Research & Trials

The research, and its limits.

Dr. Pellionisz is a psychiatrist and biomedical engineer. Alongside clinical work he is involved in three active neuromodulation studies — focused ultrasound, connectivity-guided transcranial magnetic stimulation, and kilohertz magnetic perturbation — and spent his doctoral years building an optical imaging system that reached the operating room. This page describes that work in plain language, and is explicit about where research ends and care begins.

Research is not the same as treatment

The studies described below are research protocols run at academic centres — UCSF and Stanford — under their own review boards, consent documents, funding and eligibility criteria. They are separate from this private practice. Enrolment is decided by each study team, not here; taking part in a study is not a substitute for clinical care; and nothing on this page is an offer of treatment.

[State plainly, in one or two sentences, which modalities you actually deliver to patients in this practice and which exist for you only inside a research protocol — naming TMS, focused ultrasound and kTMP individually rather than saying "some treatments". This is the single most misread thing on a neuromodulation site: readers reasonably assume that describing a technology means offering it. An implied offer of an investigational device is both a marketing-claim problem with the medical board and a consent problem with the IRB, and it is also the fastest way to fill your inbox with enquiries you have to decline. If the honest answer is "evaluation, medication management and referral only", say exactly that — it reads as confidence, not as a limitation.]

Active studies

What is under investigation now.

Three studies, each asking whether stimulation delivered to a more precisely chosen target changes a symptom that has resisted conventional treatment. None of them has answered that question yet; that is what makes them studies.

UCSF · Clinical trial

Low-intensity focused ultrasound for treatment-resistant depression

Focused ultrasound aims sound waves from outside the head so that they converge on one small region of brain tissue. At low intensity the intent is not to heat or destroy anything, but to change — briefly and reversibly — how active that region is. Because sound passes through the skull and can be focused at depth, it can reach structures that magnetic stimulation applied at the scalp cannot. This trial is testing whether that approach helps people whose depression has not responded to standard treatment.

Sub-investigator · industry-sponsored (Attune Neuroscience) · UCSF · 2024–present

Stanford · Clinical trial

fMRI connectivity-guided TMS for chronic tinnitus

Transcranial magnetic stimulation delivers brief magnetic pulses through the scalp to alter activity in the cortex beneath it. Conventional targets are found by scalp measurement and are the same for everyone, but the circuit of interest sits in a slightly different place in each brain. This study uses each participant's own functional MRI — a map of which regions rise and fall in activity together — to decide where the coil goes. The question is whether individualised targeting changes the response in chronic tinnitus, a symptom generated in the brain rather than in the ear.

Stanford University · with neurology, radiology and head & neck surgery · ongoing. [Give your role of record on this protocol exactly as it appears on the IRB approval — principal investigator, co-investigator or sub-investigator. Role titles are the one line on a research page that colleagues and journalists check, and an upgraded title is a correction you never fully undo.]

NIMH SBIR · Clinical trial

kTMP for anhedonic major depression

Kilohertz transcranial magnetic perturbation is a newer form of noninvasive magnetic stimulation. Rather than the discrete, tapping pulses of conventional TMS, it applies a continuous field oscillating in the kilohertz range, producing a weaker electric field intended to shift how readily neurons fire rather than to force them to fire. The study focuses on anhedonia — the loss of pleasure and motivation — because it is among the dimensions of depression that respond least well to medication. It is treated here as a target in its own right rather than as one line on a depression rating scale.

Sub-investigator · NIMH SBIR · 2026–present

[For each of the three studies above, tell me its current status — open to enrolment, enrolment paused, closed to accrual, or follow-up only — plus one line on who is broadly eligible (age range, diagnosis, and the city where visits happen) and the ClinicalTrials.gov identifier if there is one. Then decide who owns keeping it current. A stale "now enrolling" is worse than no status at all: a patient who arranges time off and travels for a screening visit that no longer exists remembers that, and an out-of-date public claim about an investigational study is exactly what a sponsor and an IRB both take seriously. If you would rather not maintain a status line by hand, say so and I will replace it with links to the ClinicalTrials.gov records, which update themselves.]

Prior work

Seeing tissue without staining it.

Before psychiatry, Dr. Pellionisz built imaging instruments. His doctoral work in the UCLA–Caltech Medical Scientist Training Program produced Dynamic Optical Contrast Imaging (DOCI), a label-free fluorescence-lifetime imaging method for use during head and neck surgery.

Most tissues glow faintly when illuminated, and the glow fades at a rate that differs between tissue types. DOCI measures that rate of decay rather than the colour or brightness of the glow, which means it can distinguish tissues without injecting a dye or waiting for a pathologist. The two problems it was built for are practical ones: telling a surgeon in real time where a tumour ends and normal tissue begins, and identifying parathyroid glands, which are small, variable in position, and consequential to injure.

The work was funded by an NIH-NCI R01, produced two patents, and was translated into clinical use at UCLA. The through-line to the current neuromodulation work is not the technology but the premise: that a treatment gets better when you can measure what you are aiming at.

Programme MD–PhD, UCLA–Caltech Medical Scientist Training Program; PhD in Bioengineering, UCLA (dissertation 2020)
Funding NIH-NCI R01
Intellectual property Two patents in optical imaging for image-guided surgery
Translation Clinically translated at UCLA

Selected publications

Peer-reviewed work.

More than twenty peer-reviewed publications and conference proceedings, three book chapters, and two patents, spanning brain stimulation and optical imaging for surgery.

[Supply the selected-publications list you want shown here — five to ten entries, each copied exactly as it appears in your CV or on PubMed: the full author string in its published order, title, journal, year, volume, issue, pages, and a DOI or PubMed link. Do not let me reconstruct any of this from memory, and do not let a citation go up that you have not opened and checked. On a practice site a citation is a credential claim, not decoration: a referring colleague verifies it in a single search, a wrong volume or page range reads as carelessness, and a wrong authorship position — a paper listed as first-author when you were third — is misrepresentation that a medical board treats as an advertising violation and that stays cached long after the page is corrected. Tell me explicitly which entries are first-author if you want that noted; otherwise I will publish author strings only and let readers see the position for themselves. If you would rather not maintain a list by hand, say so and I will replace this section with a single link to your PubMed or ORCID profile, which stays current without you touching it.]

[Optional: if you want a CV posted here, supply the PDF and the filename you want it served under, and I will add a download link in this spot. Before you hand the file over, strip your home address, personal mobile number, date of birth, scanned signature, and any licence or NPI numbers. An academic CV is written for search committees who already have a duty of confidentiality; a file linked from a public practice site is crawled, cached and archived within days, and those exact fields are what credential-fraud and identity-theft collectors harvest. If stripping it properly is more work than it is worth this month, no CV link is the safer default — "full bibliography available on request" costs you nothing.]

For referring clinicians

What a referral can be for.

Written for colleagues in psychiatry, primary care, neurology and otolaryngology who are weighing what to do next for a patient who has not responded to the obvious steps.

Questions this practice is suited to

Diagnostic clarification in mood disorders that have failed several adequate medication trials; neuropsychiatric presentations where the boundary between a psychiatric and a neurological explanation is unclear; interpretation of a prior course of device-based treatment, including whether the target, protocol and dose were adequate before it was called a failure; and orientation for a patient asking about focused ultrasound, TMS or research participation who needs an accurate account of what those options currently are.

[Confirm which of these you actually want to take on, and in what form: informal curbside, a formal one-off consultation with a written note back to the referrer, or ongoing co-management with you handling one part of the treatment. Colleagues will read this list as an offer and refer accordingly, and each of those three carries a different documentation, billing and liability posture. Delete anything you do not want in your inbox.]

Records that make the visit useful

A medication history with doses, durations, response and the actual reason each agent was stopped is worth more than any other document. Beyond that: records of previous device or interventional treatment (TMS, ECT, ketamine or esketamine) including protocol, target, number of sessions and the response measured at the time; any standardised rating scores; recent laboratory work and, where relevant, imaging reports; the audiology and ENT workup for a tinnitus referral; and a short note on safety history and prior levels of care.

Please send records through a secure, HIPAA-compliant channel rather than ordinary email, and let the patient know what is being shared.

How to make contact, and when to expect a reply

[Give the exact route you want colleagues to use — a direct line, a secure-message address on the practice domain, or a referral form — and say whether it differs from the patient-facing route, because if it does not, referrals will sit behind the general enquiry queue. Note also that ppellionisz@stanford.edu is a university address tied to your research role; using it for practice referrals mixes two record systems and places clinical correspondence on a server the practice does not control. This should be a practice-domain address.]

[State the turnaround you can genuinely honour for a colleague's call or written question — for example "within two business days" — and what happens outside that window, including who covers when you are away. Referring clinicians plan around this figure and it is the one number on the page you will be held to, so do not set it from a light week. If you would rather not commit to a number, we can say that referrals are acknowledged before an appointment is offered, which is honest and still useful.]

Questions about a study, or about a patient?

Patients and colleagues are welcome to ask what is research, what is available as care, and which of the two a particular question belongs to.

Get in touch About Dr. Pellionisz