How to create value in the lab with artificial intelligence
Laboratories can prepare for the technological shift posed by AI with a better understanding of how the technology will change the clinical lab.
A recent survey conducted by Siemens Healthineers queried hospital and laboratory senior executives as well as lab directors on the future of artificial intelligence in the in vitro diagnostic lab.
The result: 69 percent of respondents expect widespread adoption will occur within four years, and 88 percent identify AI as important for the lab. The survey was sponsored by Siemens Healthineers and conducted by ReRez Research of Dallas last May. It was completed by 200 health system and in vitro diagnostic laboratory senior leaders.
How can laboratories prepare for this significant technological shift? What challenges might labs face, and where will AI be most useful? Let’s review how lab leaders expect AI to change the clinical lab.
Added value of AI
AI is expected to have a broad impact across healthcare in general, as confirmed by 92 percent of respondents. Focusing specifically on the IVD lab, survey participants identified several key areas where they expect AI will be of most value.
There are three immediate areas that AI can impact. First, AI is expected to more accurately predict and optimize turnaround times by analyzing data, instrument performance, workflow, and staff efficiency. Next, AI has the potential to determine the most efficient testing and treatment pathways by analyzing large quantities of lab and population data that humans could not process. Finally, AI will identify health issues and predict disease emergence via insight gleaned from aggregated data.
Additionally, respondents are optimistic that AI will eventually transform population-wide testing and treatment practices. It is expected to define better medical guidelines and patient pathways and improve diagnoses of patients by finding patterns in worldwide in vivo and in vitro diagnostic and patient outcomes data.
AI implementation challenges
Some 54 percent of survey respondents are unsure where to begin with AI. Only 20 percent are at the implementation stage, and just 29 percent are discussing or trialing AI. A lack of familiarity with AI is one of the key challenges to adoption. Moving forward requires AI to be a priority in strategic planning, industry collaboration, and organization-wide changes in hiring and training.
Many hospital and lab executives are concerned about the cost of implementing AI. While the initial implementation may be expensive, the long-term benefits of optimized workflow, improved staff utilization, better treatment algorithms, and increased population health management are expected to outweigh the initial investment.
Some 81 percent of respondents expressed concern about regulation surrounding AI. Much of this concern stems from how AI is currently governed, with AI-related innovations evaluated in the same way as existing applications for new diagnostic tests and patient treatments.
The existing regulatory framework was developed for “static” innovations, requiring each new diagnostic algorithm, treatment regimen or drug to be submitted for regulatory review. But AI systems are inherently dynamic and continuously evolving, with their ability to analyze and adapt to new information and use this data to make correlations, predictions and decisions. Further, most clinical innovations are approved based on well-controlled human clinical studies, and AI-related innovations might not require those protocols.
Recognizing the need for AI in healthcare, regulators are determining how to ensure that AI-developed innovations are both “safe and effective” and respect patient confidentiality. For example, the FDA has introduced the Digital Health Innovation Action Plan to foster innovation while continuing to protect the public. The U.K.’s Committee on Artificial Intelligence has tasked the National Health Service (NHS) with outlining data-sharing plans.
Reimbursement is also critical to the adoption of AI. The potential for reduced costs, increased efficiency, and improved outcomes enabled by AI can be realized only if providers have the financial support to be able to implement it. Changes in the reimbursement policies of both public and private payers are critical to incentivizing adoption of AI-related healthcare advancements.
Preparing for AI: Industry collaboration and data standardization
To prepare for AI, shifts must occur at both the industry and organization level. To function optimally, AI requires large quantities of data, and no single entity will be able to supply an adequate volume or diversity of information. Therefore, the healthcare industry must band together to ensure data sources are standardized, curated, interoperable and accessible.
At the industry level, changes must occur regarding how data is collected and formatted. Today, data is siloed within individual systems and organizations, often in nonstandard formats. This will restrict how AI learns and hinder its effectiveness. Moving toward standardizing data structures and enabling appropriate access to healthcare systems and population data will be a significant challenge. By participating in steering and standards-setting groups, healthcare leaders can help guide the discussion and use this interaction to define the internal strategies for their organizations.
Since the lab generates much of the data that feeds AI’s decision making, lab IT systems will need to be part of an enterprise-wide digital ecosystem that provides an open and secure framework for aggregating, analyzing and operationalizing this data. Establishing a digital roadmap means uniting and educating operational and clinical teams as well as executive leadership. Because few health systems are prepared to pursue these initiatives on their own, they expect to work closely with experienced partners and vendors.
The path to the future with AI
While the predictions of AI’s impact on healthcare have been cautiously optimistic, often expressing concern about robots displacing jobs and de-humanizing patient interactions, in vitro diagnostics testing has much to gain from AI. Labs worldwide have long faced such challenges as shortages of qualified staff, increasing test volumes, and declining reimbursement. AI represents tremendous potential to optimize workflow, improve operational efficiency, and increase workforce productivity.
Lab leaders recognize that AI will enable their organizations to improve patient outcomes by operationalizing massive amounts of valuable data. Despite the many challenges ahead, the executives surveyed welcome the opportunities AI holds, both for patients and their institutions.
The result: 69 percent of respondents expect widespread adoption will occur within four years, and 88 percent identify AI as important for the lab. The survey was sponsored by Siemens Healthineers and conducted by ReRez Research of Dallas last May. It was completed by 200 health system and in vitro diagnostic laboratory senior leaders.
How can laboratories prepare for this significant technological shift? What challenges might labs face, and where will AI be most useful? Let’s review how lab leaders expect AI to change the clinical lab.Added value of AI
AI is expected to have a broad impact across healthcare in general, as confirmed by 92 percent of respondents. Focusing specifically on the IVD lab, survey participants identified several key areas where they expect AI will be of most value.
There are three immediate areas that AI can impact. First, AI is expected to more accurately predict and optimize turnaround times by analyzing data, instrument performance, workflow, and staff efficiency. Next, AI has the potential to determine the most efficient testing and treatment pathways by analyzing large quantities of lab and population data that humans could not process. Finally, AI will identify health issues and predict disease emergence via insight gleaned from aggregated data.
Additionally, respondents are optimistic that AI will eventually transform population-wide testing and treatment practices. It is expected to define better medical guidelines and patient pathways and improve diagnoses of patients by finding patterns in worldwide in vivo and in vitro diagnostic and patient outcomes data.
AI implementation challenges
Some 54 percent of survey respondents are unsure where to begin with AI. Only 20 percent are at the implementation stage, and just 29 percent are discussing or trialing AI. A lack of familiarity with AI is one of the key challenges to adoption. Moving forward requires AI to be a priority in strategic planning, industry collaboration, and organization-wide changes in hiring and training.
Many hospital and lab executives are concerned about the cost of implementing AI. While the initial implementation may be expensive, the long-term benefits of optimized workflow, improved staff utilization, better treatment algorithms, and increased population health management are expected to outweigh the initial investment.
Some 81 percent of respondents expressed concern about regulation surrounding AI. Much of this concern stems from how AI is currently governed, with AI-related innovations evaluated in the same way as existing applications for new diagnostic tests and patient treatments.
The existing regulatory framework was developed for “static” innovations, requiring each new diagnostic algorithm, treatment regimen or drug to be submitted for regulatory review. But AI systems are inherently dynamic and continuously evolving, with their ability to analyze and adapt to new information and use this data to make correlations, predictions and decisions. Further, most clinical innovations are approved based on well-controlled human clinical studies, and AI-related innovations might not require those protocols.
Recognizing the need for AI in healthcare, regulators are determining how to ensure that AI-developed innovations are both “safe and effective” and respect patient confidentiality. For example, the FDA has introduced the Digital Health Innovation Action Plan to foster innovation while continuing to protect the public. The U.K.’s Committee on Artificial Intelligence has tasked the National Health Service (NHS) with outlining data-sharing plans.
Reimbursement is also critical to the adoption of AI. The potential for reduced costs, increased efficiency, and improved outcomes enabled by AI can be realized only if providers have the financial support to be able to implement it. Changes in the reimbursement policies of both public and private payers are critical to incentivizing adoption of AI-related healthcare advancements.
Preparing for AI: Industry collaboration and data standardization
To prepare for AI, shifts must occur at both the industry and organization level. To function optimally, AI requires large quantities of data, and no single entity will be able to supply an adequate volume or diversity of information. Therefore, the healthcare industry must band together to ensure data sources are standardized, curated, interoperable and accessible.
At the industry level, changes must occur regarding how data is collected and formatted. Today, data is siloed within individual systems and organizations, often in nonstandard formats. This will restrict how AI learns and hinder its effectiveness. Moving toward standardizing data structures and enabling appropriate access to healthcare systems and population data will be a significant challenge. By participating in steering and standards-setting groups, healthcare leaders can help guide the discussion and use this interaction to define the internal strategies for their organizations.
Since the lab generates much of the data that feeds AI’s decision making, lab IT systems will need to be part of an enterprise-wide digital ecosystem that provides an open and secure framework for aggregating, analyzing and operationalizing this data. Establishing a digital roadmap means uniting and educating operational and clinical teams as well as executive leadership. Because few health systems are prepared to pursue these initiatives on their own, they expect to work closely with experienced partners and vendors.
The path to the future with AI
While the predictions of AI’s impact on healthcare have been cautiously optimistic, often expressing concern about robots displacing jobs and de-humanizing patient interactions, in vitro diagnostics testing has much to gain from AI. Labs worldwide have long faced such challenges as shortages of qualified staff, increasing test volumes, and declining reimbursement. AI represents tremendous potential to optimize workflow, improve operational efficiency, and increase workforce productivity.
Lab leaders recognize that AI will enable their organizations to improve patient outcomes by operationalizing massive amounts of valuable data. Despite the many challenges ahead, the executives surveyed welcome the opportunities AI holds, both for patients and their institutions.
More for you
Loading data for hdm_tax_topic #better-outcomes...