It is no surprise that clinical laboratories are facing myriad challenges in today’s health care climate, not the least of which involve staffing. Fewer new MT and MLS graduates combined with an increase in staff retirement is not a novel problem for clinical laboratories. However, the onset of the pandemic and its enduring impact has only served to couple these realities with further staff attrition (due to a variety of pandemic-related issues) and exacerbated pressure via louder calls for faster “STAT” turnaround times and increased on-boarding of new testing to drive revenue and accommodate all things COVID. To alleviate some of this pressure, laboratory directors and leaders across the country are seeking new ways to meet the needs of their providers and patients.
Microbiology Automation Through a Different Lens
Incorporating automation is certainly a logical method through which many laboratory practice areas have been able to counterbalance some of these challenges. In certain departments, adding new instrumentation such as front-end processors to accession, route for testing, and release results has decreased the number of staff needed in the laboratory overall. Specimen processing and core chemistry are universal departments that often benefit from this type of automation due in part to the “black and white” nature of their results. While such advancements are good for the laboratory as a whole, not all disciplines benefit equally.
Traditionally, microbiology has been one department that has not experienced the same rapid growth in automation due largely to the high complexity and acuity required for cultures work up. Accordingly, it has become increasingly important to view automation in microbiology through a different lens. Rather than solely looking to automation to fill the need for staffing, seek it out as a tool to increase the efficiency of current microbiology tasks, facilitate workflow, and perhaps most important, cultivate the skills of new microbiology technologists.
Target Specific Microbiology Tasks with Automation
Microbiology workflows often can be seen as intricate dances where a number of necessary but time-consuming side steps are incorporated to obtain the proper results of a culture report (see FIGURE 1). Through the implementation of certain automation to specific areas of microbiology workflow, valuable efficiencies can be gained that substantially mitigate the time required to issue a report. The employment of front-end microbiology processors is one example of automated instrumentation used to expedite and improve the accuracy of specimen processing. These instruments can further inoculate urine, sputum, and wound specimens, as well as transport them to incubation holding containers.
Other versions of these instruments have further capabilities to capture images of plates and enable technologists to select colonies more rapidly for ID and susceptibility testing. Furthermore, modular systems designed to improve certain aspects of culture reading also are available on the market with a mature and established presence in many microbiology labs. Automated antibiotic susceptibility testing (AST) and identification systems have helped cut reporting times down drastically from that of archaic phenotypic identification strategies that required upwards of 50 biochemical tests in tubes and evaluating a bacterial isolate’s metabolic capabilities.
Bring Artificial Intelligence to Bear
While many laboratories look at implementing automation to cut down the overall time it takes to report preliminary or final reports (and rightly so), there is another equally if not more important aspect to consider when looking at microbiology automation. This aspect addresses the question: How do we use this technology and automation to maximize the efficiency of microbiology processes, but also utilize it to sharpen and improve the capabilities and skills of our staff? This is where recent breakthroughs in the use of artificial intelligence (AI) in automation can provide significant benefits for all laboratories.
When instrumentation with plate reading capabilities is incorporated into the lab, the benefit is three-fold: First, the time it takes for a culture report to be generated is decreased by cutting down the number of human touch points and enabling techs to process cultures more quickly. Second, decreasing the overall number of cultures that need to be evaluated by a tech frees up more time to complete other, perhaps lower-priority testing, but still requires a high level of staff competency, such as ova and parasite (O&P) reading. This capability is especially valuable for laboratories with a new or less experienced workforce.
Enabling experienced techs to spend less time on routine cultures is one benefit, but there are other elements of AI automation that enrich microbiology practice. With today’s outflux of experienced techs, it is becoming increasingly difficult to ensure new staff receive the full training and assurance they need to feel confident in making appropriate clinical decisions on their own. In some respects, automation can provide positive reinforcement as new bench techs gain valuable experience. When automation is used to sort plates based on both the quantity of bacterial growth as well as the significance of that growth, techs are aware that what they are working with is of clinical importance.
Impact of AI on Microbiology Workflow
To understand how automation and AI produces these benefits, we need to look at traditional microbiology lab workflow and how it changes once AI is introduced. In basic traditional microbiology workflow, specimens are received in the lab, plated to the appropriate media, and then incubated. Since not all specimens are received at the same time, incubators are constantly being opened and closed to add new cultures throughout the day. After incubation, plates are evaluated for growth and worked up appropriately, usually at fixed intervals depending on the staffing levels of each shift. Results are then entered into the laboratory information system (LIS) and thorough preliminary or final reports are issued.
Many micro laboratories perform the bulk of plate reading during the day shift, as this is when the greatest number of staff members are on hand. Thus, regardless of incubation length, all cultures are likely to be evaluated at roughly the same time. If cultures require additional work up, such as further incubation, sub culturing for identification, or AST, multiple days’ time may be needed before a final report can be issued. Furthermore, this projection does not take into consideration the time required for specimens to be sorted and accessioned by a lab’s triage department, or the processing time required once those specimens are received into the microbiology department. Therefore, placing automated instruments into the workflow at practically any point in the pre-analytical processing stage can help increase efficiencies.
Current AI Options in Microbiology
Fortunately, there are few different options involving AI-based automation that could benefit the microbiology laboratory depending on need. Certainly constants, such as available budget and physical space, are to be considered, as well as the overall automation goals of the laboratory. While budget and space have been major decision drivers in the past, new compact technology and a wider field of available options in the market have allowed department leaders to look instead at what would be most beneficial for workflow and efficiency, rather than picking something that fits the confines of budget or footprint alone.
Today, there is a substantial amount of customization allowances built into the design of lab automation. For some labs, total lab automation (TLA) systems may prove to be the best option. These systems are designed to encompass the entire microbiology workflow from specimen processing through culture reporting. Among the key benefits of this these systems are AI-based, front-end processing capabilities combined with workflow standardization. These instruments can process many specimens in a short amount of time, with little necessary involvement of skilled lab tech. This in turn enables laboratories to minimize staffing for specimen processing, while redirecting available skilled staff toward complex work, such as plate reading. After initial set up of these automated instruments, cultures are held in incubators within the system until they have reached the minimum amount of time needed for culture growth. Plates are then sent through sophisticated imaging software that enables a tech to evaluate growth via computer and make workup decisions based on standardized lab protocols (see FIGURE 2).
Microbiology-based TLA systems help to further standardize much of the microbiology workflow, as timing is monitored by the system and cultures can be monitored continuously. While large hospital systems and reference labs may see the greatest benefit of TLA given their tendency to have higher sample volumes, more numerous staffing levels per shift, and larger spaces overall physical spaces and budgets, the time is coming when automation will proliferate all clinical laboratory phases, and the microbiology department stands to benefit greatly from this evolution.
In Part 2 of this article, we will discuss modular automation systems in microbiology, important logistical considerations, and the value of clear communications with information technology staff and other key laboratory and microbiology stakeholders.
Jacqueline Getty, MLS(ASCP), is a medical technologist in the microbiology department at Hennepin County Medical Center. She received her MLS degree from the University of Minnesota. In addition to her clinical work in microbiology Jacqueline is involved in Hennepin County’s rapid testing protocols for COVID-19 at the hospital level.