Explore recent discoveries made by innovators at Dartmouth Health and the Geisel School of Medicine at Dartmouth that could transform health and medicine.
A Kill Switch Hidden Inside Every Cell
Every cell carries the design for its own destruction. The process—apoptosis—is how the body quietly removes damaged or dangerous cells. When that mechanism breaks down, cancer cells can proliferate. Understanding how that kill switch works could help answer one of the biggest open questions in oncology today.
Postdoctoral researcher Sukrut Kamerkar, PhD, and Henry Higgs, PhD, professor of biochemistry and cell biology, both at the Geisel School of Medicine at Dartmouth, are already on it. They discovered that a protein called LACTB acts as a regulator deep inside the mitochondria, reshaping its inner membrane to release the molecules that pull the trigger on cell death. When Higgs and Kamerkar reduced LACTB levels, cells stopped dying on cue. When they boosted LACTB levels, cells died faster. Their results were published in Science Advances in December 2025.
Their findings may have unlocked a layer of control that nobody knew had been governing cell death all along. If many tumors resist treatment because their cells won’t die when they should, LACTB may be the very lever that cancer has been secretly hiding behind.
An AI Teaching Assistant That These Students Actually Trust
Medical education has a scale problem. Even at well-resourced institutions, educators can only reach so many medical students at once.
But AI tools could alleviate this scale problem, according to a study published in November 2025 in npj Digital Medicine. Thomas Thesen, PhD, associate professor of medical education at the Geisel School of Medicine at Dartmouth, and co-author Soo Hwan Park, MD, MED ’25 tracked how 190 medical students used NeuroBot TA, an AI teaching assistant built by Thesen and Park to answer students’ questions around the clock using only vetted course materials. Unlike chatbots prone to generating convincing but inaccurate information, NeuroBot TA pulled exclusively from textbooks, lecture slides, and clinical guidelines. Students overwhelmingly trusted it more than general AI tools, and nearly half found it a useful study aid.
Thesen’s lab has a history of these innovations. Its AI Patient Actor platform, developed in 2023, simulates patient conversations to help medical students sharpen diagnostic and communication skills, and is now used in medical schools across the United States and abroad. Together, the tools point toward a future where AI doesn’t replace medical educators but broadens their reach far beyond the classroom.
Tick-Tock: Cutting Time to Lyme Disease Diagnosis
Lyme disease can be notoriously difficult to diagnose quickly and accurately. Most standard tests detect the body’s immune response to infection rather than the bacterium that causes it—and immune responses can take time to develop and vary greatly between individuals. But researchers at Dartmouth Health’s Dartmouth Hitchcock Medical Center (DHMC) have developed a new tool that directly detects Borrelia burgdorferi, the bacterium responsible for most Lyme disease cases in the U.S., much earlier in the course of infection.
The team, led by Guohong (Grace) Huang, PhD, in the Laboratory for Clinical Genomics and Advanced Technology at DHMC, presented the work at the Association for Molecular Pathology’s 2025 Annual Meeting in Boston. Using droplet digital polymerase chain reaction (ddPCR) technology, the researchers created three assays that can identify the DNA of Borrelia species—including one specific for B. burgdorferi—with very high sensitivity, detecting as few as 5 to 10 bacterial cells in a sample. This represents a substantial contribution to Lyme disease patient care, as conventional antibody tests can take weeks to turn positive and often cannot distinguish between past and active infections. The team is currently investigating the potential use of this test with skin biopsies of lesions that could be associated with Lyme disease in patients who do not have the typical “bull’s-eye” rash and have not yet developed detectable antibodies.
Early Gut Microbiome Differences in Babies With Cystic Fibrosis
Researchers at the Geisel School of Medicine at Dartmouth have discovered that infants with cystic fibrosis (CF) show distinct differences in their gut microbiome within the first few months of life compared with healthy infants, and those changes may influence how the disease progresses.
In a study published in mBio in March 2025, the team analyzed stool samples from infants with CF and infants without CF. Although it is best known for damaging the lungs, CF also disrupts digestion and nutrient absorption early in life.
Led by Benjamin Ross, PhD, assistant professor of microbiology and immunology at Geisel, and clinician-scientists Juliette Madan, MD, MS, RES ’21 MED ’00 and Julie Sanville, DO, of Geisel’s Department of Pediatrics and who both practice at Dartmouth Health, the team found that babies with CF experienced delayed and disrupted development of key gut bacteria involved in digestion and immune maturation. Notably, these differences appeared before gastrointestinal symptoms typically emerge.
The findings suggest microbiome disruption may be an early feature of CF itself, pointing to possible early-life interventions, including targeted microbiome therapies or tailored nutrition strategies to improve growth and health outcomes.
The Implant That Records the Brain—While Regrowing the Skull
Foreign objects don’t usually do very well in the brain. Even electronic implants, placed against neural tissue, can risk immune response, scarring, and structural damage to the very organ they are meant to help. Such problems limit how long devices can stay and how well they perform over time.
But Dartmouth researchers have now built an implant that works with the body instead of against it. In a paper published in Advanced Materials Technologies in February 2026, Alexander Boys, PhD, assistant professors at the Thayer School of Engineering at Dartmouth, and Katie Hixon, PhD, clinical assistant professor of orthopaedics at the Geisel School of Medicine at Dartmouth and assistant professor of engineering at Thayer, combined their labs’ distinct expertise: Boys’ thin-film neural recording arrays were integrated with Hixon’s bone-regenerating scaffold material, creating a single device that records brain activity while simultaneously regrowing the skull around it. In laboratory testing, the implant produced no adverse immune response and stimulated bone formation comparable to the scaffold alone.
In the future, such a device could support brain-computer interfaces, enable real-time pain monitoring, and extend the lifespan of neural implants used to treat epilepsy and Parkinson’s disease, conditions where durable, long-term access to the brain is everything.