Menu
2024 PIT-UN Summit Voices: AI in Healthcare
from the October, 2024 PIT UNiverse Newsletter
On November 7 and 8, 2024, leaders from across academia, government, civil society and industry will gather at San José State University for the PIT-UN Summit, the marquee annual event for public interest technologists.
In the coming weeks, you’ll hear from our panelists and keynote speakers about the big issues they are excited to discuss at the Summit. In-person registration is now closed, but you can join us virtually through the registration link below.
Join Us Virtually
Register for a virtual pass and gain access to two days of dynamic, interactive keynotes, panels and presentations, plus exclusive content from the PIT-UN Summit Live! studio.
Why is technology advancement, especially applications of artificial intelligence, in healthcare delivery so important for public interest technology? What current events, policies, or big questions are you excited to dive into at the PIT-UN Summit?
Joe Grzywacz of San José State University will lead a session on Public Healthcare Delivery in the Era of AI on Day 1 of the PIT-UN Summit.
Joe Grzywacz, Assoc. Dean and Faculty, San José State University College of Health and Human Sciences
As artificial intelligence increasingly becomes integrated into healthcare delivery, its potential to improve medical diagnostics, treatment plans, and resource management has captured the public’s imagination. However, while AI promises to transform healthcare, it also raises significant ethical, social, and economic challenges.
This is especially true in relation to biases in data that disproportionately affect racial minorities, low-income individuals, rural residents, and other vulnerable populations.
"Access to quality healthcare is a fundamental human right, but innovations like AI constantly reshape the meaning of 'quality," with unanticipated consequences.
Furthermore, issues around affordability, particularly for healthcare delivery systems reimbursed primarily through Medicare and Medicaid, underscore the need for public interest technology to ensure that these innovations serve all members of society equitably.
The Promise of AI in Healthcare
AI applications in healthcare are numerous and growing. Machine learning algorithms can analyze large datasets to help identify diseases early, provide personalized treatment recommendations, and streamline hospital operations. For instance, AI has demonstrated remarkable capabilities in interpreting medical imaging, offering accurate diagnoses for cancer, cardiovascular diseases, and even neurological disorders. These advancements improve patient outcomes and reduce strain on healthcare professionals, allowing them to focus on more complex and human-centered tasks.
AI’s ability to process vast amounts of medical literature and patient data offers the potential for more personalized and precise medicine. Instead of a one-size-fits-all approach, AI can tailor treatment plans based on an individual’s genetic makeup, lifestyle, and medical history. Predictive analytics is another promising area where AI can anticipate disease outbreaks, optimize resource allocation, and enable preventive healthcare measures, thereby reducing the overall cost of care.
However, as AI becomes more embedded in healthcare systems, it’s crucial to examine these technologies’ social impact. Access to quality healthcare is a fundamental human right, but innovations like AI constantly reshape the meaning of “quality” in healthcare with the unanticipated consequences of exacerbating health inequalities rather than reducing them.
What is Public Interest Technology?
5 Keys to Institutionalizing PIT
What is PIT-UN?
The Promise of AI in Healthcare
One of the most pressing concerns with AI in healthcare is biases in the data used to train these algorithms. AI systems learn from historical data, which often reflect societal inequalities related to race, gender, and socioeconomic status. As a result, these biases, if not properly addressed, can perpetuate or even worsen disparities in healthcare outcomes.
For example, studies have shown that Black and Hispanic patients are more often underdiagnosed or misdiagnosed than white patients. If these historical patterns are not corrected in AI training data, they will be replicated and amplified by predictive algorithms. Because the depths and breadth of biases by race and ethnicity in AI models are largely unknown, health inequities could easily be exaggerated.
Similarly, low-income individuals and rural residents, who face increased barriers to accessing healthcare, are at risk for greater health equity by AI if algorithms are trained primarily on data from urban, well-resourced clinics or hospitals and fail to account for the different patterns of disease progression, types of infectious agents, or availability of treatment options in rural clinics.
The Promise of AI in Healthcare
The affordability of AI-driven healthcare technologies creates practical threats to quality healthcare for all. Many AI tools require sophisticated infrastructure, specialized training, and expensive technology to operate, making it difficult for smaller hospitals, clinics, and healthcare providers — particularly those serving low-income or rural populations — to afford these innovations.
This issue is particularly acute in the context of fixed-point pricing systems like Medicare and Medicaid. These government programs provide healthcare to some of society’s most vulnerable members, including elderly and disabled populations and children living in poverty. Healthcare delivery systems serving these groups often struggle to afford the latest medical technologies and to maintain, update, and evaluate whether AI algorithms built into the data platforms are appropriate for their clientele.
High upfront costs of acquiring AI tools, coupled with the ongoing expenses of maintaining and updating these systems, create financial barriers that limit widespread adoption of AI in settings that serve underserved populations. As a result, hospitals in affluent areas are more likely to benefit from AI innovations than hospitals that serve marginalized communities, exacerbating the healthcare inequality already present in society.
The Role of Public Interest Technology
Given these challenges, public interest technology has a critical role to play in ensuring that AI in healthcare is developed and deployed in ways that promote equity and serve the public good.
Mitigate Bias in AI
To reduce bias in AI algorithms and ensure that these tools work effectively for everyone, public interest technologists must:
- Develop tools and strategies to detect and mitigate biases inherent in every data source.
- Advocate for creating diverse datasets representing the rich tapestry of racial, ethnic, economic, and rural groups that characterize modern society.
Create Affordable Technologies
Public interest technologists and policymakers both have roles to play in making AI technology more affordable.
- Public interest technologists must design affordable AI technologies, particularly for hospitals and clinics that serve vulnerable populations.
- Policymakers must address the affordability of AI technologies, perhaps through government subsidies or incentives for facilities heavily reimbursed by Medicaid and Medicare.
- Public interest technologists can help shape policy discussions around affordability, bringing attention to the needs of marginalized groups and ensuring that technology benefits all members of society.
Learn More
Health TechQuity, a new multidisciplinary and cross-sectorial network based in San José, is looking for additional minds, ideas, and hands.
In-person attendees at the PIT-UN Summit can attend “Betwixt and Between: Public Healthcare Delivery in the Era of Artificial Intelligence” from 10 to 10:50 a.m. PST Nov. 8 to learn more.
In the weeks and months to follow, our emerging network will dig in through monthly meetings, regular homework and activities, and quarterly events to press toward our goals. Contact Joe Grzywacz ([email protected]) if you want to learn more or are interested in joining the quest.
Transparency statement: Some ideas and elements of this op-ed were supported through generative artificial intelligence tools.
PIT-UN SUMMIT
The Public Interest Technology University Network (PIT-UN) reimagines its annual convening and launches, PIT-UN Summit 2024, the marquee event for public interest technologists from across academia, government, civil society, and industry.
