Andrei Kasyanau, cofounder and CEO at Glorium Technologies. Startup advisor and an expert in health and real estate tech.
A study published in 2024 revealed that artificial intelligence (AI) tools can analyze data from individual cells within tumors to forecast whether a patient’s cancer will react to a particular drug. Another study by Google Health showed that their AI model could detect breast cancer from mammogram images with the accuracy of a trained radiologist. Their AI model reduced false positive and false negative rates and outperformed six expert radiologists. However, the researchers noted challenges in the clinical deployment of AI tools, citing issues like integrating it into existing radiology workflows and establishing regulatory approval processes.
AI In Healthcare: The Top Five Challenges
The case of AI in mammography is an excellent example of how implementing AI in healthcare can improve and streamline patient care, diagnostic procedures and more. However, it also shows that the path to successful AI implementation in healthcare is paved with obstacles. Let’s explore the top five challenges hindering the widespread adoption of AI in the healthcare industry.
1. Data Quality And Accessibility
One of the biggest stumbling blocks to real-world AI implementation has been a deceptively simple challenge: access to high-quality data. Experts predict that the CAGR of healthcare data will reach 36% by 2025. This massive increase in healthcare data poses great opportunities. At the same time, it introduces significant challenges for health tech companies adopting AI technology.
The volume of data doesn’t guarantee its quality and availability. Healthcare data is often fragmented across different systems, leading to inaccuracies and inconsistencies. These inaccuracies can adversely affect the performance and reliability of AI models, leading to flawed insights and recommendations. Addressing data quality and accessibility challenges lays the foundation for the transformative potential of AI in healthcare.
2. Implementation Cost
Cleveland Clinic has been at the forefront of implementing AI in healthcare and overcoming data quality issues. By successfully implementing AI solutions to analyze patient flow and optimize scheduling, they’ve reduced wait times by 10%.
Yet, achieving this level of AI integration requires significant investment in technological infrastructure, regulatory compliance, staff training and more. Organizations can mitigate these financial barriers by using trusted, experienced resources that reduce the cost of implementation and avoid wasteful steps.
For example, the appropriate use of RAG architecture can significantly reduce runtime expenses and increase the performance and quality of the output. Additionally, companies often don’t fully grasp the complexity of working with AI, seeing only the tip of the iceberg and missing the hidden costs. Significant additional investment is required to realize its potential.
3. Acceptance Of AI In Healthcare
Beyond the executive level, AI implementation in healthcare affects patients and healthcare professionals in terms of human acceptance and trust issues. A 2023 study found that clinical staff may struggle to accept AI due to the need to learn new skills and take on more complex tasks.
Experts focus on educational efforts and clear communication to address AI adoption concerns. Introducing AI systems that closely match current human activities—like predictive analytics for patient monitoring and virtual assistants for scheduling—demonstrates AI’s reliability and direct benefits, which may enhance overall acceptance in healthcare.
4. Lack Of Experience In Evaluating AI Performance
Without assessing AI systems’ performance, successful implementation is unlikely. This involves examining their integration with existing processes and measuring their accuracy and reliability, user satisfaction and overall impact on business goals. As AI technologies become increasingly integrated into healthcare systems, executives and clinical leaders often lack frameworks for assessing AI’s effectiveness and safety.
This gap in experience can lead to challenges in determining how well AI solutions meet clinical goals, integrate with existing health IT systems, impact patient outcomes and affect the company’s financial performance. This challenge underscores the need for comprehensive training programs, partnerships with AI experts and the development of standardized evaluation frameworks to build the necessary expertise in this rapidly evolving field.
5. Technical Complexity And Skill Gaps
Last but not least, developing and deploying AI solutions in healthcare requires specialized technical expertise in machine learning, data science and software engineering. However, there’s a shortage of skilled professionals with the necessary knowledge and experience to design, implement and maintain AI-driven systems in healthcare settings.
More importantly, there’s a gap in the skills needed to develop strategies for implementing AI from a business value perspective. The role of the chief AI officer (CAIO) has recently been rising to fill these gaps. Regardless, there’s still a significant skills gap in the workforce regarding the understanding and management of AI technologies, according to a 2023 study by GlobalData.
Before implementing AI, experts recommend building trust and confidence in AI across all healthcare workforce levels, including providers, IT staff, executives and administrators. Involving existing staff leverages their familiarity with operations, but incorporating external expertise—such as hiring a CAIO or consulting with outsourcing firms—can help navigate implementation challenges effectively.
Takeaways
The integration of AI into healthcare is met with excitement and caution. Innovative AI applications generate industrywide enthusiasm but encounter challenges like data quality, regulatory hurdles and workforce skepticism. These issues often temper initial optimism, highlighting the complex path to adoption.
Despite challenges, the increasing adoption of AI in healthcare suggests a transformative future. With the market expected to reach $188 billion by 2030, overcoming obstacles with innovation can significantly enhance healthcare quality, efficiency and accessibility, making AI a fundamental improvement rather than a fleeting trend.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
link
More Stories
Mark Cuban discusses ‘f**ked up’ pharmaceutical industry at Penn fireside chat
Sila Realty Sees Uptick in Outpatient Volumes in Healthcare Sector
US senators debut bipartisan legislation to fortify cybersecurity, protect data across healthcare sector