Using Machine Learning to Identify Patients at Risk of Acquiring HIV in an Urban Health System
In the peer reviewed Journal of Acquired Immune Deficiency Syndrome (JAIDS), the identification of patients who are at risk for contracting HIV and are candidates for preventative measures is discussed as a machine learning model to predict risk for HIV may enhance patient selection for proactive outreach. This important paper’s authors include PCCI’s Arun Kumar Nethi, MS, Data & Applied Scientist, and Albert Karam, VP, Data Strategy Analytics.
BMJ Journal – 32 Diagnostic missed opportunities for diagnosis – addressing the unknown unknown with a safety net team
In this article, co-authored by Albert Karam, Vice President, Data Strategy Analytics at PCCI, missed opportunities for diagnoses are common in diagnostic imaging, particularly when incidental findings require follow-up.
Dallas County celebrates the second anniversary of PASS, a celebrated tool helping asthma sufferers understand their vulnerability
In March of 2023, Dallas County Health & Human Services (DCHHS) announced the unveiling of the Pediatric Asthma Surveillance System* (PASS) that is helping asthma sufferers better understand the harmfulness of their environment to their condition. In the two years since its deployment on the DCHHS website, PASS has positively impacted the residents of Dallas County and has supported the efforts of Parkland Health (Parkland) and DCHHS in reducing poor health outcomes related to pediatric asthma.
“Dallas County is one of the few, if not only, communities in the country that offer this kind of support to residents suffering from asthma,” said Steve Miff, PhD, CEO and President of Parkland Center for Clinical Innovation (PCCI). “We are very proud of the success of this model program showing how properly applied artificial intelligence systems can be used to directly inform and affect those in need in our community.”
PASS is a community-wide effort between Dallas County Health and Human Services, PCCI, and Parkland. Publicly available at the DCHHS website, it has been visited by more than 6,000 Dallas County residents, was honored by the Dallas County Commissioners for its service to asthma sufferers, and described by the Dallas Morning News as “a win for Dallas County.” PASS has also been featured in the highly respected New England Journal of Medicine Catalyst.
“PASS is the direct result of the joint Parkland-DCHHS Community Health Needs Assessment, which identified the need for an actionable dashboard to help stakeholders collaborate using a single source of truth,” said Teresita Oaks, Director of Community Health Programs at Parkland Health.“By integrating health and social data, PASS enables our community to take a proactive approach in addressing pediatric asthma and reducing disparities in care.”
PASS includes a novel, validated Pediatric Asthma Vulnerability Index that integrates health and social data insights to identify communities where children have higher vulnerability to poor asthma outcomes and highlight areas of health disparities. It can also predict the probability of a community-level asthma-related emergency department (ED) visit or hospitalization within three months, by incorporating the effect of 10 community indicators, such as socioeconomic conditions, demographic characteristics, medication use patterns, health services utilization, and environmental conditions, on community-level asthma risk.
“One of the most impactful outcomes of PASS has been the identification of neighborhoods with the highest burden of pediatric asthma,” Philip Huang, MD, MPH, Director/Health Authority, DCHHS. “By leveraging real-time data, PASS has enabled us to pinpoint geographic areas where asthma prevalence and hospitalizations are disproportionately high, so we can more effectively target our resources and interventions to the communities that need them the most.”
The Pediatric Asthma Vulnerability Index and all indicators are categorized as Very High, High, Moderate, Low, or Very Low Risk based on their impact on pediatric asthma vulnerability in the community. On the website dashboard, users can drill down from the Vulnerability Index view to specific indicator views by clicking on specific indicators of interest or navigating to the “Other Indicators” tab. The user can also navigate the map from a ZIP Code view to a census tract view and can select a specific geography of interest for further analysis.
At every geographic level, a risk-driven, color-coded map is displayed in the center with demographic data included below the map. The top indicators contributing to the Vulnerability Index are displayed to the right of the map. To the right of each gauge, an impact score shows the degree to which the indicator contributes to the Vulnerability Index within a selected geographic area. Each gauge has a central black line indicating the vulnerability quintile of the indicator within Dallas County.
“The two years PASS has been running have been a huge benefit to Dallas County as community residents and stakeholders have had single source of precise, timely, and actionable data insights at a highly localized level to identify areas at high-risk for poor asthma outcomes,” said Yolande Pengetnze, MD, Senior Vice President, Clinical Leadership at PCCI. “More importantly, PASS strengthens public health leaders’ ability to plan, design, deploy, and evaluate pediatric asthma programs because they now have a single source of truth as a robust data source.”
Additional indicators are presented under “Other Indicators.” While these indicators were not retained in the Vulnerability Index prediction model, they provide the user with additional, actionable insights into drivers of asthma vulnerability in the community, such as the proportion of the population reporting tobacco smoking in the neighborhood.
“In this example,” said Dr. Pengetnze, “If a child with poorly controlled asthma lives in a census tract with a high prevalence of smokers and low controller medication use, the clinical provider might initiate caregiver education about secondhand smoking and medication adherence. Additionally, public health entities might address access to medication and smoking cessation programming at the community level.”
The PASS supports Parkland’s and DCHHS’ communitywide pediatric asthma programming and has been has also been used to engage parents or caregivers to participate in Parkland’s educational, interactive, pediatric asthma text-messaging program designed to help parents manage their child’s asthma.
*This work is funded in part by Lyda Hill Philanthropies. Lyda Hill Philanthropies encompasses the charitable giving for founder Lyda Hill and includes her foundation and personal philanthropy. Our organization is committed to funding transformational advances in science and nature, empowering nonprofit organizations and improving the Texas and Colorado communities. Because Miss Hill has a fervent belief that “science is the answer” to many of life’s most challenging issues, she has chosen to donate the entirety of her estate to philanthropy and scientific research. For more details visit lydahillphilanthropies.org.
Five years after the chaos of COVID-19 comes a new era of AI innovation
By Steve Miff, PhD, President & CEO of PCCI
Five years ago this week, the COVID-19 pandemic threw our world into chaos. But that chaos also sparked an opportunity and a drive to accelerate innovation, through leveraging artificial intelligence with clinical knowledge in unprecedented ways. The results of these Covid-driven efforts have led to a fuller understanding of our community’s health, enabling the initiation of care further upstream and enhancing management of resources and costs in ways benefiting both patients and providers.
I can still remember that week five years ago. I had my bags packed and flight booked to travel to one of the biggest healthcare technology conferences, HIMSS, in Orlando, Fla., when the word came down. The government declared a full lockdown of the country thanks to the pernicious COVID-19 outbreak that the World Health Organization had declared a full-blown pandemic.
Sadly, the cost of COVID-19 in lives— 1.2 million deaths in the U.S. alone— was devastating and the reverberations of the pandemic still affect our daily lives.
However, the chaos and heartache also stimulated a new mindset for collaboration. While the lockdown closed our office doors, it opened new windows to collaborate with other stakeholders across our community. Our teams immediately partnered with Dallas County Health and Human Services (DCHHS), Parkland Health (Parkland) , and other local providers and governmental agencies to help support our North Texas public health leaders and families manage through the ever-changing nature of the COVID-19 pandemic. We deployed a transparent, data-driven set of analytics to inform the dynamic allocation of resources, guide triage protocols in emergency rooms, identify COVID-19 community hotspots, and predict rising waves of hospitalizations and capacity challenges. These collaborations would evolve into support for testing, vaccination distribution, and measuring of community-wide immunity to the infection.
While many of the solutions developed during COVID-19 were industry firsts, we continue today to leverage and expand on many of these same novel applications in AI and non-medical drivers of health (NMDoH) analytics.
Vulnerability Index
The development of the Covid Vulnerability Index and geo-mapped hotspotting created the dynamic dashboards that guided the local pandemic response, including the placement of the testing, and later vaccination sites. This novel approach later expanded into what’s now the Community Vulnerability Index, an industry leading NMDoH analytics method modeling community and neighborhood barriers to health and wellness.
Proximity Index
The Covid Proximity Index was used to predict the risk of infection daily at the individual level using advanced geomapping and data science techniques to guide personal awareness and behaviors, county-wide contact tracing, healthcare provider virtual care scheduling, and ED triage. We are so proud of the United States Patent for AI driven proximity index system and methods (US 12,087,449 B2) that was awarded to PCCI late last year. We’re now leveraging these methods alongside mobility data to understand access to vital services and placement of new access points in the community. The Proximity Index was also featured in the highly respected peer-reviewed New England Journal of Medicine Catalyst in a paper authored by Parkland’s CEO Frederick Cerise, MD and others.
Community Protection Modeling
The national Covid Community Protection Dashboard was launched in collaboration with the Institute for Healthcare Improvement (IHI), Civitas Networks for Health, and Cincinnati Children’s Hospital to model local behaviors and immunity patterns and help communities manage through new Covid variants and waves. We learned that the dynamic nature of modeling factors and a community’s protection depends on the characteristics of the most prevalent current variant and the immunity from prior vaccination and infections, adjusting for the time elapsed from the most recent immunizing event. These innovations translated into the development and deployment of the Pediatric Asthma Surveillance System (PASS), the soon to be released Diabetes/Hypertension Surveillance System, and the development of a novel Maternal Health Surveillance System. Each of the programs are (or will be) in use in Dallas County and provide life-saving support to its residents.
Better Preparedness for Future Crises
The pandemic brought Parkland, our main public health system, closer to DCHHS and many other public health leaders, working together with PCCI to build new policies and procedures to manage the COVID-19 emergency and future public health crises. And have no doubt about it, it is only a matter of when, not if, we will have more COVID-19-like outbreaks.
In the big picture for public health innovation, COVID-19 created a necessity for us to be innovative and try new ways to solve extremely difficult problems. This same way of thinking is helping us today as we tackle ongoing health challenges through improved collaborations and tools. For example, our Community Vulnerability Compass (CVC) is a tool that can provide deeps insights into the complexities of societal challenges to health, access, and community well-being affecting our neighborhoods throughout Texas. It can tell you, for example, what local daily challenges our residents face down to the block level. The CVC has been adopted by a variety of organizations around the state, led by the United Way of Metropolitan Dallas through its Data Capacity Building Initiative that, within five years, aims to equip over 200 community-based partners with robust data insights from the CVC.
And as previously mentioned, the PASS system is celebrating its two-year anniversary of providing residents of Dallas County unprecedented understanding of their vulnerability to asthma-related risks. PASS is a community-wide effort between the county, PCCI, and Parkland and is publicly available at the DCHHS website. It has been visited by thousands of Dallas County residents and was honored by the Dallas County Commissioners for its service to asthma sufferers and was described by the Dallas Morning News as “a win for Dallas County.” PASS has also been featured in the New England Journal of Medicine Catalyst.
The pandemic did put a damper on at least one of our innovations. The book, “Building Connected Communities of Care,” was due to drop via a major release at the Orlando HIMSS convention in March, 2020. The celebration and release was changed and conducted virtually and digitally. This guidebook helps communities and public health leaders create holistic community networks that support any number of health management issues. Thanks to its insights and easy-to-follow steps, “Building Connected Communities of Care” has become an invaluable resource for many public health leaders across the country. It was also featured in the New England Journal of Medicine Catalyst for its insights it offers in handling community healthcare crises.
The COVID-19 pandemic was not easy on anyone, but we are wise to remember what happened during those tough times and leverage the innovations and progress towards bigger and broader ongoing impact. We will continue to leverage the critical clinical and community health lessons COVID-19 taught us in innovative programs incorporating the newest AI technology. As a community, while we remember the terrible price COVID-19 cost our friends and families, we should feel some level of optimism that the next time we will be better prepared and stronger together.
Pre-pandemicPost pandemic
About Steve Miff
Dr. Steve Miff is the President and CEO of Parkland Center for Clinical Innovation (PCCI), a leading, research non-profit, artificial intelligence and cognitive computing organization affiliated with Parkland Health, one of the country’s largest and most progressive safety-net hospitals. Spurred by his passion to use next-generation analytics and technology to help serve the most vulnerable and underserved residents, Steve and his team focus on leveraging technology, data science, and clinical expertise to obtain unique non-medical-determinants-of-health data and incorporate those holistic, personal insights into point-of-care interventions.
PCCI Experts Set to Deliver Major Presentations to the nation’s healthcare leaders at HIMSS25
Starting March 5, PCCI experts are joining leaders from Parkland Health to present cutting-edge AI healthcare programs in prime spots at 2025 HIMSS Global Health Conference and Exhibition in Las Vegas, the epicenter of healthcare innovation.
The presentations are:
Creating a Large Language Model to Catalog Important Radiologist Recommendations
Wednesday, March 5, 3:15 PM to 4:15 PM PACIFIC
Speakers
Alex Treacher, PhD, Senior Data and Applied Scientist – PCCI
Albert Karam, Vice President, Data Strategy and Analytics – PCCI
Brett Moran, Chief Health Officer – Parkland Health
Abstract
Medical errors, the third leading cause of death in the U.S., include wrong or delayed diagnoses, causing more serious harm than any other type of medical error. Delayed or missed opportunities for diagnoses (MOD) are particularly common in diagnostic imaging, where incidental findings often require further evaluation. At Parkland Health, a major safety-net public health system, 1.7 percent of all CT and MRI studies involve such findings. To address this, a large language model (LLM) is developed that identifies and flags delayed surveillance recommendations from radiologists’ interpretations. These delayed recommendations result in MODs 17 percent of the time. This LLM has been integrated into the electronic health record (EHR) of Parkland Health, enabling centralized management and navigation of these cases. Our results demonstrate 95 percent accuracy in identifying imaging that requires follow-up based on physician notes and 85 percent accuracy in determining the appropriate timing for follow-up. This work outlines the process, development, tools, current performance, and future plans for building an automated system to enhance image surveillance and mitigate MODs in diagnostic imaging.
Yusuf Tamer, PhD, Principal Data and Applied Scientist- PCCI
Michael Lane, Senior Vice President, Chief Quality and Safety Officer – Parkland Health
Teresita Oaks, Director, Community Health Programs – Parkland Health
Abstract
In Dallas County’s safety-net population, an AI/machine learning-driven unsupervised clustering algorithm identifies clusters of diabetic and hypertensive patients with a combination of social and clinical risk factors associated with suboptimal quality of care (e.g., inadequate of Hemoglobin A1C monitoring) and poor disease control. Clusters analyses uncover underlying, actionable risk drivers such as criminal justice involvement and immigration concerns that require innovative, culturally-responsive approaches for a sustainable engagement of these vulnerable populations into effective preventive care. Additional in-depth analyses identify missed and potential opportunities for care engagement that inform innovative workflow modifications leveraging traditional (e.g., EHR-based standing orders) and nontraditional (e.g., telehealth modalities and mobile units) approaches to effectively engage and support these vulnerable populations and improve health quality, outcomes and equity countywide. The data sets and analytical approaches are scalable and replicable to other vulnerable populations nationwide.
PCCI Sets Friday, Aug. 8 for Sachs Summer Scholars Innovation Showcase
PCCI’s Sachs Summer Scholars Innovation Showcase has been set for Friday, August 8, 2025 from 9 a.m. to Noon in room 101 at Pegasus Park. PCCI’s Sachs Summer Scholars is an internship program that advances women in STEM and data science. This end-of-term showcase spotlights the work of PCCI’s elite group of interns who worked side-by-side with our clinical, technical and data science experts on real projects of substantial impact. The intern showcase will also provide an inside view into the diverse and innovative work PCCI does every day in support of underserved communities.
The Sachs Summer Scholars intern showcase demonstrates PCCI’s ongoing commitment to supporting women in the data science and technology industry. This program has become one of the premier internships in North Texas that immerses students in meaningful, real word projects with measurable impact. This includes providing each intern direct experience with innovative healthcare case studies, groundbreaking and responsible artificial intelligence applications and social determinants of health projects. The aim is to support and promote practical applications of analytics, computing, and data science all while advancing the spirit of mentorship and advancement of female students.
At the research showcase, the interns will highlight their work on core projects that include core, innovative projects PCCI created that include advanced AI and non-medical determinants of health analytics.
To learn more about PCCI’s programs and the impact on the community, take a look at last year’s program video.
Please, mark this exciting program on your calendar now!
Parkland Center for Clinical Innovation releases its 2025 annual impact report highlighting recent accomplishments
Parkland Center for Clinical Innovation (PCCI), a non-profit innovation and research institute for advanced data science and non-medical drivers of health (NMDoH) innovation, has released its annual impact report detailing its successes and activities from 2024.
“The Annual Impact Report features some of the outstanding results the team at PCCI had in 2024,” said Steve Miff, PhD, President and CEO of PCCI. “The report highlights our expanding AI in medicine for underserved populations solutions through novel programs and partnerships in North Texas and around the state.”
These AI-focused programs featured in the report include PCCI’s leadership in advanced AI and NMDoH analytics for clinical decision support and population health and surveillance systems. The report also features the national activities and engagements of PCCI experts and leaders who drive the organization’s programs and mission to support the underserved.
“I am very excited to share the accomplishments and leadership of our team and our partners,” Miff said. “Our mission of driving innovation in AI for underserved populations is at the heart of everything we do. This past year our team built and deployed novel and scalable solutions and contributed to the emerging knowledge and guidelines for ethical, responsible and equitable uses of AI in healthcare.”
The Parkland Center for Clinical Innovation (PCCI) is a not-for-profit, mission-driven organization with industry-leading expertise in the practical applications of artificial intelligence, machine learning and NMDoH data modeling to address the needs of vulnerable populations. PCCI started as a department within Parkland Health and was spun out as an independent organization in 2012. PCCI strives to leapfrog the status quo by harnessing the transformative potential of data. Our unique capabilities allow us to provide innovative, actionable solutions that more effectively identify needs, prioritize services, empower providers, and engage patients.
PCCI named to Dallas 500 list of leading organization in North Texas
D CEO’s signature special publication, the Dallas 500, features profiles of the most influential North Texas executives in more than 50 industry categories. The 2025 book, our 10th edition, includes well known, emerging, and behind-the-scenes leaders who make the regional economy tick. Editors make selections after months of research and hundreds of personal interviews. Along with providing business advice and industry insights, this national award-winning book gives readers a glimpse at the human side of DFW’s most powerful leaders.
Joining this list for the fifth year in a row is PCCI. CEO Steve Miff, PhD, who attended the annual event recognizing these outstanding leaders.
Disparities Next Door: PCCI’s Community Vulnerability Compass spotlights areas of increased vulnerability across Dallas County
Dallas, Texas(Sept. 19, 2024) – Researchers at Parkland Center for Clinical Innovation (PCCI) have highlighted areas of increased vulnerability within Dallas County after conducting an analysis using PCCI’s Community Vulnerability Compass (CVC). The findings include a number of pockets of highly vulnerable areas within many low vulnerability ZIP Codes.
PCCI’s CVC analyzes, at the ZIP Code, census tract, and block-group level, where a community’s most vulnerable residents live and the root cause factors limiting these residents’ ability to thrive. The CVC analyzes 26 clinical and socio-economic indicators that reveal the health, resiliency, and economic vibrancy of neighborhoods. CVC’s analyses provide true, holistic pictures of who needs the most assistance and where to find them, enabling proactive support of those in need.
It is well-documented that many ZIP Codes in the south and southeast areas of Dallas County (see image 1) are home to long-standing socio-economically disadvantaged populations and have the highest vulnerability rating. However, PCCI’s analysis underscores what Parkland and other officials have long stated: that in addition to the south and southeast regions, there are other ZIP Codes in Dallas County with low vulnerability but are home to highly vulnerable block groups. Specifically, these include 18 Dallas County ZIP Codes with an overall very low or low vulnerability rating that contain within them 48 block groups with high or very high vulnerability ratings. The 18 ZIP Codes have a total population of 590,971, with 72,954 (12.3%) of their residents living in these very high vulnerability blocks groups.
“In some cases, we find that a very high area of vulnerability is across the street from a very low area of vulnerability.“
Steve Miff, phd, President and CEO of PCCI
The “very high” designation indicates that these neighborhoods are in the top 20% of vulnerability when compared to the rest of Dallas County. CVC vulnerability groupings are localized to the Dallas County geography and broken into quintiles and clustered into very low, low, moderate, high and very high areas.
“For good reason we focus a lot of our efforts on supporting the underserved communities in south and southeast Dallas County, but CVC empowers us with refined geographic precision to examine our broader community and uncover data that provides a high-resolution understanding of social vulnerability factors within other micro-geographies in our community,” said Steve Miff, PhD, President and CEO of PCCI. “In this examination, CVC highlights areas of deprivation that may be masked by gentrification and/or historical surrounding prosperity. In some cases, we find that a very high area of vulnerability is across the street from a very low area of vulnerability.”
For example, ZIP Code 75206, located in the Greenland Hills area that runs along Highway 75 and is bordered by Highway 12 and Skillman Avenue, is rated very low vulnerability by the CVC, but includes four block groups that are rated by the CVC as high and very high vulnerability. In this ZIP Code, which has a total population of 38,209, there are 2,661 (7%) who experience high or the very highest levels of vulnerability as measured by the CVC.
One ZIP Code of note, 75254, located north of the Galleria between Dallas North Tollway and Coit Road, has an overall low vulnerability rating, but includes 8 high or very high vulnerability blocks. Of this area’s total population of 24,047, 12,522 (52.1%) residents experience high or very high vulnerability. (see Image 2)
Additionally, ZIP Code 75230, which is considered to be an economically advantaged area in North Dallas between Walnut Hill and Interstate 635, includes a very high vulnerability block group.
“The lesson is that we need to look beyond our expectations and understand that disparities and vulnerability are all around us no matter how prosperous a given area may appear,” Miff said. “We hope this will bring an additional level of high-resolution, hyper-localized understanding of social vulnerability factors and empower leaders across all communities with relevant information to improve and sustain the lives of all Dallas County residents. North Texas has outstanding public health leaders who are doing a great job supporting vulnerable populations throughout the county and we hope this helps them as well as policymakers, community-based organizations and others who are vested in helping mitigate disparity.”
IMAGE 1
IMAGE 2
About PCCI
The Parkland Center for Clinical Innovation (PCCI) is a not-for-profit, mission-driven organization with industry-leading expertise in the responsible application of artificial intelligence, machine learning and NMDOH data modeling to address the needs of vulnerable populations. PCCI started as a department within Parkland Health and was spun out as an independent organization in 2012. PCCI strives to leapfrog the status quo by harnessing the transformative potential of data. Our unique capabilities allow us to provide innovative, actionable solutions that more effectively identify needs, prioritize services, empower providers, and engage patients.
AI Is Working For Us Now, Helping Us Heal and Prevent Harm
By Steve Miff, PhD, President & CEO of PCCI
We can’t let artificial intelligence (AI) continue to be defined by science fiction, deep fakes, and plagiarized term papers, because today―right now― responsible uses of AI in medical care are helping prevent harm, improve clinical decisions, and reduce costs of care. Is there too much unjustified hype and exuberance over AI? Absolutely! Are there unsubstantiated claims and sometimes dangerous promises? Unfortunately, yes. However, there are also many examples of ethical, responsible and equitable uses of AI that are streamlining operations, improving efficiencies, and yes, even helping physicians and clinical teams augment their insights and expertise to save lives.
AI is also not new. While the history of artificial intelligence (AI) can be traced to antiquity hypothesized by master craftsmen, modern AI really started with the invention of the programmable digital computer based on mathematical reasoning in the 1940s. In the last decade, the true potential (and risk) of AI and Machine Learning algorithms have accelerated and can simply be classified into predictive, prescriptive and generative AI/ML, with the latter creating the most excitement and controversy over the last 12-24 months.
At PCCI, we have been researching, developing, deploying and testing applications of AI and machine learning for a decade with a focus on applying these innovative concepts to those who serve the most vulnerable individuals and communities. With every project that we’ve embarked on, we’ve followed sound scientific principles and a prescriptive approach to development, testing and implementation that keeps the clinician in full control and with the utmost transparency, and the patient at the center. While the AI part is sexy, the governance, processes, and discipline you put around it makes it reliable, responsible and effective. We developed and optimized our own processes over type and rely on a core set of core principles:
There needs to be a clearly articulated problem and need that the AI is trying to solve/augment. If you don’t have a real problem to solve, it’s just cool math.
A multi-disciplinary team that includes an engaged and passionate lead clinician is required from day 1. The front-line staff already have the knowledge and intuition of what is needed and what could work.
More than half the time is spent on ensuring data quality and staging – researching, curating, studying, aggregating, ingesting, validating, augmenting and analyzing data from multiple sources is the lifeblood of any successful model. The old saying of “junk in, junk out” is even more crucial with AI.
Focus just the data required for the initial development, but understanding the required data latency, refresh frequency, storage and compute requirements to understand deployment requirement and cost to operationalize and sustain.
You need your own AI sandbox and your EHR is not it. PCCI developed Isthmus™, a secure, digital data environment leveraging established cloud technology and optimizing open-source tolls for end-to-end, secure data orchestration, modeling and deployment.
Data security is paramount, hence why PCCI deploys Isthmus™ within the firewalls of a health system/payer/provider (no protective data is moving out)
There is a prescriptive stage-gate development and deployment process that requires patience and time:
Build models leveraging historical data; T
Test the preferred model with a reserved data set
Optimize the performance of the model leveraging clinician input to optimize how the model is to perform and be used in practice
Deploy the model with real-time data and run it silent mode. No decisions are made using the model, but the performance, stability, expected output is evaluated and monitored. Evaluate for equity and expected performance on the respective patient population. This could take months, or years. If you’re trying to predict a rare event, it will require extensive time to ensure adequate amount of time and data is being processed. To ensure proper evaluation for the PCCI Parkland Trauma Mortality (PTIM), we went into a full silent mode on every patient every hour for 3 full months before we moved to provider facing production.
Design the output and display of the model and integrate into the EHR or case management system where clinical teams do their work (no extra clicks or sign-on to get to the information). Integrate and deploy systems that give clinical teams insights into the metrics behind the model. PCCI developed and deployed a web-based interface called Islet™ that provides real-time insights into a model. This is critical not only to ensure transparency to clinical teams, but it’s extremely valuable to point to the top factors and the data driving the algorithms to point to the factors that can be actioned
Design a stages go-live deployment process and incrementally expend teams and/or department while evaluating each deployment
Ensure a robust and coordinate response process to address off-cycle failures. While the models are designed to augment clinical decisions, the power of AI is tangible and useful and clinical teams rely on their assistance. It’s like getting used to driving on automatic transmission car and then having to go back to driving stick. You can still drive, but it’s not as easy.
This is not a one-and-done. Far from it. Develop and stick to a regular cycle of model evaluation, testing and updates
At PCCI speed to market is a self-imposed urgency and profits are not a measure of success. What our AI and machine learning systems aim to do is help providers make more informed clinical decisions, optimize resources to expand access, offer synthesized data insights that help improve health outcomes, and educate our region’s health leaders and residents.
Some of these AI predictive model applications came to life during the COVID-19 pandemic. We were able to use the established infrastructure, processes and expertise to create several programs that made significant impact on how Dallas County managed through the pandemic. A few of these AI and predictive models included:
Capacity Forecasting Model. Leveraging an AI algorithm based on geo mapping, we helped Parkland manage resources by creating an effective model that helped forecast possible needs for hospital beds, ICU, staffing needs, and development of workflows.
Proximity Index (PI). This AI predictive triage model empowered frontline clinical and population health teams to proactively manage high-risk patients and reduce potential COVID spread from asymptomatic patients in hospital and community settings.
Vulnerability Index (VI). This AI predictive model allowed public health leaders to understand the hardest hit populations in Dallas County and then drive community response.
Beyond COVID, PCCI’s AI-driven clinical decision support and population health programs guide policy and direct resources where they are needed most. PCCI’s team members have extensive experience deploying, testing, and validating a wide array of multi-institution, algorithm-driven protocols in clinical settings. Examples include:
Parkland Trauma Index of Mortality (PTIM)
The PTIM machine learning algorithm is the only known model that uses electronic medical record data to predict ―every hour ― 48-hour mortality during the first 72 hours of hospitalization, thus evolving with the patient’s physiologic response to trauma. Over a one-year period, PTIM has correctly identified 89% of the high-risk trauma patients and 92% of the low-risk trauma patients.
Inpatient (IP) Medicine Sepsis AI Program
The sepsis management system is real-time predictive model that identifies and triggers interventions of patients who are at high risk for sepsis both upon presentation and after being admitted to the hospital. The IP Sepsis model is unique as it enables Parkland to identify cases of sepsis that were not present on admission.
STI-HIV PrEP Model
The model will be able to predict those at-risk for hiv infection and eligible for prep in order to reduce HIV transmission. The PrEP predictive model is being developed using data from Parkland’s EHR, pertinent public open-source data on NMDOH, and DCHHS STI data. The model will be able to predict those at-risk for HIV infection and eligible for PrEP to reduce HIV transmission.
Preterm Birth Prevention (PTB) AI-driven Program
To better serve pregnant women, PCCI and the Parkland Community Health Plan developed and implemented an innovative Preterm Birth (PTB) Prevention program that uses a machine learning algorithm, healthcare data, and Non Medical Drivers of Health to identify pregnant women who are at a higher risk of PTB within vulnerable communities. The PTB program consistently risk stratifies about 13,000 pregnant women from vulnerable communities per year.
Pediatric Asthma Surveillance System (PASS)
PCCI developed the PASS to help prevent pediatric asthma from harming children in Dallas by providing a single source of community insights to identify children that have higher vulnerability to poor asthma outcomes and highlight areas of health disparities. The PASS interactive dashboard has been visited by nearly 3,000 Dallas County residents.
Many of these programs have proven to be cost effective, scalable and successful. We will continue to invest and explore new ways to apply AI solutions in collaboration with some of the top healthcare providers and public health organizations in the nation.
Whether everyone knows it, understands it or even likes it, AI is here to stay and is exploding in healthcare, and making a huge difference in our lives. At PCCI we will continue to focus on applying and localizing these powerful concepts with those who serve the most vulnerable individuals and communities. That’s our mission and focus and will remain that way. This specifically means continuing our work with Parkland Health (Parkland), PCHP, Dallas County Health & Human Services Department (DCHHS), the United Way of Metropolitan Dallas, and expanding our partnerships with other similar organizations in San Antonio, Austin, Atlanta and other communities.
Because of the responsibility that has been entrusted in us, we feel the urgency and responsibility to not only continue to learn from others pioneering responsible applications of AI, but contribute to the knowledge, voice and narrative of AI nationally. That’s why we’re appreciative to be included and actively participate in national AI committees and collaboratives, such as the Health AI Partnership advisory board, the National Academy of Medicine AI Adoption and Code of Conduct committee, the White House Health AI commitments executive group, as well as several other national meetings and forums. We bring the important perspective and the unique challenges of leveraging these new and exciting capabilities in an equitable was and make these innovations accessible to underrepresented and underserved communities.
For many the promise of AI is about what’s on the horizon, but I see the promise of AI delivering results and innovations today. With responsible, equitable, reliable and transparent applications, the future is now.
About Steve Miff
Dr. Steve Miff is the President and CEO of Parkland Center for Clinical Innovation (PCCI), a leading, non-profit, artificial intelligence and cognitive computing organization affiliated with Parkland Health, one of the country’s largest and most progressive safety-net hospitals. Spurred by his passion to use next-generation analytics and technology to help serve the most vulnerable and underserved residents, Steve and his team focus on leveraging technology, data science, and clinical expertise to obtain unique non-medical-determinants-of-health data and incorporate those holistic, personal insights into point-of-care interventions. Steve was the recipient of the 2020 Dallas Business Journal Most Inspiring
