 and presence of village health teams (VHTs) (aOR 0.02; 95% CI 0.01-0.45), p = 0.001. Sensitivity and specificity were higher for the P. falciparum/pLDH RDTs compared to HRP2-only RDTs, 90% (95% CI 86.5-93.5) and 95.2% (95% CI 92.8-97.7), p = 0.001, respectively. CONCLUSION: Asymptomatic malaria infections were present in the study population and this varied with place and person in the different age groups. Plasmodium falciparum was the dominant parasite species however the presence of P. malariae and Plasmodium ovale was observed, which may have implication for the choice and deployment of diagnostic tools. Individuals who slept under mosquito net or had presence of functional VHTs were less likely to have asymptomatic malaria infection. P.f/pLDH RDTs performed better than the routinely used HRP2 RDTs. In view of these findings, investigation and reporting of asymptomatic malaria reservoirs through community surveys is recommended for accurate disease burden estimate and better targeting of control.

Health care provider practices in diagnosis and treatment of malaria in rural communities in Kisumu County, Kenya.

BACKGROUND: Accurate malaria diagnosis and appropriate treatment at local health facilities are critical to reducing morbidity and human reservoir of infectious gametocytes. The current study assessed the accuracy of malaria diagnosis and treatment practices in three health care facilities in rural western Kenya. METHODS: The accuracy of malaria detection and treatment recommended compliance was monitored in two public and one private hospital from November 2019 through March 2020. Blood smears from febrile patients were examined by hospital laboratory technicians and re-examined by an expert microscopists thereafter subjected to real-time polymerase chain reaction (RT-PCR) for quality assurance. In addition, blood smears from patients diagnosed with malaria rapid diagnostic tests (RDT) and presumptively treated with anti-malarial were re-examined by an expert microscopist. RESULTS: A total of 1131 febrile outpatients were assessed for slide positivity (936), RDT (126) and presumptive diagnosis (69). The overall positivity rate for Plasmodium falciparum was 28% (257/936). The odds of slide positivity was higher in public hospitals, 30% (186/624, OR:1.44, 95% CI = 1.05-1.98, p < 0.05) than the private hospital 23% (71/312, OR:0.69, 95% CI = 0.51-0.95, p < 0.05). Anti-malarial treatment was dispensed more at public hospitals (95.2%, 177/186) than the private hospital (78.9%, 56/71, p < 0.0001). Inappropriate anti-malarial treatment, i.e. artemether-lumefantrine given to blood smear negative patients was higher at public hospitals (14.6%, 64/438) than the private hospital (7.1%, 17/241) (p = 0.004). RDT was the most sensitive (73.8%, 95% CI = 39.5-57.4) and specific (89.2%, 95% CI = 78.5-95.2) followed by hospital microscopy (sensitivity 47.6%, 95% CI = 38.2-57.1) and specificity (86.7%, 95% CI = 80.8-91.0). Presumptive diagnosis had the lowest sensitivity (25.7%, 95% CI = 13.1-43.6) and specificity (75.0%, 95% CI = 50.6-90.4). RDT had the highest non-treatment of negatives [98.3% (57/58)] while hospital microscopy had the lowest [77.3% (116/150)]. Health facilities misdiagnosis was at 27.9% (77/276). PCR confirmed 5.2% (4/23) of the 77 misdiagnosed cases as false positive and 68.5% (37/54) as false negative. CONCLUSIONS: The disparity in malaria diagnosis at health facilities with many slide positives reported as negatives and high presumptive treatment of slide negative cases, necessitates augmenting microscopic with RDTs and calls for Ministry of Health strengthening supportive infrastructure to be in compliance with treatment guidelines of Test, Treat, and Track to improve malaria case management.

Diagnostic performance and comparison of ultrasensitive and conventional rapid diagnostic test, thick blood smear and quantitative PCR for detection of low-density Plasmodium falciparum infections during a controlled human malaria infection study in Equatorial Guinea.

BACKGROUND: Progress towards malaria elimination has stagnated, partly because infections persisting at low parasite densities comprise a large reservoir contributing to ongoing malaria transmission and are difficult to detect. This study compared the performance of an ultrasensitive rapid diagnostic test (uRDT) designed to detect low density infections to a conventional RDT (cRDT), expert microscopy using Giemsa-stained thick blood smears (TBS), and quantitative polymerase chain reaction (qPCR) during a controlled human malaria infection (CHMI) study conducted in malaria exposed adults (NCT03590340). METHODS: Blood samples were collected from healthy Equatoguineans aged 18-35 years beginning on day 8 after CHMI with 3.2 × 103 cryopreserved, infectious Plasmodium falciparum sporozoites (PfSPZ Challenge, strain NF54) administered by direct venous inoculation. qPCR (18s ribosomal DNA), uRDT (Alere™ Malaria Ag P.f.), cRDT [Carestart Malaria Pf/PAN (PfHRP2/pLDH)], and TBS were performed daily until the volunteer became TBS positive and treatment was administered. qPCR was the reference for the presence of Plasmodium falciparum parasites. RESULTS: 279 samples were collected from 24 participants; 123 were positive by qPCR. TBS detected 24/123 (19.5% sensitivity [95% CI 13.1-27.8%]), uRDT 21/123 (17.1% sensitivity [95% CI 11.1-25.1%]), cRDT 10/123 (8.1% sensitivity [95% CI 4.2-14.8%]); all were 100% specific and did not detect any positive samples not detected by qPCR. TBS and uRDT were more sensitive than cRDT (TBS vs. cRDT p = 0.015; uRDT vs. cRDT p = 0.053), detecting parasitaemias as low as 3.7 parasites/µL (p/µL) (TBS and uRDT) compared to 5.6 p/µL (cRDT) based on TBS density measurements. TBS, uRDT and cRDT did not detect any of the 70/123 samples positive by qPCR below 5.86 p/µL, the qPCR density corresponding to 3.7 p/µL by TBS. The median prepatent periods in days (ranges) were 14.5 (10-20), 18.0 (15-28), 18.0 (15-20) and 18.0 (16-24) for qPCR, TBS, uRDT and cRDT, respectively; qPCR detected parasitaemia significantly earlier (3.5 days) than the other tests. CONCLUSIONS: TBS and uRDT had similar sensitivities, both were more sensitive than cRDT, and neither matched qPCR for detecting low density parasitaemia. uRDT could be considered an alternative to TBS in selected applications, such as CHMI or field diagnosis, where qualitative, dichotomous results for malaria infection might be sufficient.

Analysis of malaria clinical-epidemiological predictors in individuals from Brazilian Amazon.

This research aims to determine whether the combination of epidemiological and clinical features can predict malaria. Diagnostic investigation detected 22.3% of individuals with Plasmodium vivax (P. vivax) malaria, with significant predominance of the male gender. The malaria triad (fever, chills and headache) had a more expressive frequency (81.1%) in individuals with positive thick blood than those with negative thick blood smear (65.1%), although there was no statistical significance. Among the variables analysed as predictive for positive thick blood smear, it was observed that personal history of travel to an endemic malaria area and past malaria infection (PMI) were significantly associated with malaria, even in multiple logistic regression. Fever had the higher sensitivity (94.6%) and past malaria history had the greater specificity (68.2%), with accuracy of 23.5% and 67.5%, respectively. In combined analysis, fever with chills had the highest sensitivity (91.9%), but low accuracy (38.5%). High specificity (91.5%) was found in the association of malaria triad, PMI and history of travel to endemic malaria area (which along with anorexia, was higher 94.6%), with good accuracy (80.7%), suggesting that the screening of patients for performing thick blood smear can be based on these data. The epidemiological features and the malaria triad (fever, chills and headache) can be predictors for identification of malaria patients, concurring to precocious diagnosis and immediate treatment of individuals with malaria.

Difficulties in diagnosing Malassezia furfur bloodstream infection and possibility of spontaneous resolution in a patient undergoing chemotherapy for neuroblastoma: A case report.

Malassezia furfur is a lipophilic, yeast-like fungus that forms part of the normal human skin microflora and is associated with a wide range of infections, such as pityriasis versicolor, folliculitis, and systemic infections in immunocompromised patients. Although matrix-assisted laser desorption/ionization time-of-flight mass spectrometry has enabled rapid identification of Malassezia species, it is still a challenge to diagnose systemic infections because Malassezia fungemia can often be missed by automated blood culture systems. We report a case in which M. furfur in blood was detected by the presence of yeast-like fungi in blood smears. Yeast-like organisms were observed in the blood smears of a 3-year-old boy, taken over 2 weeks without any symptoms. He had undergone several courses of chemotherapy for neuroblastoma via an indwelling central venous catheter (CVC) that was placed in his right anterior chest for 11 months. Although the blood cultures obtained from an automated blood culture system were negative, M. furfur growth was detected in the subcultured blood taken from the CVC. The CVC was removed, and the scheduled chemotherapy was postponed. No systemic M. furfur bloodstream infection occurred; the infection resolved spontaneously without any specific treatment; only prophylactic fluconazole was administered. M. furfur fungemia may not be diagnosable by an automated blood culture system. Further, M. furfur may not cause infections in humans even when administered intravenously. This report may lead to the discovery of factors related to human infectivity of this disease in the future.

Laboratory diagnosis of community-acquired bloodstream infection in therapeutic pathology.

Community-acquired bloodstream infections (CBSIs) occur in the out-of-hospital setting (44%) and increase the overall mortality from bloodstream infections (BSIs) by 7.2% per year. The development of CBSIs depends on both comorbid and polymorbid diseases and the patients' age. The causes of CBSIs are: respiratory, hepatobiliary gastrointestinal and urogenital tracts and dental interventions. The etiology of CBSIs is characterized by the isolation of coagulase-negative staphylococci (CNS) (32%), E. coli (27%). To investigate community-acquired bloodstream infection in therapeutic patients. The study included out-of-hospital patients (n=382). 4.5 ml of blood were taken intravenously into a closed vacuum system in order to obtain a buffy coat of blood, which was put on glasses for microscopy and Petri dishes with blood agar for cultivating under aerobic and anaerobic conditions. Microorganisms were identified by mass spectrometry. Microscopy of blood smears was used for rapid diagnosis of infection in the bloodstream. BSI was diagnosed in 183 (48.0%) out of 382 out-of-hospital patients. The etiology of CBSIs was studied on 297 isolated strains of microorganisms. CBSIs rather often complicated the underlying disease in women and young people. The spectrum of CBSI pathogens included aerobic and anaerobic bacteria and fungi. Gram-positive cocci with the leadership of S.epidermidis (25.7%) were more often isolated among bacteria. 70% of all isolated pathogens grew under anaerobic conditions. CBSIs were characterized by polymicrobiality (33.5%) of two to four different microorganisms in one blood culture; the species of associates of polymicrobial blood cultures are shown. Microscopic examination of blood smears revealed microorganisms in 97.1% of cases, including associations of bacteria with fungi (66.9%). CBSIs occurred after contour plastic, in diseases of the respiratory system, genitourinary system, oral cavity, skin and subcutaneous tissue. Microbiological examination of the buffy coat is an alternative microbiological method of CBSIs diagnosis, which includes microscopy and blood cultivating and has a high diagnostic efficiency (97.1% and 48% respectively). It can become an option for replacing imported blood culture automated systems.

Malaria parasite detection in thick blood smear microscopic images using modified YOLOV3 and YOLOV4 models.

BACKGROUND: Manual microscopic examination of Leishman/Giemsa stained thin and thick blood smear is still the "gold standard" for malaria diagnosis. One of the drawbacks of this method is that its accuracy, consistency, and diagnosis speed depend on microscopists' diagnostic and technical skills. It is difficult to get highly skilled microscopists in remote areas of developing countries. To alleviate this problem, in this paper, we propose to investigate state-of-the-art one-stage and two-stage object detection algorithms for automated malaria parasite screening from microscopic image of thick blood slides. RESULTS: YOLOV3 and YOLOV4 models, which are state-of-the-art object detectors in accuracy and speed, are not optimized for detecting small objects such as malaria parasites in microscopic images. We modify these models by increasing feature scale and adding more detection layers to enhance their capability of detecting small objects without notably decreasing detection speed. We propose one modified YOLOV4 model, called YOLOV4-MOD and two modified models of YOLOV3, which are called YOLOV3-MOD1 and YOLOV3-MOD2. Besides, new anchor box sizes are generated using K-means clustering algorithm to exploit the potential of these models in small object detection. The performance of the modified YOLOV3 and YOLOV4 models were evaluated on a publicly available malaria dataset. These models have achieved state-of-the-art accuracy by exceeding performance of their original versions, Faster R-CNN, and SSD in terms of mean average precision (mAP), recall, precision, F1 score, and average IOU. YOLOV4-MOD has achieved the best detection accuracy among all the other models with a mAP of 96.32%. YOLOV3-MOD2 and YOLOV3-MOD1 have achieved mAP of 96.14% and 95.46%, respectively. CONCLUSIONS: The experimental results of this study demonstrate that performance of modified YOLOV3 and YOLOV4 models are highly promising for detecting malaria parasites from images captured by a smartphone camera over the microscope eyepiece. The proposed system is suitable for deployment in low-resource setting areas.

Applications of Digital Microscopy and Densely Connected Convolutional Neural Networks for Automated Quantification of Babesia-Infected Erythrocytes.

BACKGROUND: Clinical babesiosis is diagnosed, and parasite burden is determined, by microscopic inspection of a thick or thin Giemsa-stained peripheral blood smear. However, quantitative analysis by manual microscopy is subject to error. As such, methods for the automated measurement of percent parasitemia in digital microscopic images of peripheral blood smears could improve clinical accuracy, relative to the predicate method. METHODS: Individual erythrocyte images were manually labeled as "parasite" or "normal" and were used to train a model for binary image classification. The best model was then used to calculate percent parasitemia from a clinical validation dataset, and values were compared to a clinical reference value. Lastly, model interpretability was examined using an integrated gradient to identify pixels most likely to influence classification decisions. RESULTS: The precision and recall of the model during development testing were 0.92 and 1.00, respectively. In clinical validation, the model returned increasing positive signal with increasing mean reference value. However, there were 2 highly erroneous false positive values returned by the model. Further, the model incorrectly assessed 3 cases well above the clinical threshold of 10%. The integrated gradient suggested potential sources of false positives including rouleaux formations, cell boundaries, and precipitate as deterministic factors in negative erythrocyte images. CONCLUSIONS: While the model demonstrated highly accurate single cell classification and correctly assessed most slides, several false positives were highly incorrect. This project highlights the need for integrated testing of machine learning-based models, even when models in the development phase perform well.

Hyper-prevalence of submicroscopic Plasmodium falciparum infections in a rural area of western Kenya with declining malaria cases.

BACKGROUND: The gold standard for diagnosing Plasmodium falciparum infection is microscopic examination of Giemsa-stained peripheral blood smears. The effectiveness of this procedure for infection surveillance and malaria control may be limited by a relatively high parasitaemia detection threshold. Persons with microscopically undetectable infections may go untreated, contributing to ongoing transmission to mosquito vectors. The purpose of this study was to determine the magnitude and determinants of undiagnosed submicroscopic P. falciparum infections in a rural area of western Kenya. METHODS: A health facility-based survey was conducted, and 367 patients seeking treatment for symptoms consistent with uncomplicated malaria in Homa Bay County were enrolled. The frequency of submicroscopic P. falciparum infection was measured by comparing the prevalence of infection based on light microscopic inspection of thick blood smears versus real-time polymerase chain reaction (RT-PCR) targeting P. falciparum 18S rRNA gene. Long-lasting insecticidal net (LLIN) use, participation in nocturnal outdoor activities, and gender were considered as potential determinants of submicroscopic infections. RESULTS: Microscopic inspection of blood smears was positive for asexual P. falciparum parasites in 14.7% (54/367) of cases. All of these samples were confirmed by RT-PCR. 35.8% (112/313) of blood smear negative cases were positive by RT-PCR, i.e., submicroscopic infection, resulting in an overall prevalence by RT-PCR alone of 45.2% compared to 14.7% for blood smear alone. Females had a higher prevalence of submicroscopic infections (35.6% or 72 out of 202 individuals, 95% CI 28.9-42.3) compared to males (24.2%, 40 of 165 individuals, 95% CI 17.6-30.8). The risk of submicroscopic infections in LLIN users was about half that of non-LLIN users (OR = 0.59). There was no difference in the prevalence of submicroscopic infections of study participants who were active in nocturnal outdoor activities versus those who were not active (OR = 0.91). Patients who participated in nocturnal outdoor activities and use LLINs while indoors had a slightly higher risk of submicroscopic infection than those who did not use LLINs (OR = 1.48). CONCLUSION: Microscopic inspection of blood smears from persons with malaria symptoms for asexual stage P. falciparum should be supplemented by more sensitive diagnostic tests in order to reduce ongoing transmission of P. falciparum parasites to local mosquito vectors.

Diagnosis of Leukaemia in Blood Slides Based on a Fine-Tuned and Highly Generalisable Deep Learning Model.

Leukaemia is a dysfunction that affects the production of white blood cells in the bone marrow. Young cells are abnormally produced, replacing normal blood cells. Consequently, the person suffers problems in transporting oxygen and in fighting infections. This article proposes a convolutional neural network (CNN) named LeukNet that was inspired on convolutional blocks of VGG-16, but with smaller dense layers. To define the LeukNet parameters, we evaluated different CNNs models and fine-tuning methods using 18 image datasets, with different resolution, contrast, colour and texture characteristics. We applied data augmentation operations to expand the training dataset, and the 5-fold cross-validation led to an accuracy of 98.61%. To evaluate the CNNs generalisation ability, we applied a cross-dataset validation technique. The obtained accuracies using cross-dataset experiments on three datasets were 97.04, 82.46 and 70.24%, which overcome the accuracies obtained by current state-of-the-art methods. We conclude that using the most common and deepest CNNs may not be the best choice for applications where the images to be classified differ from those used in pre-training. Additionally, the adopted cross-dataset validation approach proved to be an excellent choice to evaluate the generalisation capability of a model, as it considers the model performance on unseen data, which is paramount for CAD systems.

MALDI TOF Mass Spectrometry Imaging of Blood Smear: Method Development and Evaluation.

The aim of this study was to develop and evaluate matrix assisted LASER desorption ionization (MALDI) time-of-flight (TOF) mass spectrometry imaging (MSI) of blood smear. Integrated light microscope and MALDI IT-TOF mass spectrometer, together with a matrix sublimation device, were used for analysis of blood smears coming from healthy male donors. Different blood plasma removal, matrix deposition, and instrumental settings were evaluated using the negative and positive ionization modes while agreement between the light microscopy images and the lateral distributions of cellular marker compounds served as the MSI quality indicator. Red and white blood cells chemical composition was analyzed using the differential m/z expression. Five seconds of exposure to ethanol followed by the 5 min of 9-aminoacridine or α-cyano-4-hydroxycinnamic acid deposition, together with two sets of instrumental settings, were selected for the MALDI TOF MSI experiments. Application of the thin and transparent matrix layers assured good correspondence between the LASER footprints and the preselected regions of interest. Cellular marker m/z signals coincided well with the appropriate cells. A metabolite databases search using the differentially expressed m/z produced hits which were consistent with the respective cell types. This study sets the foundations for application of blood smear MALDI TOF MSI in clinical diagnostics and research.

Case of Babesia crassa-Like Infection, Slovenia, 2014.

We report a case of Babesia crassa-like infection in an asplenic patient in Slovenia in 2014. We diagnosed the infection using microscopy, 18S rRNA sequencing, and serology and monitored parasitemia using digital PCR. With its increasing occurrence, babesiosis should be included in differential diagnoses for immunocompromised patients displaying fever.

Cost-effectiveness of rapid diagnostic tests, compared to microscopic tests, for the diagnosis and treatment of gestational malaria in Colombia from an institutional perspective.

BACKGROUND: Gestational malaria is associated with negative outcomes in maternal and gestational health; timely diagnosis is crucial to avoid complications. However, the limited infrastructure, equipment, test reagents, and trained staff make it difficult to use thick blood smear tests in rural areas, where rapid testing could be a viable alternative. The purpose of this study was to estimate the cost-effectiveness of rapid tests type III (Plasmodium falciparum/Plasmodium spp P.f/pan) versus microscopic tests for the diagnosis and treatment of gestational malaria in Colombia. METHODS: Cost-effectiveness analyses of gestational malaria diagnosis from an institutional perspective using a decision tree. Standard costing was performed for the identification, measurement and assessment phases, with data from Colombian tariff manuals. The data was collected from Health Situation Analysis, SIVIGILA and meta-analysis. Average and incremental cost-effectiveness ratio were estimated. The uncertainty was assessed through probabilistic sensitivity analysis. RESULTS: The cost of rapid diagnostic tests in 3,000 pregnant women with malaria was US$66,936 and 1,182 disability adjusted life years (DALYs) were estimated. The cost using thick blood smear tests was US$50,838 and 1,023 DALYs, for an incremental cost-effectiveness of US$ 101.2. The probabilistic sensitivity analysis of rapid diagnostic tests determined that they are highly cost-effective in 70% of the cases, even below the US$1,200 threshold; also, they showed an incremental net monetary benefit of $150,000 when payer's willingness is US$1,000. CONCLUSION: The use of rapid diagnostic tests for timely diagnosis and treatment of gestational malaria is a highly cost-effective strategy in Colombia, with uncertainty analyses supporting the robustness of this conclusion and the increased net monetary benefit that the health system would obtain. This strategy may help in preventing the negative effects on maternal health and the neonate at a low cost.