A 20-minute ultrasound for addiction turns neuromodulation into an outpatient line item

Researchers at West Virginia University are testing focused ultrasound, beamed through the skull, as a treatment for people with Alzheimer's disease and with addiction ^[1]. What makes that a category change rather than another lab result is the delivery profile: the ultrasound takes about 20 minutes and the patient goes home the same day ^[2].

The useful comparison is the incumbent procedure. Deep brain stimulation, which the trials' neurosurgeon Ali Rezai helped pioneer, means cutting into the skull and implanting electrodes that reach neurons deep in the brain ^[4]. Rezai has performed more than 3,000 of those operations across a 30-year career ^[5]. His first, as a resident at NYU in 1996, ran 16 hours and required dozens of computers crammed into a basement room to record from the patient's brain ^[6]; implants now go in a fraction of that time with a few machines ^[7]. A 20-minute transcranial session is roughly one forty-eighth the duration of that first operation ^[8], and it does not break the skin ^[3].

Getting here took seventy years. In the 1950s William and Francis Fry built a rig that focused ultrasound to a single point, controls upstairs and probes coming down through the ceiling, and used the heat from high-intensity beams to burn pinprick lesions in animal brains without damaging surrounding tissue, then did the same in patients with Parkinson's disease ^[9][10]. Those early attempts still required brain surgery ^[11]. The shift is not in the physics but in the removal of the incision, which moves the thing out of an operating-room block and into something a clinic can schedule.

Demand is not the constraint. Almost 45 million Americans meet the criteria for a substance use disorder ^[12]. Overdose deaths have declined for three years and still ran to nearly 70,000 in 2025 ^[13], and West Virginia has the highest rate by population, more than double the national average ^[14], which is why Rezai took a post his colleagues called "academic suicide" ^[18]. The existing pharmacological stack churns rather than resolves: methadone left Erin McNulty tired, so she used cocaine to stay awake, and suboxone ended the heroin use before methamphetamine started ^[22][17].

Access is the constraint, and at present it looks like broadcast television. Linda McNulty found the trial in early 2024 by flipping channels onto a 60 Minutes segment ^[15], then had to locate a daughter who had been missing for weeks and move her more than 600 miles from Vermont to West Virginia ^[20][16]. That is not a referral pathway with any throughput in it.

Wired's headline states that Erin's brain "was fixed with sound, not a scalpel" ^[19]. The published account is a family narrative and does not report enrollment, endpoints or durability ^[21], so the clinical claim should be read as unsettled while the delivery claim is the part with operational consequence.

Watch three things. Whether the 20-minute figure survives contact with the full cycle time, once imaging, targeting and screening are counted. Whether the capability travels beyond a single academic site, or whether the 600-mile trip stays the price of entry. And whether anyone builds a reimbursement path for a same-day brain procedure in addiction, because without one this is a device that works and cannot be scheduled.