Leveraging artificial intelligence in the laboratory

The healthcare industry is facing a shortage of laboratory technicians and medical laboratory scientists.1 In the search for answers, does the revolutionary power of artificial intelligence (AI) have a role to play? 

Laboratory personnel utilizing the power of artificial intelligence (AI) can expect to see many transformative benefits. Whether taking over the performance of repetitive and mundane tasks or easing the workload of personnel in understaffed laboratories, artificial intelligence (AI) is a tool that is reshaping the ways in which laboratory personnel perform their work.

Article highlights:
  • AI can support laboratory personnel to overcome some of their most pressing challenges, including understaffing.
  • The benefits of AI for in the laboratory extend beyond operational efficiency and resource management.
  • AI is propelling laboratory personnel to the forefront of scientific and medical innovation and empowering them to foster meaningful breakthroughs that will shape our world.
animated speaking bubble

Join our community and stay up to date with the latest laboratory innovations and insights.

Subscribe now

Managing laboratory operations so personnel don’t have to

A key advantage of AI for laboratory personnel is that it allows for the automation of routine and repetitive tasks in the laboratory.2 AI is able to extract data from a range of sources, such as test reports or observations from hand-written notes, and populate electronic records, reducing the need for laboratory personnel to perform manual data entry.3

Through the analysis of equipment’s historical data and usage patterns, AI systems can perform automatic scheduling of repairs and maintenance, thereby avoiding equipment breakdown and costly delays.4

In terms of inventory management, AI-driven systems monitor the use of products and materials and detect shortages and expirations in real time.5 When inventory levels fall below a certain threshold, AI systems trigger automatic orders to replenish stock.5

By managing operations tasks like these, laboratory personnel are given more time to focus on meaningful work such as research, data interpretation, and supporting decision-making.

Advanced image analysis to support rapid and accurate diagnosis

AI is an incredibly powerful tool for image analysis, enabling us to explore and extract information beyond what is possible with visual human perception.6 AI can identify regions of interest on whole slide images and quantify parameters such as nuclear morphology and tissue architecture. AI thereby improves reproducibility and precision, while freeing up the laboratory team from mundane and time-consuming imaging tasks.2,6,7

In oncology, AI-based algorithms have enabled pathologists to quantify breast cancer markers, grade prostate cancer, predict responses to chemoradiotherapy in rectal cancer, and predict recurrence in early-stage non-small cell lung cancer through the identification of cell and tissue morphological features.6,7,8 By increasing the amount of information available for analysis and the speed with which images are analyzed, AI supports pathologists to make more rapid and accurate diagnoses.

Making personalized medicine the standard of healthcare

AI is supporting the advancement of personalized medicine in many ways and will place laboratory personnel at the forefront of this exciting frontier. For example, AI algorithms will be used in the laboratory to analyze genetic sequences and identify disease-causing mutations in patient specimens. This information, when combined with data about the patient such as medical history, lifestyle, and symptoms, will allow physicians to provide increasingly personalized diagnoses and treatment recommendations. These developments also pave the way for novel prognostic scores.7,9,10

AI algorithms that can analyze genetic mutations in tumor cells and recommend targeted therapies based on the specific genetic alterations are already a reality. This approach, known specifically as precision oncology, is being used by medical and laboratory teams to improve treatment outcomes by targeting the underlying genetic drivers of the disease.11,12

Capturing actionable insights from big data

AI techniques, including machine learning, are being leveraged by researchers to identify meaningful patterns in complex datasets, that will provide insights relevant for healthcare and research purposes.13 Genomic data is a prime target for AI analysis and accordingly, it is being used to unravel the secrets held within our genomes by identifying genetic variations associated with diseases like cancer, heart disease, and rare genetic disorders. 

Despite AI being in its infancy, researchers are already benefitting from its use in healthcare. Machine learning is paving the way for the detection and profiling of cancer in humans using blood samples.14 AI is also being used by laboratory personnel to predict how cancer will progress in patients15 and understand how the genome of influenza will vary in the future, to assist public health efforts.13

Pushing the needle in drug discovery

The process of discovering and developing new drugs has traditionally centered on high-throughput screening for known disease-associated targets. As a result, the drug discovery process has typically been long, expensive, and relatively ineffective.16 On average, the cost of the R&D process is estimated at over US$2 billion per drug.16

In February 2023, the FDA granted the first orphan drug designation for an investigative compound which was designed using a generative AI platform.17 AI algorithms provide drug discovery scientists with a powerful tool that can predict how investigational compounds might interact with biological targets. AI also assists with virtual screening and increases the accuracy of predictions of the safety and efficacy of drugs. As a result, laboratory personnel are able to identify development candidates with a high potential for success prior to initiating a clinical trial. Ultimately, this will make clinical trials more predictable, leading to a reduction of time and costs in the drug development pipeline.16

Barriers to implementing artificial intelligence in the laboratory

While AI has many benefits to offer, some challenges remain regarding its implementation. As a first step, it will be crucial to ensure that AI-based algorithms are trained using representative and unbiased data. Without such a foundation, AI may produce unreliable and inaccurate results which may have significant consequences in the healthcare space. Mistrust of AI is a barrier to its use. AI systems must therefore be transparent and explainable. 

There is also a common belief that AI will result in job displacement. This must be tackled to ensure AI is seen as a tool that can enhance human capabilities rather than replace them. Finally, the implementation of AI-driven technologies in laboratory settings introduces new vulnerabilities, as sensitive data becomes a valuable target for cyberattacks. It will be key to strike a balance between harnessing the benefits of AI whilst ensuring data and patients are well protected.

A symbiotic collaboration between humans and artificial intelligence

The benefits of artificial intelligence (AI) in the laboratory extend beyond operational efficiency and resource management. From accelerating drug discovery by predicting molecular interactions to revolutionizing diagnostics with swift and precise image analysis, AI is propelling laboratory personnel to the forefront of scientific and medical innovation and empowering them to foster meaningful breakthroughs that will shape our world.

  1. Robinson A. (2023). Biomed J Sci & Tech Res 48, 39710-39713. Paper available from [Accessed September 2023] 

  2. The Royal College of Pathology.(2023). Article available from [Accessed September 2023] 

  3. Rossum. (2023). Article available from [Accessed September 2023] 

  4. Marescotti D et al. (2022). SLAS Technology 27,195-203. Paper available from [Accessed September 2023] 

  5. McKinsey. (2021). Article available from [Accessed September 2023] 

  6. Baxi V et al. (2022). Modern Pathology 35, 23-32. Paper available from [Accessed September 2023] 

  7. Niazi M et al. (2019). Lancet Oncol 20, e253-e261. Paper available from [Accessed September 2023] 

  8. Roche Diagnostics. (2023). Article available from [Accessed September 2023] 

  9. Volynskaya Z. (2018). Arch Pathol Lab Med 142, 369-382. Paper available from [Accessed September 2023] 

  10. Johnson K et al. (2021). Clin Transl Sci 14, 86-93. Paper available from [Accessed September 2023] 

  11. Yala A et al. (2019). Radiology 292, 60-66. Paper available from [Accessed September 2023] 

  12. Coudray N et al. (2018). Nature Medicine 24, 1559-1567. Paper available from [Accessed September 2023] 

  13. National Human Genome Research Institute. (2022). Article available from,will%20progress%20in%20a%20patient.  [Accessed September 2023] 

  14. Cristiano S et al. (2019). Nature 570, 385-389. Paper available from  [Accessed September 2023] 

  15. Caravagna G et al. (2018). Nature Methods 15, 707-714. Paper available from  [Accessed September 2023] 

  16. Deloitte. (2019). Article available from  [Accessed September 2023] 

  17. Genetic engineering and biotechonology news. (2023). Article available from  [Accessed September 2023]