Search

Background A risk assessment model for predicting the risk of haemolytic disease of the fetus and newborn (HDFN) in future pregnancies following the transfusion of Rh(D)-positive red blood cell (RBC)-containing products to females of childbearing potential (FCP) was developed, accounting for the age that the FCP is transfused in various countries. Methods The HDFN risk prediction model included the following inputs: risk of FCP death in trauma, Rh(D) alloimmunization rate following Rh(D)-positive RBC transfusion, expected number of live births following resuscitation, probability of carrying an Rh(D)-positive fetus, the probability of HDFN in an Rh(D)-positive fetus carried by an alloimmunized mother. The model was implemented in Microsoft R Open, and one million FCPs of each age between 18 and 49 years old were simulated. Published data from eight countries, including the United States, were utilized to generate country-specific HDFN risk estimates. Results The risk predictions showed similar characteristics for each country in that the overall risk of having a pregnancy affected by HDFN was higher if the FCP was younger when she received her Rh(D)-positive transfusion than if she was older. In the United States, the overall risk of HDFN if the FCP was transfused at age 18 was 3·4% (mild: 1·20%, moderate: 0·45%; severe: 1·15%; IUFD: 0·57%); the risk was approximately 0% if the FCP was 43 years or older at the time of transfusion. Conclusion This model can be used to predict HDFN outcomes when establishing transfusion policies as it relates to the administration of Rh(D)-positive products for massively bleeding FCPs.

Flow cytometric (FC) immunophenotyping is critical but time-consuming in diagnosing minimal residual disease (MRD). We evaluated whether human-in-the-loop artificial intelligence (AI) could improve the efficiency of clinical laboratories in detecting MRD in chronic lymphocytic leukemia (CLL). We developed deep neural networks (DNN) that were trained on a 10-color CLL MRD panel from treated CLL patients, including DNN trained on the full cohort of 202 patients (F-DNN) and DNN trained on 138 patients with low-event cases (MRD < 1000 events) (L-DNN). A hybrid DNN approach was utilized, with F-DNN and L-DNN applied sequentially to cases. “Ground truth” classification of CLL MRD was confirmed by expert analysis. The hybrid DNN approach demonstrated an overall accuracy of 97.1% (95% CI: 84.7–99.9%) in an independent cohort of 34 unknown samples. When CLL cells were reported as a percentage of total white blood cells, there was excellent correlation between the DNN and expert analysis [r > 0.999; Passing–Bablok slope = 0.997 (95% CI: 0.988–0.999) and intercept = 0.001 (95% CI: 0.000–0.001)]. Gating time was dramatically reduced to 12 s/case by DNN from 15 min/case by the manual process. The proposed DNN demonstrated high accuracy in CLL MRD detection and significantly improved workflow efficiency. Additional clinical validation is needed before it can be fully integrated into the existing clinical laboratory practice.

Context.—Recent developments in machine learning have stimulated intense interest in software that may augment or replace human experts. Machine learning may impact pathology practice by offering new capabilities in analysis, interpretation, and outcomes prediction using images and other data. The principles of operation and management of machine learning systems are unfamiliar to pathologists, who anticipate a need for additional education to be effective as expert users and managers of the new tools.Objective.—To provide a background on machine learning for practicing pathologists, including an overview of algorithms, model development, and performance evaluation; to examine the current status of machine learning in pathology and consider possible roles and requirements for pathologists in local deployment and management of machine learning systems; and to highlight existing challenges and gaps in deployment methodology and regulation.Data Sources.—Sources include the biomedical and engineering literature, white papers from professional organizations, government reports, electronic resources, and authors' experience in machine learning. References were chosen when possible for accessibility to practicing pathologists without specialized training in mathematics, statistics, or software development.Conclusions.—Machine learning offers an array of techniques that in recent published results show substantial promise. Data suggest that human experts working with machine learning tools outperform humans or machines separately, but the optimal form for this combination in pathology has not been established. Significant questions related to the generalizability of machine learning systems, local site verification, and performance monitoring remain to be resolved before a consensus on best practices and a regulatory environment can be established.

The full potential of predictive maintenance has not yet been utilised. Current solutions focus on individual steps of the predictive maintenance cycle and only work for very specific settings. The overarching challenge of predictive maintenance is to leverage these individual building blocks to obtain a framework that supports optimal maintenance and asset management. The PrimaVera project has identified four obstacles to tackle in order to utilise predictive maintenance at its full potential: lack of orchestration and automation of the predictive maintenance workflow, inaccurate or incomplete data and the role of human and organisational factors in data-driven decision support tools. Furthermore, an intuitive generic applicable predictive maintenance process model is presented in this paper to provide a structured way of deploying predictive maintenance solutions.

Objectives Despite more than 40 years of experience performing the Bethesda assay (BA), poor intra- and interlaboratory precision remains the biggest laboratory challenge to date. Methods The BA procedure was modeled using stochastic simulation techniques to determine the precision of the BA up to dilutions of 1:4,096, to estimate the minimum significant relative change at various inhibitor titers, and to understand the laboratory procedural variables that could significantly affect the performance of the BA at high dilutions. Results Selecting the lowest dilution tube with a residual activity closest to 25% for calculating the reported Bethesda titer (BT), using a factor activity assay with a coefficient of variation less than or equal to 7.5% in the range of 15% to 50% factor activity level, performing the factor activity measurement in replicates, and minimizing pipette volumetric error resulted in the lowest imprecision in the reported BT. The factor neutralization kinetics of the inhibitor appear to have little impact on the precision of the assay if the incubation time is greater than 90 minutes. Conclusions This in silico model will assist future laboratory efforts in standardizing the quantification of specific coagulation factor inhibitors and improving the precision of the reported results.

Item does not contain fulltext X, 301 p.

We give a formal verification procedure that decides whether a classifier ensemble is robust against arbitrary randomized attacks. Such attacks consist of a set of deterministic attacks and a distribution over this set. The robustness-checking problem consists of assessing, given a set of classifiers and a labelled data set, whether there exists a randomized attack that induces a certain expected loss against all classifiers. We show the NP-hardness of the problem and provide an upper bound on the number of attacks that is sufficient to form an optimal randomized attack. These results provide an effective way to reason about the robustness of a classifier ensemble. We provide SMT and MILP encodings to compute optimal randomized attacks or prove that there is no attack inducing a certain expected loss. In the latter case, the classifier ensemble is provably robust. Our prototype implementation verifies multiple neural-network ensembles trained for image-classification tasks. The experimental results using the MILP encoding are promising both in terms of scalability and the general applicability of our verification procedure.

Contains fulltext : 225791.pdf (Publisher’s version ) (Closed access) ISoLA

Damage control resuscitation (DCR) of the massively bleeding trauma patient comprises early intervention with blood products to replete intravascular volume, restore tissue oxygenation, and correct coagulopathy until surgical hemostasis can be achieved. Blood product resuscitation in the emergency department (ED), and also in the prehospital setting, must be coordinated, prompt, multifaceted, and evidence-based and should incorporate the appropriate use of laboratory tests and adjunct therapies, which facilitate DCR and damage control surgery. Hemostatic resuscitation should also address the management of the acute and delayed complications of massive transfusion including electrolyte imbalances, interstitial edema, acidosis, and hypothermia.

Objectives To determine a quantitative herpes simplex virus (HSV) DNA threshold in lower respiratory tract specimens that correlates with positive viral culture and clinical outcomes. Methods Bronchoalveolar lavage and bronchial wash samples from 53 HSV culture-positive and 61 culture-negative matched controls were tested using HSV-1 and HSV-2 quantitative polymerase chain reaction (qPCR). Results Median viral culture turnaround time was 21.8 days and 9.9 days for culture-negative and culture-positive specimens, respectively. Using an HSV-1 viral load threshold of 1.62 × 103 copies/mL, there was 93% agreement with viral culture. An HSV-1 viral load ≥1.3 × 104 copies/mL was associated with worse clinical outcome compared to a viral load

Background The serological safety of transfusing ≤4 units of low titer group O whole blood (LTOWB) in civilian trauma patients has been demonstrated. This study investigated clinical outcomes of LTOWB recipients compared to patients who received only conventional blood components during their resuscitation. Study design and methods A retrospective analysis of trauma patients' medical records who received LTOWB during the first 24 hours of their admission was performed. Using a 12-parameter propensity matching strategy, LTOWB recipients were matched to other patients who received at least one red blood cell (RBC) unit during their first 24 hours of admission but not LTOWB. The primary outcomes were mortality and blood use. Results A total of 135 patients who received LTOWB (median 2 units) were matched to 135 patients who received conventional components. There were no significant differences in the matching parameters between the groups. There were no significant differences in outcomes between the conventional component and LTOWB groups: median (interquartile range) in-hospital mortality, 24.4% vs. 18.5% (respectively, p = 0.24); 24-hour mortality, 12.6% vs. 8.9% (respectively, p = 0.33). The hospital and intensive care unit lengths of stay were not significantly different between groups. The median number of RBC units transfused, including the contribution from the LTOWB, was not significantly different between the groups. The time to normalization of elevated plasma lactate levels tended to be shorter among the LTOWB recipients compared to the conventional component recipients (median 8.1 [3.7-15.4] hr vs. 13.2 [4.4-26.8] hr, respectively, p = 0.05). Conclusion The LTOWB recipients had similar clinical outcomes compared to recipients of conventional component therapy.

Thyrotoxic periodic paralysis is a rare complication of hyperthyroidism where increased influx of potassium into skeletal muscle cells leads to profound hypokalaemia and paralysis. Most cases arise sporadically in Asians; however, it is being increasingly reported in Caucasians. It is regarded as a channelopathy where a genetic and/or acquired defect in the sodium-potassium (Na/K-ATPase) pump renders it more sensitive to excess thyroid hormone in susceptible individuals. Because the clinical presentation is similar to familial hypokalaemic periodic paralysis, genes implicated in this autosomal-dominant condition became candidates for thyrotoxic periodic paralysis, particularly if they were known to have thyroid hormone-responsive elements. These include the voltage-gated calcium (CACNA1S) and sodium (SCN4A) channel genes, KCNJ18 which encodes the inwardly rectifying potassium channel Kir2.6, and subunits of the Na/K-ATPase genes. Although no single pathogenetic mutation has been identified in thyrotoxic periodic paralysis, several single-nucleotide polymorphisms in these genes have been associated with it. We describe a 27-year-old Caucasian Irish male who presented with acute onset limb paralysis and severe hypokalaemia. He was diagnosed as having thyrotoxic periodic paralysis secondary to Graves’ disease based on clinical presentation, biochemical findings and rapid response to intravenous potassium. Genetic analysis identified heterozygous variants in three candidate genes: KCNJ18 (c.576G>C), SCN4A (c.2341G>A) and CACNA1S (c.1817G>A). Since these variants are not disease causing and occur at high prevalences of 50%, 2–3% and 1%, respectively, in the normal population, they do not explain the clinical phenotype in our patient suggesting that acquired environmental triggers or as-yet unidentified gene mutations remain as leading pathogenetic co-factors in thyrotoxic periodic paralysis.

BACKGROUND The use of cold-stored low-titer group O whole blood (LTOWB) for civilian trauma patients is gaining popularity. However, hemolysis might occur among non-group O recipients. This study evaluated the serologic safety of transfusing up to 4 units of LTOWB. STUDY DESIGN AND METHODS Hypotensive male and at least 50-year-old female trauma patients who received leukoreduced, uncrossmatched, group O+, low-titer (

Background Bleeding emergencies can complicate pregnancies. Understanding the disposition of the products that are issued in this clinical setting can help inform inventory levels at hospitals where obstetric patients are seen. Study design and methods Patients who had an obstetric hemorrhage of any etiology between January 2013 and June 2017, and whose resuscitation began with uncrossmatched red blood cells (RBCs) or emergency-issued plasma or platelets (PLT), were included. The disposition of all blood products issued within 6 hours of the first uncrossmatched or emergency-issued product was documented, as was basic patient demographic information. Results In total, 301 women with an obstetric hemorrhage from seven academic institutions were identified. Their mean ± standard deviation age was 30.9 ± 6.1 years, 45.2% delivered by Cesarean section, and 40.5% delivered vaginally, while 12% did not deliver. The largest single etiology of hemorrhage was related to abnormal placentation. Of the 2280 issued RBC units, 55% were transfused, 43% were returned, and 2% were wasted. The rates of transfusion of the other blood products ranged from 58% for plasma units to 82% for cryoprecipitate. Seventeen percent of the issued cryoprecipitate units were wasted, the highest of any blood product. The rate of a patient receiving a transfusion when at least one blood product had been ordered ranged from 74% for PLTs to 91% for cryoprecipitate. Conclusion Although the rates of receiving a transfusion of at least one blood product when one is ordered was high, many of the issued units were returned, especially for RBCs.

BACKGROUND There are many influences on a hospital's demand for plasma. Pharmaceuticals are now being administered for many indications instead of plasma, although trauma resuscitation now emphasizes increased and early intervention with plasma. This multinational study evaluated changes in blood center plasma unit distributions over a 10-year period. STUDY DESIGN AND METHODS Data on the total number and the ABO groups of plasma unit distributions were obtained from nine American blood collectors (ABCs) and nine national or provincial blood services (NPBS) from 2007 through 2016. Plasma distributions to trauma hospitals by five ABCs and four NPBS were also analyzed. RESULTS The overall number of plasma unit distributions from ABCs decreased by 23.1% from 2007 to 2016, but the relative proportion of distributed AB plasma units increased during the same period. The NPBS (excluding the Japanese Red Cross [JRC]) also had a 35.4% decrease in the overall number of plasma unit distributions with an increase in the relative proportion of AB plasma distributions between 2007 and 2016. The JRC, however, reported an increase in the overall number of plasma distributions by 13.5% in 2016 compared to 2007. The proportion of low-titer A plasma distributions increased to 1.6% of total plasma distributions by ABCs in 2016. There was a trend of distributing increasing proportions of group AB plasma units to trauma hospitals over the 10-year period. CONCLUSION Although the number of plasma unit distributions has decreased at many blood collectors over time, the proportion of AB units has increased at both ABCs and NPBS.