Signals: what's Now, Near and Next in health & BIomedicine

November 2026 · Curated by Daniel Kraft, MD

Between newsletters, I keep a running list of the developments that matter most across health and medicine. Here is the full set behind the November issue: 125 signals across seven themes, newest first within each, plus 15 developments on the horizon. Most are from the past two months, with a few earlier standouts.

Jump to: AI co-clinician · Trust and governance · Care models · Monitoring · Programmable biology · Longevity · From the lab · What's Next

1. AI is moving from assistant to co-clinician

AI is moving past note-taking into nurse handoffs, decision support, screening and patient-facing agents. The open questions are accuracy, workload and who stays in the loop.

2. Trust and governance: who sets the rules for health AI

Regulators, professional bodies and states are drawing the lines on liability, privacy and oversight, while patients say they want a human involved.

3. Care models are being rebuilt, not just digitized

Payment is starting to follow prevention and care outside the hospital, and new models are proving they can scale.

4. Monitoring is going continuous

Phones, watches, rings and blood tests are turning into screening tools. Validation and reimbursement are the bottlenecks.

5. Biology is becoming programmable

Personalized vaccines, gene editing inside the body and AI-designed molecules are reaching patients, with real safety signals to watch.

6. Longevity is becoming clinical

Midlife prevention, GLP-1s and better ways to measure biological aging are pulling longevity into mainstream care, along with a healthy dose of reality checks.

7. From the lab: early science worth watching

Robots, organoids, implants and new imaging that are still in animals or first small studies, but point to where care could go.

What's Next: 15 developments on the horizon

Not in the clinic yet, but closer than you might think. Each one is anchored in a real milestone from 2026. Horizons are my best estimate of when each could reach routine use for meaningful numbers of patients.

Arriving soon: 1–3 years

Blood pressure at your fingertip, no cuff required

A pocket device reads blood pressure from a fingertip press, with no inflatable cuff and no calibration against one. In May 2026 the FDA cleared the first such over-the-counter monitor for home spot checks, after accuracy testing in 264 adults; it is limited to ages 22 to 59 without arrhythmias. What has to be true: real-world accuracy, and expansion to older adults and people with atrial fibrillation.

Source: Medical Device Network

A lifespan pill for older dogs

A daily pill meant to extend healthy life in senior dogs, and a test case for regulating a drug that targets aging itself. In January 2026 FDA veterinary regulators accepted the safety section of its application, which the developer calls a first for a lifespan drug, while a 1,300-dog efficacy trial continues. What has to be true: the manufacturing review and the efficacy trial both have to come through; people are a much longer road.

Source: dvm360

3–5 years

Drones that race the ambulance to a cardiac arrest

Drones launched with 911 dispatch carry a defibrillator to the address, so a bystander can shock the heart before the ambulance arrives. A US multi-site study now flies them on live cardiac arrest calls in North Carolina and Virginia, where early simulations put the drone on scene about 6 minutes ahead of ground crews. No survival data yet. What has to be true: FAA rules for routine beyond-line-of-sight flights, and survival data showing drones change outcomes.

Source: WHRO

Pregnancy ultrasound anyone can do, with AI guidance

Health workers with no ultrasound training sweep a low-cost handheld probe across the belly, and AI on a phone estimates how far along the pregnancy is. In a 2026 study in Chicago and Nairobi, it dated 385 pregnancies with a 4.2-day average error, as accurate as standard clinical assessment. What has to be true: regulatory clearance, durable low-cost hardware, and a place in prenatal care programs.

Source: AuntMinnie · JAMA Network Open

A digital twin of your heart guides the procedure

A personalized computer model built from a patient's heart MRI simulates dangerous rhythms to show cardiologists exactly where to ablate. In a 10-patient first-in-human trial reported in April 2026, all were free of ventricular tachycardia a year later apart from brief episodes in two during healing, versus roughly 60% long-term success with standard ablation. Early data with no control group. What has to be true: a randomized multicenter trial, and twins that are fast and cheap to build.

Source: Associated Press · NEJM

An AI-discovered drug reaches Phase 3

A pill for lung scarring whose biological target and molecule were both found with generative AI. In July 2026 a 320-patient, year-long Phase 3 trial began in idiopathic pulmonary fibrosis, after a small Phase 2a showed improved lung capacity at the top dose over 12 weeks. What has to be true: Phase 3 has to confirm the benefit seen in a small, short study.

Source: Clinical Research News · Phase 2a paper, Nature Medicine

5–10 years

Paramedics who fly to the emergency

Small electric vertical-takeoff aircraft carry a paramedic and lifesaving gear over roads and water to rural emergencies. In 2026 a rural North Carolina county became the first in the US to use one in routine 911 response; on its early live calls, the paramedic reached one patient about 20 minutes ahead of the ground ambulance. What has to be true: aviation rules, weather limits and cost per mission, and evidence that faster arrival improves outcomes.

Source: JEMS

A wireless brain implant that gives back a voice

A fully implanted brain-computer interface reads individual neurons and turns attempted speech into a synthesized voice in real time. In September 2026 the first participant in an FDA-approved feasibility trial, a woman in her 60s with motor neuron disease, talked with her family through the device. One patient so far. What has to be true: long-term safety and durability, a larger pivotal trial, and a way to pay for it.

Source: ALS News Today

Sound waves that retune mood circuits, no surgery

Low-intensity focused ultrasound, aimed through the skull, changes activity in a targeted brain region with no implant, incision or anesthesia. In a 36-patient sham-controlled trial in China, 10 days of treatment roughly halved depression scores, with effects lasting 4 weeks. What has to be true: larger multicenter trials showing durable benefit, and a cleared device.

Source: HCPLive · Journal of Affective Disorders

Surgery from another hospital

A surgeon at one hospital controls a robot operating on a patient at another site over a secure network, with a team at the bedside. On Oct 25, 2026, a Florida surgeon removed a 66-year-old patient's prostate at a sister hospital 35 miles away, under an FDA investigational exemption. What has to be true: network reliability and cybersecurity, FDA clearance, and rules for licensing and liability across sites.

Source: American College of Surgeons

CAR-T made inside the body

One infusion turns a patient's own T cells into targeted CAR-T cells inside the body, skipping weeks of custom lab manufacturing. In September 2026 researchers in China reported that it worked in 16 patients with treatment-resistant autoimmune nerve diseases, including multiple sclerosis, with improvements lasting six months so far. Small study, no control group. What has to be true: controlled trials showing durable benefit and long-term safety.

Source: Singularity Hub · NEJM

A pig kidney as a bridge to a human one

Kidneys from gene-edited pigs as a stopgap, and perhaps one day an alternative, to scarce donor organs. In September 2026 Boston surgeons reported that a pig kidney kept a 66-year-old man off dialysis for a record 271 days, before a human kidney transplant that worked immediately. What has to be true: multi-patient trials showing pig kidneys last for years on tolerable immunosuppression.

Source: Mass General Brigham · The Lancet

A one-time shot to lower cholesterol

A single CRISPR infusion permanently switches off a liver gene that raises LDL cholesterol and triglycerides. In August 2026 researchers reported that the top dose cut LDL by 52.5% and triglycerides by 47.8% after one year in 15 people, with no serious treatment-related side effects. What has to be true: outcome trials showing fewer heart attacks and strokes, long-term safety of a permanent edit, and a workable price.

Source: Cleveland Clinic · NEJM

10+ years

Rewinding cell age in the eye

A gene therapy that switches on reprogramming genes in a controlled way to restore younger gene activity in damaged optic nerve cells. In June 2026 the first volunteer, a person with glaucoma, was dosed in a Phase 1 safety trial, widely described as the first human test of partial reprogramming. No efficacy data yet. What has to be true: safety with no tumor risk, then measurable vision benefit in controlled trials.

Source: MIT Technology Review

Needle-free glucose on your wrist: still not here

A watch or ring that reads blood sugar through the skin, with no needle or sensor underneath. As of mid-2026 no such device has FDA authorization, and the agency's warning against products claiming to do it still stands. What has to be true: a sensor as accurate as today's under-skin monitors across skin tones, motion and temperature, which no one has shown yet.

Source: eWeek

See it in person. NextMed Health 2027 is May 17–20, 2027, in Silicon Valley, with clinicians, scientists, founders and investors exploring what's Now, Near and Next. Learn more and apply

Back to top