TL;DR:
- Industry experts recommend using existing SDLC models for AI/ML medical product development.
- Eric Henry and Kip Wolf emphasize the need for decisiveness in choosing SDLC models and gaining executive support.
- A compelling business case within the existing framework can pave the way for AI/ML product development.
- People, processes, and technology should be integral considerations in AI/ML software development.
- Agile development methodology should be applied to stakeholder management throughout the process.
- The medtech industry is catching up in adopting process maturity and SDLC models.
Main AI News:
In the fast-paced world of artificial intelligence and machine learning (AI/ML) medical products, industry experts advise software developers to leverage their existing software development lifecycle (SDLC) models. Eric Henry, a seasoned quality and regulatory compliance advisor at King & Spalding, and Kip Wolf, the President of Thaumazo Bioscience Management, shared their insights at the 2023 AI Summit, hosted by the AFDO/RAPS Healthcare Products Collaborative.
Navigating the complex landscape of AI/ML product development often leaves software developers grappling with the choice of SDLC models and obtaining buy-in from company executives. Despite these challenges, Henry and Wolf advocate for decisiveness and action, rather than reinventing the wheel.
“Don’t be cavalier, but if you can find a way to be proactive and maybe a little provocative and find yourself in a schism in your organization where everyone is on hold in analysis paralysis… what we’re saying is just do it,” emphasized Wolf, drawing from extensive experience. He encouraged software developers to recognize the empowerment afforded by their quality system management (QMS) and to utilize these established models confidently.
Henry echoed this sentiment, emphasizing that developers should make compelling business cases for leadership within the existing framework. “If you present a compelling business argument to your leadership that says, ‘Within the existing effective framework that we have that I’m trained in at my organization, I think we can do XYZ, and here are the risks,’ I don’t see any leaders standing in the way,” stated Wolf.
Regardless of the chosen SDLC model, Henry stressed the importance of considering people, processes, and technology throughout AI/ML software development. “Keep in mind, whether you’re developing in weeks or months or years, that we should always engage the people,” he advised. “Make sure you’ve got a good stakeholder map.”
Based on his experience, Henry pointed out that a gap typically exists between developers and stakeholders. Waiting until the end of the development process to address stakeholder needs can be detrimental. In today’s environment, where AI/ML technology is in high demand, the urgency to develop such products is greater than ever. Henry recommended applying the agile development methodology to stakeholder management throughout the process.
Reflecting on his transition into the medical device software space in the early 2000s from other industries using SDLC models, Henry noted that the medtech industry lagged behind by five to ten years in terms of adopting process maturity and SDLC models. However, the industry is now catching up, recognizing the need for agility and efficiency in AI/ML product development.
Conclusion:
Leveraging established SDLC models for AI/ML product development is essential for accelerating market entry. The advice from industry experts underscores the importance of decisiveness, stakeholder engagement, and agility in navigating the evolving landscape of AI/ML medical products. As the medtech industry catches up with process maturity and SDLC adoption, it can better meet the growing demand for innovative AI/ML solutions in healthcare.
