Researchers at Baylor College of Medicine’s Human Genome Sequencing Center have trained an AI assistant to explain genetic test results to patients. Photo via Getty Images

Artificial intelligence in the health care setting has a lot of potential, and one Houston institution is looking into one particular use.

Researchers at Baylor College of Medicine’s Human Genome Sequencing Center have trained an AI assistant to explain genetic test results to patients. According to findings published in the Journal of the American Medical Informatics Association (JAMIA), the team has developed generative AI to understand and interpret genetic tests. They have also tested its accuracy against Open AI’s ChatGPT 3.5.

“We created a chatbot that can provide guidance on general pharmacogenomic testing, dosage implications, and the side effects of therapeutics, and address patient concerns,” explains first author Mullai Murugan in a press release. Murugan is director of software engineering and programming at the Human Genome Sequencing Center. “We see this tool as a superpowered assistant that can increase accessibility and help both physicians and patients answer questions about genetic test results.”

The initial chatbot training specifically targeted pharmacogenomic testing for statins, meaning a patient’s potential response to cholesterol-lowering drugs, as dictated by genetics.

Murugan explains why they decided to create their own chatbot in the key publication on statin pharmacogenomics was published in May 2022, four months after the training cutoff date for ChatGPT 3.5 in January 2022. Alternatively, her team’s technology uses Retrieval Augmented Generation (RAG) and was trained on the most recent guidelines.

How did the two AI assistants compare? Four experts on cardiology and pharmacogenomics rated both chatbots based on accuracy, relevancy, risk management, and language clarity, among other factors. Though the AI scored similarly on language clarity, Baylor’s chatbot scored 85 percent in accuracy and 81 percent in relevancy compared to ChatGPT’s 58 percent in accuracy and 62 percent in relevancy when asked questions from healthcare providers.

“We are working to fine-tune the chatbot to better respond to certain questions, and we want to get feedback from real patients,” Murugan says. “Based on this study, it is very clear that there is a lot of potential here.” Nonetheless, Murugan emphasized that there is much work still to be done before the program is ready for clinical applications. That includes training the chatbot to explain results in the language used by genetic counselors. Funds from the NIH’s All of Us Research Program helped to make the research possible.

Ad Placement 300x100
Ad Placement 300x600

CultureMap Emails are Awesome

University of Houston med school secures full accreditation

med school milestone

The University of Houston’s Tilman J. Fertitta Family College of Medicine has officially earned full accreditation after launching six years ago.

The designation comes from the Liaison Committee on Medical Education, which granted UH the maximum five-year review period available to a new medical school. To achieve the status, the committee reviewed the Fertitta College of Medicine's facilities, faculty and student outcomes, and curriculum to ensure the college met the highest standards in medical education.

“Full accreditation is not the finish line — it’s the foundation for everything we hope to accomplish in the years ahead,” Dr. Jonathan McCullers, vice president for health affairs at UH and dean of the Fertitta College of Medicine, said in a news release. “It strengthens our ability to attract outstanding students and faculty, expand our partnerships and prepare the next generation of physicians.”

Fertitta College of Medicine has graduated three classes of physicians, and every graduate from its most recent class secured a residency position. Enrollment has expanded from 30 students in its inaugural class to 65 students in its current class.

According to UH, about 70 percent of graduates have chosen to remain in Texas for further training, which is key to addressing the estimated 10,000 doctors needed to meet growing needs.

Fertitta College of Medicine opened for its first students in 2020, and now joins 14 other Texas schools as a fully LCME-accredited medical school. It also joins UTHealth Houston's McGovern Medical School and Baylor College of Medicine as one of three Houston-based LCME-accredited campuses.

The news comes after UH launched its health enterprise initiative, UH Health, in July and broke ground on its $77.5 million Medical Research Building next to the Fertitta College of Medicine last month.

“Building a medical school from the ground up is an extraordinary undertaking, and this milestone reflects the dedication of everyone who helped make it happen,” Dr. Stephen Scott, vice dean for education at the Fertitta College of Medicine, added in the release. “From developing an innovative curriculum to establishing the systems and partnerships that support our students, this has been a true team effort.”

Woodlands startup Mavria raises $130M to mass-produce AI data centers

seed funding

The Woodlands-based Mavria, which mass-produces factory-built modular AI data centers, just emerged from stealth mode with $130 million in funding at a $1 billion valuation.

Mavria raised the money in pre-seed and seed rounds led by Paradigm, an investment firm that concentrates on early-stage startups developing “pioneer” technology. Participants in the funding rounds include AMP PBC, Critical Frontier, Dream Ventures and Liquid 2 Ventures, co-founded by former NFL quarterback Joe Montana.

The company’s inaugural product, Terrain, is a prefabricated data center designed for AI labs, cloud computing providers and data center developers. Mavria says its platform avoids ground-up construction, delivering infrastructure more quickly and at a lower cost.

Mavria tells the Wall Street Journal that the first Terrain unit is being built at a Houston-area factory. The 5-megawatt Terrain system features power distribution, direct liquid cooling, controls and server racks.

“America sets the pace at the intelligence frontier. But the race is moving from the lab into the physical world,” Adam Blake, co-founder and CEO of Mavria, wrote in a blog post.

“Intelligence needs data centers. Data centers need power. Power needs factories, equipment, supply chains, and skilled workers, but much of that infrastructure is still built one custom project at a time, on timelines measured in years.”

Blake and Mavria’s two other co-founders, Nick Poradek and Seth Cohen, most recently worked at the U.S. Department of Energy. There, they focused on reforming the nuclear power sector.

From start to finish, it can take roughly three to six years to build a data center at an estimated cost of $10 million to $37 million per megawatt. Based on those figures, a 100-megawatt data center might cost as much as $3.7 billion.

Here are brief bios of Blake, Poradek and Cohen, who co-founded Mavria in May.

Adam Blake

Blake most recently was a senior advisor to U.S. Energy Secretary Chris Wright. He is co-founder and former CEO of Zego, a provider of energy management technology for property managers. Zego spun out from Brightergy, a commercial solar company. Blake is founder and former CEO of Brightergy.

Nick Poradek

Poradek, chief operating officer at Mavria, was also an advisor to the U.S. energy secretary. He is founder and former CEO of Spring-based Edera Energy, co-founder of The Woodlands-based craft-ice brand Clarity Ice, and founder and former CEO of The Woodlands-based Edera Capital. Poradek sold Edera Energy, a manufacturer of oilfield equipment, to Liberty Energy, a fracking company. Wright is Liberty’s founder and former chairman and CEO.

Seth Cohen

Cohen, chief strategy officer and chief legal officer at Mavria, was head of nuclear law and policy at the Department of Energy. He is also a former NASA advisor and a former attorney at Kirkland & Ellis.

“We believe in rapidly deploying infrastructure as a product, rather than custom-building every project,” Blake said. “We’re bringing together engineering, manufacturing, supply chains and capital so that every project makes us better at delivering the next.”