The Brain Scan Promise: When Imaging Meets Psychiatry's Uncertainty

This article examines whether brain imaging can diagnose psychiatric conditions like depression, what the current science actually shows, and why the promise of "seeing" mental illness on a scan remains largely unfulfilled.

Someone asked me last week if I'd ever ordered a brain SPECT scan for depression. The question caught me off guard - not because it's unreasonable, but because it highlights something fascinating about how we think about psychiatric illness. We want to see what's wrong. We want the problem to be visible, measurable, something we can point to on a screen and say, "There. That's it."

SPECT vs. Standard Psychiatric Diagnosis: A Clinical Comparison

Feature	SPECT Brain Imaging	Standard Psychiatric Assessment
What It Measures	Blood flow patterns in brain; ~2 minute snapshot showing regional perfusion	Symptom patterns, duration, severity, and functional impairment using validated scales
Technology Used	Nuclear medicine scan with radioactive tracer injection	Clinical interview, standardized questionnaires (PHQ-9, GAD-7), behavioral observation
Time Required	45-minute wait post-injection + 30-minute scan	45-90 minute initial evaluation; 15-30 minute follow-ups
Cost	$1,500-$4,000+ per scan (usually not covered by insurance for psychiatric use)	$100-300 for initial visit (usually covered by insurance)
Evidence Base for Psychiatric Use	No large randomized controlled trials; mostly case series and retrospective analyses	Decades of validation studies; established treatment guidelines based on symptom patterns
Professional Guidelines	APA and AAN recommend against routine use for psychiatric diagnosis/treatment planning	Standard of care endorsed by all major psychiatric organizations
Proven Clinical Uses	Epilepsy (seizure focus), stroke assessment, certain dementias, brain trauma evaluation	Depression, anxiety, ADHD, bipolar disorder, schizophrenia diagnosis and treatment
Key Limitation	Cannot reliably predict medication response or improve treatment outcomes for individual patients	Relies on subjective symptom reporting; no biological confirmation
Best Suited For	Neurological conditions with structural or perfusion abnormalities	All standard psychiatric diagnoses when neurological causes ruled out

I understand that impulse completely. Maybe too much.

What SPECT Actually Shows

SPECT imaging - Single Photon Emission Computed Tomography, which is a mouthful - is a nuclear medicine study that looks at blood flow patterns in the brain. You inject a small amount of radioactive tracer, wait about 45 minutes, then scan. What you get is essentially a map of where blood is flowing, which theoretically correlates with brain activity. It captures roughly a two-minute snapshot, though proponents argue the patterns remain fairly consistent unless you actively change something about your brain health.

It's different from functional MRI in a subtle but important way. fMRI catches changes in real-time during specific tasks - you think about something, certain areas light up. SPECT gives you more of an average over that brief window. Both are trying to answer the same question: what's happening in there?

The technology itself isn't controversial. SPECT has legitimate uses in neurology - detecting seizure foci in epilepsy, assessing blood flow after stroke, looking for signs of certain dementias. The controversy comes when you use it for psychiatric diagnosis and treatment planning.

The Clinical Dilemma We Don't Talk About Enough

Here's the uncomfortable truth: psychiatry really is the only medical specialty that usually makes diagnoses without looking at the organ it treats. We diagnose diabetes by checking blood sugar. We diagnose heart failure with echocardiograms and BNP levels - a blood marker that rises when the heart is stretched and failing. We diagnose pneumonia with chest X-rays and inflammatory markers.

But depression? Anxiety? ADHD? We use symptom checklists. The PHQ-9 for depression, the GAD-7 for anxiety - standardized questionnaires that quantify subjective experience. The DSM-5, our diagnostic manual, contains almost no neuroscience. It's symptom clusters, duration criteria, functional impairment. Same basic framework we've used since the 1950s.

That bothers people. It bothers patients who want concrete answers. Sometimes it bothers me, though I've learned to sit with that discomfort rather than rush to resolve it.

The argument for imaging goes something like this: if we could see what's actually happening in the brain, we could treat more precisely. Less trial and error with medications. Better outcomes. More personalized medicine. It sounds logical. Compelling, even.

What The Evidence Actually Says

This is where it gets messy.

The major psychiatric and neurological organizations - the American Psychiatric Association, the American Academy of Neurology - have explicitly stated that SPECT (and similar functional imaging) should not be routinely used for psychiatric diagnosis or treatment planning. Not because the scans are fake or useless, but because we don't have good evidence that using them improves patient outcomes.

There are no large randomized controlled trials showing that treatment decisions guided by SPECT lead to better results than standard psychiatric care. None. The evidence base consists largely of case series, retrospective analyses, and claims based on clinical experience. Valuable in their own right, but not sufficient to overturn decades of established practice guidelines.

A 2020 review in the Journal of Nuclear Medicine noted that while SPECT can show differences in brain perfusion patterns, "the clinical utility for individual patient management in psychiatric disorders remains unproven." That's academic language for: interesting, but we can't recommend this yet.

The counter-argument usually invokes experience. "I've looked at hundreds of thousands of scans. I see patterns. My patients improve." And maybe they do. Clinical intuition built over decades matters. But it's not the same as systematic evidence. It's not reproducible in the way we need medical practice to be.

The Weight Question (And Why It Actually Matters)

One thing that does hold up: the relationship between obesity and brain health. Multiple large studies - including imaging studies using MRI and PET, not just SPECT - show associations between higher BMI and reduced brain volume, particularly in areas like the hippocampus and frontal lobes. A 2019 study in Radiology with over 12,000 participants found measurable decreases in gray matter volume as weight increased.

The inflammatory hypothesis makes biological sense. Adipose tissue - especially visceral fat around organs - secretes pro-inflammatory cytokines like IL-6 and TNF-alpha. Chronic low-grade inflammation affects the brain. We know inflammation plays a role in depression; it's why some people with treatment-resistant depression respond to anti-inflammatory interventions.

So the advice to address weight, diabetes, sleep, and substance use in the context of mental health? That's solid regardless of imaging. You don't need a SPECT scan to know that untreated diabetes, chronic sleep deprivation, and alcohol use disorder make depression worse. We have excellent evidence for that already.

The question is whether the scan adds anything beyond what we can assess clinically.

What I Think About When Someone Brings This Up

I've had patients ask about brain scans. Usually after reading something online or hearing about someone who "finally got answers" after imaging. The desire is genuine. They've tried three antidepressants, therapy isn't quite working, and they want to know why. They want proof that it's not just in their head - which is ironic, because of course it is in their head, just not in the dismissive way that phrase usually implies.

What I don't do is dismiss the question. Because underneath it is something important: the frustration with psychiatric diagnosis feeling vague, subjective, somehow less legitimate than "real" medical problems. And that feeling isn't wrong. There is something unsatisfying about diagnosing Major Depressive Disorder based on nine symptom criteria and two-week duration thresholds.

But imaging isn't the solution we're hoping for. Not yet, anyway.

What actually helps - at least in my experience, which is admittedly limited and not generalizable - is addressing the biology we can measure. Checking for hypothyroidism, which can mimic depression. Testing vitamin D and B12 levels. Asking about sleep architecture, not just hours slept. Reviewing inflammatory markers in people with chronic physical illness. Looking at glucose regulation. These aren't fancy. They're basic. But they're evidence-based and actionable.

And sometimes the answer isn't more testing - it's better listening. Understanding that when someone says they're "tired all the time," they might mean profound anhedonia, or they might mean actual sleep deprivation, or both, and parsing that takes time. You can't see it on a scan.

The Supplement Industry Angle (Which Complicates Everything)

I'd be dishonest if I didn't mention that some clinics offering SPECT-based psychiatry also sell proprietary supplements. Not universal, but common enough to note. That doesn't automatically invalidate the approach, but it does create perverse incentives. If your clinic's revenue stream depends on supplement sales linked to scan findings, your threshold for recommending scans - and supplements - might shift in ways you don't consciously recognize.

Evidence for most psychiatric supplements is mixed at best. Omega-3 fatty acids have modest support for depression, particularly formulations higher in EPA. Some studies on SAMe, methylfolate for certain MTHFR variants, occasional signals for NAC in specific contexts. But "brain health formulas" bundled into proprietary blends? Usually not well-studied as combinations.

I'm not opposed to supplements categorically. I recommend magnesium for patients with chronic stress and muscle tension. Vitamin D for deficiency. Omega-3s sometimes, when diet alone isn't cutting it. But I recommend them based on established deficiencies or specific evidence, not because a scan suggested a pattern.

Why Outcomes Haven't Improved (And What That Means)

The claim that psychiatric outcomes haven't improved since the 1950s is... partially true, and partially more complicated than that. Antidepressants today aren't dramatically more effective than imipramine - the first tricyclic from 1957 - but they're vastly more tolerable. Fewer cardiac side effects, less lethality in overdose, less sedation. That matters. It means more people can actually stay on treatment.

We also have better psychotherapy models now. Dialectical Behavior Therapy for borderline personality disorder didn't exist in the 1950s. Trauma-focused CBT for PTSD. Acceptance and Commitment Therapy. These aren't just rebranding - they represent genuine advances in how we help people.

But the core problem remains: we're still treating based on syndromes, not mechanisms. We don't treat "high blood sugar syndrome" - we treat diabetes with specific pathophysiological targets. Psychiatry doesn't have that yet. We have serotonin, dopamine, norepinephrine - neurotransmitter systems we can modulate - but we're not precision targeting specific deficits in individual patients.

Maybe imaging will get us there eventually. Maybe not SPECT, but some future combination of imaging, genetics, and biomarkers. I don't know. I hope so.

What I Actually Do (When Someone Is Struggling)

When I see someone whose depression isn't responding to standard treatment, I think about context first. What else is happening? New stressors, relationship breakdown, chronic pain, medication side effects from something else entirely. The sleep history - always the sleep history. Substance use, because people don't always volunteer that information upfront.

Then I think about comorbidities. Thyroid function. Anemia. B12 deficiency. Testosterone in men, sometimes. PCOS in women. Undiagnosed sleep apnea. Chronic inflammation from autoimmune conditions or obesity. All of these can present as or worsen psychiatric symptoms.

If those are addressed and we're still stuck, then it's medication optimization - dose adjustments, augmentation strategies, switching classes. Or deeper psychotherapy if that hasn't been tried. Or reassessing the diagnosis entirely, because sometimes what looks like depression is actually bipolar disorder, and antidepressants alone make things worse.

It's systematic. Not glamorous. Often slow.

Would a brain scan occasionally show something surprising? Maybe. Rarely. But the likelihood of finding actionable information that changes management is low enough that I can't justify the cost - or the radiation exposure, which is small but not zero - based on current evidence.

The Bigger Question (That No One Really Asks)

Why are we so drawn to the idea that seeing the brain would solve psychiatric illness?

Part of it is scientism - the belief that if something is measurable and visualizable, it's more real, more legitimate. Part of it is desperation. When you've suffered for years, tried multiple treatments, and nothing quite works, you grasp at anything that promises clarity.

But I think part of it is also discomfort with subjectivity. With the reality that mental suffering, while biologically rooted, is experienced and expressed through individual narrative, context, meaning. That it's both physical and existential. Both chemical and phenomenological.

A scan can't tell you why your depression feels like drowning versus feeling like numbness. It can't explain why your anxiety manifests as chest tightness or why certain memories trigger disproportionate fear responses. The biology matters, absolutely. But the lived experience - the part that actually determines whether someone can function, find meaning, feel joy - that exists in the space between neurons.

Maybe we don't need to see it. Maybe we need to get better at listening to it.

I don't know. I go back and forth.

Some days I think imaging will eventually revolutionize psychiatry. Other days I think we're looking for answers in the wrong place - that the real advances will come from better understanding stress biology, trauma, inflammation, the microbiome, the social determinants of health. Things we already know matter but don't integrate well into clinical practice.

Probably it's some combination. Probably I'm overthinking it.

What This Means For You (If You're Struggling)

If someone offers you a brain SPECT scan for psychiatric diagnosis, ask questions. Ask about the evidence base. Ask what specific treatment decisions would change based on the results. Ask about cost - these scans aren't cheap, and insurance often doesn't cover them for psychiatric indications.

If the answers are vague ("We'll personalize your treatment"), be skeptical. If there's a linked supplement company, be more skeptical.

That doesn't mean dismiss it entirely. Just weigh it appropriately. Understand that you're making a decision with limited evidence, potentially significant cost, and uncertain benefit.

What I'd prioritize instead: a thorough medical workup for treatable contributors to psychiatric symptoms. Adequate trials of evidence-based treatments - both medication and therapy. Addressing sleep, exercise, diet, substance use. Building support systems. Sometimes the basics work, when given time and attention.

And if you've done all that and you're still suffering? Then talk to your doctor about options. Maybe experimental treatments in the context of clinical trials. Maybe alternative approaches if you've exhausted conventional ones. But go in informed about what's proven and what's hopeful speculation.

The light in my office always seems slightly too harsh this time of year. I noticed that yesterday, talking to someone about whether to try a fourth antidepressant. Couldn't tell you why I thought of it. Sometimes your mind wanders to irrelevant details when the relevant ones feel inadequate.

---

Dr. Terry Nguyen is a family medicine physician who makes house calls in Sydney. He writes about medicine, humanity, and the spaces where certainty fails.