From microscope to micropixels: A rapid review of artificial intelligence for the peripheral blood film.

Artificial intelligence (AI) and its application in classification of blood cells in the peripheral blood film is an evolving field in haematology. We performed a rapid review of the literature on AI and peripheral blood films, evaluating the condition studied, image datasets, machine learning models, training set size, testing set size and accuracy. A total of 283 studies were identified, encompassing 6 broad domains: malaria (n = 95), leukemia (n = 81), leukocytes (n = 72), mixed (n = 25), erythrocytes (n = 15) or Myelodysplastic syndrome (MDS) (n = 1). These publications have demonstrated high self-reported mean accuracy rates across various studies (95.5% for malaria, 96.0% for leukemia, 94.4% for leukocytes, 95.2% for mixed studies and 91.2% for erythrocytes), with an overall mean accuracy of 95.1%. Despite the high accuracy, the challenges toward real world translational usage of these AI trained models include the need for well-validated multicentre data, data standardisation, and studies on less common cell types and non-malarial blood-borne parasites.

Comparison of Different Laboratory Tests to Identify "Aspirin Resistance" and Risk of Vascular Events among Ischaemic Stroke Patients: A Double-Blind Study.

"Aspirin resistance" (AR) is associated with increased risk of vascular events. We aimed to compare different platelet function tests used in identifying AR and assess their implications on clinical outcome. We performed platelet aggregation studies on non-cardioembolic ischaemic stroke patients taking aspirin 100 mg/day and 30 non-stroke controls. Data were collected on demographics, vascular risk factors, and concomitant medications. Cut-offs for AR were (1) light transmission aggregometry (LTA) of ≥20% using arachidonic acid (AA), ≥70% using ADP, or ≥60% using collagen; and (2) VerifyNow® assay ≥ 550 ARU. Telephone follow-ups were conducted by study staff blinded to AR status to ascertain the occurrence of vascular outcomes (stroke, myocardial infarction, amputation, death). A total of 113 patients were recruited, mean age 65 ± 8 years, 47% women, 45 ± 15 days from index stroke. 50 (44.3%, 95% CI 34.9-53.9) had AR on at least 1 test. Frequency of AR varied from 0% to 39% depending on method used and first vs. recurrent stroke. There were strong correlations between LTA AA, VerifyNow® and Multiplate® ASPItest (r = 0.7457-0.8893), but fair to poor correlation between LTA collagen and Multiplate® COLtest (r = 0.5887) and between LTA ADP and Multiplate® ADPtest (r = 0.0899). Of 103 patients with a mean follow up of 801 ± 249 days, 10 (9.7%) had vascular outcomes, of which six had AR by LTA-ADP. AR by LTA-ADP is associated with increased risk of vascular outcome (p = 0.034). Identification of AR is not consistent across different platelet function tests. LTA of ≥70% using 10 µM ADP in post-stroke patients taking aspirin is associated with increased risk of vascular outcome.

Venous Thrombosis in Acquired Hemophilia: The Complex Management of Competing Pathologies.

Introduction Venous thrombosis is rare in the setting of factor VIII (FVIII) deficiency. Cases of deep vein thrombosis (DVT) have been described in hemophiliacs after recent major surgery, or in association with the administration of FVIII concentrate and activated prothrombin complex concentrates, but occurrence of spontaneous DVT is even more uncommon. Aim We describe the challenging management of extensive DVT in a patient with acquired hemophilia A with concurrent hemorrhagic manifestations and review similar published cases. Methods We summarize a series of 10 cases with the following demographics: 6 males and 4 females; median age at presentation of 65 (21-80); mean inhibitor titer of 68.5 Bethesda Units (BU 1.9 to BU 350). Results Four cases were idiopathic and six had associated conditions (cancer [two cases], recent pregnancy [two cases], and recent surgery [two cases]). Three cases had an inferior vena cava filter inserted for acute lower limb DVT/pulmonary embolism. Inhibitor eradication was achieved with high-dose steroids with or without cyclophosphamide, and adjunct Rituximab administration was used in three cases. One patient received concurrent therapeutic plasma exchange (TPE). Inhibitor eradication was fastest with concurrent TPE at 6 days (range: 6-733 days). The 30-day survival was 90%. Conclusions There was adequate response of inhibitors to immunosuppression with steroids and cyclophosphamide therapy. For more refractory disease, Rituximab is emerging as a beneficial and cost-effective adjunct with better rates of complete remission, and the threshold for its use may be lowered in this complex cohort with dual competing pathologies.

Large artery ischaemic stroke in severe acute respiratory syndrome (SARS).

Of the 206 patients who contracted Severe Acute Respiratory Syndrome (SARS) in Singapore five developed large artery cerebral infarctions. Four patients were critically-ill and three died. Intravenous immunoglobulin was given to three patients. An increased incidence of deep venous thrombosis and pulmonary embolism was also observed among the critically-ill patients. We believe our experience warrants an increased vigilance against stroke and other thrombotic complications among critically-ill SARS patients in future outbreaks, especially if treatment such as intravenous immunoglobulin, that increases pro-thrombotic tendency, is contemplated.

Consensus recommendations for preventing and managing bleeding complications associated with novel oral anticoagulants in singapore.

INTRODUCTION: Novel oral anticoagulants (NOACs) have at least equivalent efficacy compared to standard anticoagulants with similar bleeding risk. Optimal management strategies for bleeding complications associated with NOACs are currently unestablished. MATERIALS AND METHODS: A working group comprising haematologists and vascular medicine specialists representing the major institutions in Singapore was convened to produce this consensus recommendation. A Medline and EMBASE search was conducted for articles related to the 3 available NOACs (dabigatran, rivaroxaban, apixaban), bleeding and its management. Additional information was obtained from the product monographs and bibliographic search of articles identified. RESULTS: The NOACs still has substantial interactions with a number of drugs for which concomitant administration should best be avoided. As they are renally excreted, albeit to different degrees, NOACs should not be prescribed to patients with creatinine clearance of <30 mLs/min. Meticulous consideration of risk versus benefits should be exercised before starting a patient on a NOAC. In patients presenting with bleeding, risk stratification of the severity of bleeding as well as identification of the source of bleeding should be performed. In life-threatening bleeds, recombinant activated factor VIIa and prothrombin complex may be considered although their effectiveness is currently unsupported by firm clinical evidence. The NOACs have varying effect on the prothrombin time and activated partial thromboplastin time which has to be interpreted with caution. Routine monitoring of drug level is not usually required. CONCLUSION: NOACs are an important advancement in antithrombotic management and careful patient selection and monitoring will permit optimisation of their potential and limit bleeding events.

Minimal residual disease-guided treatment deintensification for children with acute lymphoblastic leukemia: results from the Malaysia-Singapore acute lymphoblastic leukemia 2003 study.

PURPOSE: To improve treatment outcome for childhood acute lymphoblastic leukemia (ALL), we designed the Malaysia-Singapore ALL 2003 study with treatment stratification based on presenting clinical and genetic features and minimal residual disease (MRD) levels measured by polymerase chain reaction targeting a single antigen-receptor gene rearrangement. PATIENTS AND METHODS: Five hundred fifty-six patients received risk-adapted therapy with a modified Berlin-Frankfurt-Münster-ALL treatment. High-risk ALL was defined by MRD ≥ 1 × 10(-3) at week 12 and/or poor prednisolone response, BCR-ABL1, MLL gene rearrangements, hypodiploid less than 45 chromosomes, or induction failure; standard-risk ALL was defined by MRD ≤ 1 × 10(-4) at weeks 5 and 12 and no extramedullary involvement or high-risk features. Intermediate-risk ALL included all remaining patients. RESULTS: Patients who lacked high-risk presenting features (85.7%) received remission induction therapy with dexamethasone, vincristine, and asparaginase, without anthracyclines. Six-year event-free survival (EFS) was 80.6% ± 3.5%; overall survival was 88.4% ± 3.1%. Standard-risk patients (n = 172; 31%) received significantly deintensified subsequent therapy without compromising EFS (93.2% ± 4.1%). High-risk patients (n = 101; 18%) had the worst EFS (51.8% ± 10%); EFS was 83.6% ± 4.9% in intermediate-risk patients (n = 283; 51%). CONCLUSION: Our results demonstrate significant progress over previous trials in the region. Three-drug remission-induction therapy combined with MRD-based risk stratification to identify poor responders is an effective strategy for childhood ALL.