AI in laboratory services – A perspective

Artificial intelligence (AI) refers to sophisticated software systems that enable computers to augment, or even emulate, human intelligence and decision making. In a survey from a pharma company of 200 hospital or laboratory senior executives and lab directors explored the future of the impact of AI on the in vitro diagnostic (IVD) laboratory. As per the survey, 69 percent expect widespread adoption of AI in the IVD lab within the next 4 years and 92 percent expect AI to have a significant impact on healthcare eventually.

The first devices based on AI have been applied to routine laboratory data management. Many digital microscopy systems are rapidly making their place in the laboratory. These systems while enabling morphological analysis of blood smears in an automated manner offer many advantages, e.g., enabling remote review, digital archiving, easy retrieval of blood films, reduced fatigue and eyestrain, facilitation of morphology education, QA training, and peer review in digital format. Paige.AI that stands for Pathology AI Guidance Engine is a new startup that uses artificial intelligence to fight cancer. It is a startup in computational pathology, focused on building AI, and which has received the breakthrough device designation from the USFDA. This device is focused on providing artificial intelligence tools to pathologists that will enable them to become faster and more accurate in their diagnosis and treatment recommendations, and in the process has collaborated with Memorial Sloan Kattering.

There are several factors driving adoption of AI and deep learning in healthcare: the strengths of digital imaging over human interpretation; the digitization of health-related records and data sharing; the adaptability of deep learning to analysis of heterogeneous data sets; the capacity of deep learning for hypothesis generation in research; the promise of deep learning to streamline clinical work flows and empower patients; the rapid-diffusion open source and proprietary deep learning programs; and the adequacy of today’s basic deep-learning technology to deliver improved performance as data sets get larger. Early AI systems represented human reasoning with symbolic logic. The first deep-learning algorithms were supervised in that human experts continued to label the training data, and the deep-learning algorithms learned the features and weights directly from the data.

Algorithms that learn from human decisions will also learn human mistakes: overtesting and overdiagnosis; failure to notice those who lack access to care; undertesting those who cannot pay; and mirroring race or gender biases. Ignoring these facts will automate and even magnify problems in our current health system. Noticing and undoing these problems requires a deep familiarity with clinical decisions and the data they produce – a reality that highlights the importance of viewing algorithms as thinking partners, rather than replacements, for doctors. Consideration of the ethical challenges inherent in implementing machine learning in healthcare is also warranted if the benefits are to be realized.

Outlook
There is little doubt that algorithms will transform the thinking underlying medicine. The only question is whether this transformation will be driven by forces from within or outside the field. If medicine wishes to stay in control of its own future, doctors and laboratory managers will not only have to embrace algorithms, they will have to excel at developing and evaluating them, bringing machine-learning methods into the medical domain. Since deep-learning systems are trained on data from the past, they are not prepared to reason in the way humans do about conditions that have not been seen before. Last but not the least, defying the ultimate creator, i.e., human intellect may be impossible, and AI be used as a tool to supplement humans. Every time humans have tried to defy nature’s rules, havoc has happened and thus, the ultimate decision making should remain in the hands of humans. The limits of AI are defined by human intellect while human intellect itself is limitless.