In April 2002, in a lab on the north end of the University of Michigan campus, a graduate student named Zhen Xu plugged her ultrasound transducer into an amplifier nobody else wanted to use because it was too loud. She aimed it at a pig heart standing in for human tissue and switched it on. When she looked, there was a clean hole punched straight through the muscle. Nothing had touched it. No blade, no heat, no incision. Just sound, tuned until it tore the tissue apart from the inside.

Her advisor, the late Charles Cain, gave the effect a name the next year: histotripsy, from the Greek for “soft tissue” and “breaking.” For most of the two decades since, it was a curiosity that lived in acoustics journals and a single Michigan lab. Then, over the past three years, the patents started arriving in a rush.

Sixty percent of the record, in thirty-six months

Across the full US grant record, 47 issued patents mention histotripsy by name. Twenty-nine of them, nearly two-thirds, were granted after January 2023. The annual count went from four grants in 2024 to twelve in 2025, and eight more have already issued in the first half of 2026. On the current pace this year will set another record.

That is a small pile of paper compared with the thousands of grants that land every week in batteries or semiconductors. But histotripsy is not a crowded field. It is an owned one. Of the 35 histotripsy grants that name an assignee, 29 belong to either the University of Michigan or HistoSonics, the company Michigan engineers and physicians spun out in 2009 to commercialize the work. That is roughly 83 percent of the assigned record held by one university and its captive startup. The remaining handful is split among the University of Washington, Philips, Siemens Healthineers, and a few others working the edges. There is no second stack of comparable depth anywhere.

For a scout trying to read where a medical technology is heading, that concentration is the signal. Histotripsy is not a phrase that got popular. It is a specific physical trick, and almost everyone who can legally build it works for one institution or licenses from it.

The bug that became the feature

To see why this is more than a vocabulary count, read what the patents actually claim, because they describe the same machine pointed at wildly different problems.

One 2025 Michigan grant covers histotripsy therapy for the brain: an ultrasound transducer paired with a drainage catheter that doubles as a sensor, firing pulses that “generate cavitation that liquefies a target tissue in the brain,” then correcting for the way the skull smears the beam. Another, “Histotripsy for thrombolysis,” aims the same cavitation at blood clots, eroding a deep vein thrombosis or a pulmonary embolus with no drug and no catheter tip in the vessel. A University of Washington patent describes a transrectal probe running at 1 to 2.8 megahertz to boil and fracture prostate tissue. A HistoSonics grant is concerned with something as mundane as measuring the buoyancy forces on a treatment head sitting in its water bath so the robot arm does not drift off target.

Delete the word histotripsy from all of them and the family still holds together, because the mechanism is identical every time. You focus sound to a point, drive it hard enough that dissolved gas flashes into a cloud of microbubbles, and let those bubbles collapse with enough violence to shred tissue mechanically rather than cook it. That last distinction is the whole game. Ordinary high-intensity focused ultrasound heats tissue until it dies, which chars a margin and makes damage hard to see in real time. Histotripsy leaves the target liquefied and the boundary sharp, cell by cell, cold.

Here is the part that reframes the story. Cavitation was, for most of the history of therapeutic ultrasound, the thing engineers were desperate to avoid. Bubbles collapsing near a transducer pit the metal and scatter the beam unpredictably; in lithotripsy for kidney stones they were the source of collateral tissue injury nobody could control. Cain’s move, and Xu’s, was to stop treating cavitation as a hazard and start treating it as the instrument. The failure mode became the product. That is the adjacent-possible in one sentence: the tool was already sitting inside the problem, filed under “danger.”

A $2.25 billion vote, and a stranger second act

The market has now weighed in with real money. In October 2023 the FDA authorized HistoSonics’ Edison system to destroy liver tumors, on the strength of the international #HOPE4LIVER trials. By the company’s own account the platform has since treated more than 2,000 patients across over 50 US medical centers. In August 2025 an investor consortium including K5 Global, Bezos Expeditions, and Wellington Management bought a majority stake at a $2.25 billion valuation, one of the larger private medical-device deals of the year. This past February, Xu landed on Time’s list of the hundred most influential people in health.

But the liver clearance may turn out to be the least interesting thing histotripsy does. The active clinical pipeline shows the mechanism migrating outward fast. In the ClinicalTrials.gov registry, the Edison system is now enrolling for renal tumors, pancreatic cancer, and benign prostatic hyperplasia, and there is a completed study using it on the plantar fascia of the foot. A demolition tool built for hepatology is being aimed at the pancreas, the kidney, the prostate, and a runner’s heel.

The strangest turn is immunological, and it is the one worth watching. When histotripsy liquefies a tumor mechanically instead of burning it, it does not denature the debris. It sprays intact tumor antigens into the surrounding tissue, along with the danger signals, calreticulin surfacing on membranes, HMGB1 dumped into the interstitium, that the immune system reads as an alarm. In mouse work published in Frontiers in Immunology and elsewhere, treating one tumor with histotripsy drove CD8 T cells into distant, untreated tumors and slowed their growth, an abscopal effect that was absent when the two tumors were genetically unrelated. In other words, the sound field turns a solid tumor into its own vaccine, and the effect is antigen-specific.

That is why two of the newest trials pair the Edison system not with a scalpel but with drugs: one combines histotripsy with chemotherapy in advanced disease, another is an early-phase study of histotripsy alongside immune checkpoint inhibitors. The thesis, still unproven in humans, is that thermal ablation cooks the antigens that immunotherapy needs, while mechanical ablation preserves them. If that holds, the machine that Xu built to punch holes stops being a competitor to surgery and becomes an adjuvant to Keytruda-class drugs, a physical way to make cold tumors hot.

Who cares

For an R&D director in medical devices, the concentration is the takeaway: this is not a field you enter by hiring three engineers. The foundational and applied stack is held by one university and one now well-capitalized company, and the fresh grants are still landing on their side of the ledger. Entry means licensing or designing around a deep, actively growing portfolio, and the recent patents on beam aberration correction and treatment-head force sensing are exactly the unglamorous control problems that separate a lab demo from a device a hospital will buy.

For an oncology investor, the signal is the pivot from ablation to immunology. A $2.25 billion valuation on liver ablation is a bet on a procedure. A checkpoint-inhibitor combination that works would be a bet on a platform, and the patents filed for thrombolysis and brain tissue suggest the same consortium is already provisioning for a machine that outgrows cancer entirely. The pig heart in that Michigan lab was a stand-in for human tissue. Twenty-four years later, the list of tissues it stands in for is still getting longer.

Method note

Counts come from roughly 9.3 million US utility patent grants sourced from USPTO bulk grant data, searched for the term “histotripsy” in title, abstract, and full text, covering grants issued from 1976 through mid-June 2026. Assignee tallies combine variant spellings and subsidiary filings for the University of Michigan and HistoSonics; the 83 percent figure is share of the 35 histotripsy grants that carry a named assignee, and a small number of grants list more than one. Clinical trial status is drawn from the ClinicalTrials.gov registry. The immune-response findings are from published preclinical and early clinical literature indexed in OpenAlex and PubMed, including work in Frontiers in Immunology; the abscopal and checkpoint-combination results cited here are, to date, largely from animal models and early-phase human studies and should be read as promising rather than established. Acquisition, funding, and FDA details are from HistoSonics, the University of Michigan, and contemporaneous trade coverage.