Psychiatry,
measured.
A consultative psychiatric practice for adults with complex mood and neuropsychiatric presentations, where the diagnosis is rebuilt from the evidence and every treatment decision is tied to something that can be measured.
[Two facts belong here before launch: whether the practice is open to new patients, and the states in which you hold an active licence — with telehealth, licensure follows the patient's location at the time of the visit, so a reader in another state needs to know immediately. This is the single most common reason a promising first enquiry turns into a wasted exchange of emails.]
Physician–Scientist · Measurement-Based Care · Neuropsychiatric Focus
An engineer's habits, brought to psychiatric care.
[This pull quote was drafted for you, not quoted from you, and it is currently attributed to you by name. The proposed sentence was: “Bringing measurement to medicine’s least measurable specialty, so that the right circuit gets the right treatment.” Approve it verbatim, replace it with a sentence you have actually said or written, or delete the block — a quotation carrying a named physician’s signature on his own practice site should be his own words. Note also that “the right circuit gets the right treatment” reads as an outcome promise and sits directly above the research-versus-clinical boundary this site is otherwise careful to hold, so it may be worth rewording even if you keep it.]
I am a psychiatrist and biomedical engineer. I trained in the UCLA–Caltech Medical Scientist Training Program and earned a PhD in bioengineering for work on label-free fluorescence-lifetime imaging — an instrument built to tell surgeons, in real time, where a tumour ends and healthy tissue begins. It was funded by an NIH–NCI R01, produced two patents, and was translated into clinical use at UCLA.
I came to psychiatry with the habits of that work intact: define the target, measure what you are doing, change one variable at a time. I completed residency at UCSF on the research track and graduated with distinction in neuropsychiatry and interventional psychiatry. My research is in imaging-guided neuromodulation, and it shapes how I think about ordinary clinical problems even when no device is involved.
This practice is built for people with complicated histories — a diagnosis that has changed several times, a medication list that is long and poorly documented, or a decision that deserves a careful second reading before it is made. It is a consultative practice, not an urgent-care or crisis service.
[Say plainly whom you see and whom you do not: age range (adults only? 18 and over? older adolescents?), and the presentations you decline or refer on — primary substance use disorders, eating disorders needing medical monitoring, forensic or court-ordered evaluations, disability paperwork. Naming the boundary here prevents the referral that wastes an evaluation slot and the difficult first visit that follows it.]
What this practice offers.
Six kinds of work, all of which begin with the same step: establishing what is actually going on before deciding what to do about it.
[Confirm that these six lines match what you are genuinely prepared to take on in the first year, and delete any you are not. A service listed here is a service a patient will arrive expecting; it is also the list a payer, a board, or an attorney would read as a description of your scope.]
Diagnostic evaluation
Most treatment that goes wrong went wrong at the diagnosis. An evaluation here starts from the beginning: a full psychiatric and medical history, a written timeline of every prior medication with its dose and duration, outside records and collateral where you can provide them, and standardised rating scales so that the starting point is documented rather than remembered. The output is a formulation you can read, disagree with, and take elsewhere.
Medication management and pharmacogenomics
Medication decisions are made deliberately and reviewed against a measure rather than an impression: one change at a time, with the expected effect and the timeframe agreed in advance. Where prior trials have failed for reasons nobody could explain, pharmacogenomic testing may occasionally identify a metabolic outlier. It is one input among many and it does not choose the drug.
[State whether you order pharmacogenomic panels yourself, which panel, and how it is billed — these labs frequently bill the patient separately from your fee, and patients who learn that afterwards are justifiably angry. If you interpret outside panels rather than ordering them, say that instead.]
Treatment-resistant mood disorders
When two or more adequate trials have not worked, the useful questions are usually whether those trials were in fact adequate, whether the diagnosis still fits, and what has never been tried. That means re-examining the bipolar spectrum, attention and trauma histories, sleep, thyroid and metabolic contributors, and substance use before reaching for the next prescription. Evidence-based options are then sequenced explicitly, with what would count as a response written down beforehand.
Neuromodulation consultation
Whether a device-based or interventional treatment belongs in a plan — and if so which one, in what order, and after what else — deserves an unhurried hour rather than a referral reflex. My research is in imaging-guided neuromodulation, which informs how I read the evidence for these treatments and, just as often, where the evidence does not yet support them.
[List exactly which interventional treatments you deliver yourself in this practice, if any, and which you only advise on and refer out for. TMS, ketamine, esketamine and ECT each carry different site, monitoring and regulatory requirements, and naming them on a practice page without that distinction reads as an offer of availability. This is the paragraph most likely to be quoted back at you.]
Second-opinion consultation
A time-limited review for patients, families, or treating clinicians who want another reading of a complicated record before the next decision. The record is reviewed before we meet; the visit is spent on the reasoning rather than on repeating the history.
[Describe the deliverable and the boundary: does a second opinion end in a written summary, does that summary go to the referring clinician, and do you take over ongoing care afterwards or return the patient to their treater? Ambiguity here is what turns a clean consultation into an unintended treatment relationship.]
Collaboration with existing treaters
Many people arrive already working with a therapist, a primary care physician, a neurologist, or a psychiatrist they intend to keep. With your written permission I correspond directly with them, so that the plan is shared rather than reconstructed from memory at each visit. Good psychiatric care is usually a small team that agrees on the formulation.
How care begins.
Four steps, in order, with no obligation to continue at any of them.
Enquiry
You send a short note through the contact form describing, in general terms, what you are looking for. No confidential clinical detail is needed at this stage, and none should be sent by email.
[Say who reads enquiries — you personally or an assistant — and how long a reply usually takes. Patients reasonably read silence as rejection, and a stated turnaround you can actually meet is worth more than a fast one you cannot.]
Brief introductory call
A short conversation to establish what you are hoping for and whether this practice is a sensible fit. If it is not, I will say so and, where I can, suggest a better direction.
[Give the length of this call, whether there is any charge for it, and who conducts it. If it is free, say so explicitly — it materially changes how many people make the first contact.]
Evaluation
The full assessment: history, timeline of prior treatment, review of outside records, baseline rating scales, and any medical workup that the picture calls for.
[Specify how many sessions the evaluation takes and how long each runs, what it costs, what records you want in advance and how to send them securely, and whether the first session is in person or by video. Under the federal No Surprises Act and California AB 1020 a self-pay patient must receive a Good Faith Estimate before the visit, so this figure has to match what you actually charge.]
Formulation and plan
You receive a written formulation: what I think is going on, what I am uncertain about, the options in the order I would try them, and what each is intended to achieve and by when. If ongoing care follows, that plan is the document we revise together.
[State the cadence and cost of follow-up, how prescriptions and refills are handled between visits, how you can be reached and how quickly, and your cancellation policy in plain figures. Every one of these is a question you would otherwise answer by email several times a week.]
The research behind the thinking.
Three active neuromodulation studies, run through academic centres. They are the reason this practice thinks in circuits and measurement — they are not services offered here.
Focused ultrasound for treatment-resistant depression
Low-intensity focused ultrasound (LIFU) is a non-invasive way of reaching deep targets that transcranial magnetic stimulation cannot. This trial evaluates it in major depressive disorder that has not responded to standard treatment.
fMRI connectivity-guided TMS for chronic tinnitus
Individual functional connectivity is used to place the stimulation target rather than relying on a scalp landmark, in collaboration with neurology, radiology, and head and neck surgery.
kTMP for anhedonic major depression
Kilohertz transcranial magnetic perturbation is a newer class of non-invasive electromagnetic stimulation. This study evaluates it against the anhedonic dimension of major depression specifically.
[Write one plain sentence, in your own words, drawing the line between research and clinical care: that these protocols run through UCSF and Stanford under their IRBs, that becoming a patient here does not enrol anyone in them and confers no advantage in screening, and how someone should actually enquire about a trial. Without that sentence a reader will assume an appointment is the route to focused ultrasound, which is a consent problem long before it is a marketing one.]
[The MedicalBusiness structured-data block at the top of this file (index.html, in the <head>) is what Google, Bing, Apple Maps and every healthcare directory read to build a listing for this practice. It currently carries no address and no telephone number, because a bracketed placeholder in that block would have been published as your real address and phone without ever appearing on the page. Before launch, supply four things and have them written into that block: street address, city, ZIP, and the practice telephone number.]
[Supply the state as well. The record previously asserted “CA”, which was never confirmed — the state of practice is an open placeholder everywhere else on this site (the hero above, about.html and contact.html). A state of licensure asserted in machine-readable data and not actually held is a board matter, not a typo, so this one must come from you rather than be inferred.]
Start with a conversation.
If you are weighing whether this is the right practice for a complicated history, a short note is the whole first step. Please keep it general — this website is not a secure channel for clinical information.