Since emerging in China in December 2019, the infection caused by SARS‐CoV‐2, COVID‐19, has infected more than 3 million people worldwide (Johns Hopkins University and Medicine: Coronavirus Resource Center). Since then, clusters of infections have emerged in Singapore, particularly among younger workers in communal living environments. The spectrum of COVID‐19 infection ranges from mild respiratory illness, to viral pneumonia and life‐threatening acute respiratory distress syndrome in up 15‐20% (Huang, et al 2020, Wu and McGoogan 2020).

Background: Morphologic rare cell detection is a laborious, operator-dependent process which has the potential to be improved by the use of image analysis using artificial intelligence. Detection of rare hemoglobin H (HbH) inclusions in red cells in the peripheral blood is a common screening method for alpha-thalassemia. This study aims to develop a convolutional neural network-based algorithm for the detection of HbH inclusions. Methods: Digital images of HbH-positive and HbH-negative blood smears were used to train and test the software. The software performance was tested on images obtained at various magnifications and on different scanning platforms. Another model was developed for total red cell counting and was used to confirm HbH cell frequency in alpha-thalassemia trait. The threshold minimum red cells to image for analysis was determined by Poisson modeling and validated on image sets. Results: The sensitivity and specificity of the software for HbH+ cells on images obtained at ×100, ×60, and ×40 objectives were close to 91% and 99%, respectively. When an AI-aided diagnostic model was tested on a pilot of 40 whole slide images (WSIs), good inter-rater reliability and high sensitivity and specificity of slide-level classification were obtained. Using the lowest frequency of HbH+ cells (1 in 100,000) observed in our study, we estimated that a minimum of 2.4 × 106 red cells would need to be analyzed to reduce misclassification at the slide level. The minimum required smear size was validated on 78 image sets which confirmed its validity. Conclusions: WSI image analysis can be utilized effectively for morphologic rare cell detection. The software can be further developed on WISs and evaluated in future clinical validation studies comparing AI-aided diagnosis with the routine diagnostic method.

Vaccine-induced immune thrombotic thrombocytopenia (VITT) is a serious and life-threatening complication occurring after adenovirus-vector COVID-19 vaccines, and is rarely reported after other vaccine types. Herein, we report a case of possible VITT after the Pfizer-BioNTech mRNA vaccine booster, who presented with extensive lower limb deep vein thrombosis, severe thrombocytopenia, markedly elevated D-dimer and positive anti-PF4 antibody occurring 2 weeks post-vaccination, concurrent with a lupus anticoagulant. A complete recovery was made after intravenous immunoglobulin, prednisolone and anticoagulation with the oral direct Xa inhibitor rivaroxaban. The presenting features of VITT may overlap with those of antiphospholipid syndrome associated with anti-PF4 and immune thrombocytopenia. We discuss the diagnostic considerations in VITT and highlight the challenges of performing VITT confirmatory assays in non-specialized settings. The set of five diagnostic criteria for VITT is a useful tool for guiding initial management, but may potentially include patients without VITT. The bleeding risks of severe thrombocytopenia in the face of thrombosis, requiring anticoagulant therapy, present a clinical challenge, but early recognition and management can potentially lead to favorable outcomes.

A high incidence of venous thromboembolism (VTE) in COVID-19 has led to international recommendations for thromboprophylaxis. With limited data on Asian patients with COVID-19, the role of thromboprophylaxis remains unclear. To investigate the in-hospital incidence of VTE in an Asian COVID-19 cohort, describe the VTE trend through successive COVID-19 waves (wild-type, delta, and omicron), and characterize the risk factors for VTE. We performed a retrospective observational cohort study of hospitalized COVID-19 adults from January 2020 to February 2022. Objectively confirmed VTE were reviewed to obtain VTE incidence and trend. Subset analysis of critical (intensive care), moderate (oxygen supplementation), and mild cases hospitalized ≥5 days was performed to investigate risk factors and in-hospital hazards of VTE. Sixteen VTE events occurred among 3574 patients. Overall, VTE incidence was 0.45%, or 0.21% in mild, 3.60% in moderate, and 5.38% in critical infection. The maximum cumulative risk of VTE was 1.14% at 14 days for mild, 8.13% at 21 days for moderate, and 11.55% at 35 days for critical infection. Thromboprophylaxis use in mild, moderate, and critical cases was 5.7%, 28.8%, and 81.7%, respectively. In multivariable analysis, the severity of infection remained the strongest independent predictor of VTE. Compared with mild infection, the relative risk was 8.26 (95% CI, 2.26-30.16) for critical infection and 6.29 (95% CI, 1.54-25.67) for moderate infection. Overall, VTE incidence in Asian patients with COVID-19 is <1% across successive waves. Patients with moderate and critical infections are at greater risk for VTE and should be considered for routine thromboprophylaxis.

Objective: Enzyme-linked immunosorbent assay (ELISA) is the established method for detecting antiphospholipid antibodies (aPL) in the diagnosis of antiphospholipid syndrome (APS) but is labor-intensive compared with the newer automated chemiluminescent assay (CLIA). This study aims to evaluate CLIA versus ELISA for aPL, correlate each method with clinical manifestations and perform a comprehensive literature review. Methods: Patient samples were concurrently tested by ELISA (QUANTA Lite ^® ) and CLIA (ACL AcuStar ^® ) for anti-cardiolipin antibody (aCL) and anti-β2-glycoprotein-I (aβ2GPI) IgG and IgM. Assay results were correlated with any of the revised Sapporo APS clinical criteria. Results: Of the 107 patients, 67% fulfilled at least one clinical criterion. 38 patients (35.5%) had APS. For aCL IgG, aCL IgM and aβ2GPI IgM, CLIA showed above 77% concordance and fair to excellent agreement (Cohen's kappa 0.39–0.86) with moderate/high positive ELISA of ≥40 units. Both methods showed good correlation (Spearman's r 0.60–0.80, p  < 0.0001) that was non-linear over the range of titers. CLIA sensitivity and specificity was 46%–100% and 68%–95%, with AUROC ranging from 0.80–0.93. For aβ2GPI IgG, concordance was 36.7% and agreement was low (kappa −0.23). Correlation with clinical criteria revealed no statistically significant difference in the occurrence of clinical manifestations in ELISA-positive versus CLIA-positive groups. Conclusions: aPL detection by CLIA showed close but incomplete concordance with ELISA. CLIA positivity correlated well with moderate/high ELISA positivity, but antibody titers should not be directly compared across systems. CLIA is an acceptable alternative to ELISA in the routine non-research setting. Our findings are congruent with the reviewed literature.