Automation–The next big thing

With the advancements in AI and ML, algorithms that can track multiple parameters simultaneously throughout a diagnosis, and find out patient specific patterns that can aid in pinpointing proper treatment or underlying causes, will become one of the biggest developments in coagulation measurement technologies in the upcoming years.

In recent years, blood coagulation monitoring has become crucial to diagnosing causes of hemorrhages, developing anticoagulant drugs, assessing bleeding risk in extensive surgery procedures and dialysis, and investi­gating the efficacy of hemostatic therapies. Hemostasis diagnostics devices have come a long way from quantifying optical density of clot in a cuvette to detection of clotting factors. These include flow cytometry, chromogenic assays, molecular typing, immunologic assays, functional assays of specific coagulation proteins, and platelet function analyzers. These advances in technology have resulted in greater capability, productivity, sensitivity, specificity, and ultimately improvement in the clinical care of patients.

Global hemostasis analyzers market is projected to reach USD 5 billion by 2025 from USD 3.8 billion in 2020, at a CAGR of 5.7 percent between 2020 and 2025. Emerging economies, such as India, China, and Brazil are expected to provide a wide range of opportunities for players in the industry. Furthermore, the increasing prevalence of cardiovascular diseases and blood disorders and the rising geriatric population are expected to boost the growth of this market in the coming years.

In 2020, the prothrombin-time testing segment accounted for the largest share of the hemostasis analyzers market. Growth in this market segment can be attributed to the high usage of prothrombin-time tests in measuring the time it takes for blood to clot. These tests are used to monitor patients implanted with mechanical heart valves, ventricular assist devices, chronic atrial fibrillation, venous embolism, and thrombosis of deep vessels of lower extremities, deep venous thrombosis, and pulmonary embolism.

In 2020, the clinical laboratories segment accounted for the largest share of the hemostasis analyzers market, by end-user. This can be attributed to the large volume of coagulation testing being performed at these facilities.

The hemostasis analyzers market in the Asia-Pacific is expected to grow at a high CAGR in the coming years. The factors contributing to the high growth of this market include the rising incidence/prevalence of blood-related disorders, increasing government initiatives for the improvement of healthcare infrastructure, rising number of general surgeries, and growth in research activities in several Asia-Pacific countries. Moreover, the rapidly growing geriatric population (and the subsequent increase in the incidence of thalassemia, sickle-cell anemia, and leukemia) and healthcare infrastructure improvements in countries, such as Japan, China, and Australia are expected to support the growth of this market in the coming years.

Abbott Laboratories, Danaher, Siemens Healthineers Group, Sysmex Group, Thermo Fisher Scientific, Horiba, Nihon Kohden, Roche Diagnostics, Maccura Biotechnology Co. Ltd., Bio Group Medical System, Genrui Biotech Inc., A&T Corporation, Beijing Succeeder Technology Inc., Diagnostica Stago SAS, Helena Laboratories, Hycel, Eurolyser, Bpc Biosed Srl, ERBA Diagnostics, Inc., Sclavo Diagnostics International S.r.l., Haematonics, Biosystems S.A., and ACON are the leading players in this segment.

Automation has prepotently revolutionized workflow and sample management in modern laboratories, most of which takes profit from the concept of lean manufacturing, which entails eliminating or minimizing all steps and activities that do not add value to the whole process. Several technical solutions exist that can be customized around specific needs of organization and testing volume. Hemostasis testing has remained virtually apart from this innovation for a long time, and only recently has the concept that coagulation analyzers may be combined within a model of (total) laboratory automation started to emerge. Some recent studies have shown that automation solutions and connection of coagulation analyzers to track-line systems are reliable strategies for improving efficiency (i.e., throughput and turnaround time) and decreasing costs in clinical laboratories, with negligible effects on the quality of routine hemostasis testing.

Recent years have seen the advancement of the use of robotics in the laboratory setting. The use of automation allows for many processes that once required a considerable amount of resources dedicated to them, such as time and human workers, to be conducted at faster speeds, with greater levels of consistency, permitting the conservation of time and human resources within the laboratory.

Hemostatic testing is one such procedure that laboratory robotics has automated. Previously, it has been explored whether the process of analyzing hemostatic samples could reliably be automated.

The key concern was the vulnerable nature of the samples; they are prone to detrimental impacts via the movements that automation introduces to samples, such as track, belt, or conveyor transportation.

Early studies demonstrated that robotic handling might potentially affect the quality of the hemostatic testing results. Samples for hemostatic testing are centrifuged before analysis. Automated transportation systems of these centrifuged samples may cause partial resuspension of pellet cells within the upper plasma phase. This has been of concern when the samples were transported at high speeds. Previously, automated systems have been compared to manual direct loading into the analyzer. Although significant differences were found between the samples that were directly loaded into the analyzer compared to samples that were entered via the automated track-line system, the results of the track-line system were still within the total allowable error.

It has been demonstrated that the processing of plasma samples through extensive track-line systems affects the results of routine hemostasis testing. Nevertheless, as the differences were not focused clinically significant, the automated processing of hemostasis testing remains a viable option.

The impact of movement on the samples must be considered to limit the negative effects on the quality of the results.

Indeed, the automation and high throughput means that extra care needs to be placed on ensuring sample integrity. Thus, it is not good enough to get a fast PT or APTT (for example), one need to ensure that these test results are accurate and reflect the patient’s status, and not just the sample status. Modern instruments are generally very reliable; nonetheless, errors in test results can occur. Importantly, pre-analytical issues account for the majority of test errors in modern times. Such pre-analytical issues often arise due to poor blood collection, and the arising sample activation and clotting prior to receipt by the laboratory. The major outcome is the development of hemolysis and also partial or complete clots. Neither is easily identified by visual inspection of the whole blood, unless the clot is large enough to prevent blood mixing; however, both can certainly compromise sample quality. Typically, samples arrive in the laboratory, patient details and ordered tests are entered into the laboratory information system (LIS), and the collection tubes are quickly centrifuged. The primary sample, with the plasma supernatant resting on top of the centrifuged cells, is typically placed on the analyzer and the tests performed, usually after automatic query of the test(s) required through interrogation of the LIS, and then test results are outputted from the analyzer to await verification. This process may be manual, or automated through a series of rules. In general, normal test results can usually be verified automatically, but the abnormal test results require a higher level of scrutiny. For any individual patient, the abnormal test result may actually be expected (e.g., on anticoagulant therapy), or else unexpected. In the latter case, the question arises: is the test result truly reflective of the patient status, or is the result an artefact of a pre-analytical issue?

Modern instruments now come with several checks for pre-analytical problems, including the so-called HIL (hemolysis, icterus, lipemia) triad. In such cases, the instrument can usually overcome the spectral effects and get an accurate test result reflective of the test sample; however, the instrument cannot account for any biological effects of HIL, and a recollection may still be recommended. In the case of potential sample clotting causing a spurious coagulation test result, the usual check requires closer visual inspection of the sample, potentially using wooden sticks to physically check the sample for visual clots. This is a time-consuming process, and even if well performed may miss small clots, which could still compromise sample quality.

The concept of ML is not in itself new. Indeed, a PubMed search using machine learning yields over 55,000 hits, dating back to at least 1967. Adding the terms coagulation or hemostasis, however, cuts the yield down to 105 papers, with over 80 of these published in the last 5 years; hence the concept is certainly new for this field. Till date, no prior paper has ever explored machine learning to identify clotted specimens in coagulation testing.

Fang et al. in their publication decided to use back­propagation neural networks (BPNNs), which is a widely used approach in neural networks classifiers. Fang et al. retrospectively retrieved from their LIS the results of coagulation testing, with 192 clotted and 2889 no-clot-detected (NCD) samples to form datasets for training and testing. Standard and momentum BPNNs were trained and validated using the training dataset with five-fold cross-validation, and then the predictive performances of the models were assessed on a testing dataset. The standard and momentum BPNNs could identify the sample status (clotted and NCD) with areas under the receiver operating characteristic (ROC) curves (AUCs) of 0.966 (95% CI, 0.958–0.974) and 0.971 (95% CI, 0.9641–0.9784), respectively. These represent extraordinarily high AUC values and indicate both high sensitivity and specificity for identification of clotted vs unclotted samples.

So, what does this mean, a method for identification or exclusion of clotted samples, and no longer need to visually inspect samples? Well, not just yet. As a caveat to the application of this process, the dataset comprised samples with five tests performed, being PT, APTT, thrombin time (TT), fibrinogen, and D-dimer. In other words, the model requires results from all five tests to enable the prediction to occur. In most laboratories, PT and APTT are the most-often performed tests, and TT, fibrinogen, and D-dimer are performed far less frequently, and generally in <10 percent of patients who have coagulation studies performed. Nonetheless, TT, fibrinogen, and D-dimer are often those tests that expert rules will reflex to in the case of unexplained prolongation of PT and/or APTT, and also with reference to patient information suggesting their potential utility. Thus, the work of Fang et al. should be seen as an important first step in the path toward an automated identification (or exclusion) of clots as the cause of unexpected abnormal coagulation test times. Further work can be look forward in this area over coming years.

The future of the hemostasis laboratory is currently unwritten and can not be accurately predicted. Currently, automation is being adopted clinically due to its numerous benefits, including consistency, faster turnaround time, and reduced overall cost. Furthermore, there are several additional indirect benefits, such as permitting the allocation of skilled staff to other areas, further reducing costs, and an improvement in patient care.

For automated hemostasis testing to become widely adopted in the future, current challenges need to be considered. The high cost of maintenance, the dependence on IT and the manufacturer, and the reduced flexibility of the testing process need improvement. The set-up costs of automation systems are high, which may prevent laboratories from investing in automation equipment.