Full-body MRI with diffusion-weighted imaging (DWI) combines anatomic precision with a functional 'lump detector' — no radiation, no contrast injection — and in a cohort of 1,000 scanned individuals produced only two false positives while identifying 8 intracranial aneurysms.
2
Mammography sensitivity collapses from ~95% in fatty breasts to ~55% in dense breasts — roughly coin-flip accuracy — which means a significant fraction of women receiving a 'clear' mammogram actually have inadequate surveillance and need ultrasound or MRI.
3
The prostate-cancer work-up has been quietly transformed: PSA alone is unreliable (elevated by cancer, prostatitis, or benign enlargement), 4Kscore blood test adds specificity, and DWI-MRI can often make biopsy unnecessary when results are clearly negative.
4
MRI resolution and capability are almost entirely determined by hardware tuning and physics knowledge, not by magnet size — a meticulously calibrated 1.5T scanner can outperform a stock 3T system on whole-body DWI for the same reason a tuned 1.5-liter F1 engine outperforms an unoptimized 5-liter block.
Protocols
Concrete recipes — what, when, how much, and why
6 items
Full-body MRI with DWI/DWIBS for cancer screening
WhatA single-session (~55 min) whole-body MRI incorporating multiple anatomic sequences (T1, T2, fat-sat) plus isotropic diffusion-weighted imaging from head to foot. No radiation, no contrast injection required. Produces an anatomic map at near-millimeter resolution combined with a functional cellular-density map (DWIBS rotating display).
WhenAttia's framing: appropriate for health-conscious individuals who want a comprehensive cancer-screening baseline; not a population-level recommendation. Repeating scans allows detection of change over time (Delta imaging), which is even more sensitive than a single cross-section.
Dose~55 minutes in the scanner. Significant warmth/heat discomfort expected in final 10-15 minutes (equivalent SAR to ~4 hours of mobile phone use concentrated across the body).
For whomAdults seeking comprehensive cancer surveillance, especially those with family histories of cancer or aneurysm. Contraindicated with metallic implants, pacemakers. Patients should disclose all metal implants and claustrophobia.
WhyNo ionizing radiation; covers organs poorly served by existing screening modalities (prostate, kidneys, liver, brain); catches incidental aneurysms, early fibrosis, occult malignancies. The false-positive rate in Attariwala's cohort of 1,000 was two cases, both non-malignant.
CaveatsMRI quality is highly scanner/operator-dependent. Not standardized across institutions. Small calcifications (e.g., in breast tissue) are a blind spot for MRI — mammography still needed in women. Claustrophobia is a barrier. Clinical-grade DWI is not available off-the-shelf from most scanners.
Attariwala's hardware approach: ~50 custom options added at purchase that the vendor had 'never seen,' followed by iterative tuning of 150+ parameters per sequence with academic MRI physicists. The rotating DWIBS display — a proprietary visualization tool — is adapted from PET/CT workflow. The goal was to unify head-to-foot coverage in a single pass without gaps in the spinal signal (a common artifact in 3T scanners where the shorter wavelength causes field inhomogeneity at body transitions). Machine learning is expected to accelerate image reading, starting with paired/longitudinal Delta comparisons.
Mechanism
DWI signal: Fick's law of diffusion — water molecules move stochastically in normal tissue; tight cell packing in tumors restricts that motion, creating restricted-diffusion signal (apparent diffusion coefficient, ADC). T1 highlights fat (anatomic detail); T2+fat-sat highlights water/edema; susceptibility-weighted imaging (SWI) differentiates oxygenated arterial from deoxygenated venous blood.
What I kind of thought okay what would I want if I'm a patient what would I want to know — well number one I want to know that my brain's okay, I want to know that the arteries in my brain are okay... and then I basically want to say whenever I go into any test of any kind the first question on their mind is do I have cancer yes or no.
Also said
“In our thousand people that we did — the fascinating thing about this — we actually had two false positives... one was a male with asymmetric breast tissue... and the second case was actually a woman who basically had a seatbelt injury to her breast.”— Real-world false-positive rate from n=1,000 cohort: 2/1,000 (0.2%), both non-malignant on biopsy.
Prostate cancer screening: PSA to 4Kscore to DWI MRI decision pathway
WhatStep 1: PSA test. If elevated, PSA alone is insufficient because it can reflect prostatitis or BPH. Step 2: 4Kscore blood test — provides probability of having aggressive/metastatic prostate cancer over a 20-year horizon, not just the presence of any cancer. Step 3: If 4Kscore is elevated (e.g., 16-20% risk), obtain DWI MRI of the prostate before considering biopsy. If MRI is clearly negative, biopsy can often be deferred. If MRI shows a suspicious dense lesion, targeted biopsy follows.
WhenMen with elevated PSA. Attia describes using this three-step pathway in clinical practice, with the MRI step allowing two patients to avoid biopsy entirely.
DoseDedicated prostate MRI at high-quality centers: typically 40+ minutes for prostate alone at 3T; Attariwala's whole-body protocol captures the prostate as part of the full scan in 55 minutes.
For whomMen with elevated PSA, high 4Kscore, or strong family history of prostate cancer. Men who wish to avoid biopsy and are willing to accept ongoing surveillance imaging.
WhyProstate cancer is near-universal at autopsy in elderly men — distinguishing aggressive from indolent is the clinical goal, not simply finding any cancer. DWI MRI maps cellular density spatially and allows size/density change tracking over serial scans.
CaveatsDWI MRI quality for prostate is highly dependent on scanner tuning. Off-the-shelf 3T scanners in standard clinical settings may not provide the DWI resolution needed for confident negative reads. The de facto standard in Australia, UK, and Scandinavia; not yet routine in US or Canada.
Attariwala describes the shift: 'Up until MRI with DWI came into effect there was no real way' to separate indolent from aggressive prostate cancer. The old pathway was biopsy on any elevated PSA, which carries infection, bleeding, and psychological harm risks. Many men are now asking to 'see what's going on' and monitor — if there's a lesion and it's not growing, watchful waiting; if it's growing, treat. This mirrors the breast DWI data from UCSF and Memorial Sloan Kettering showing DWI sensitivity equivalent to contrast-enhanced MRI. The 4Kscore result Attia uses: ~16-20% risk of metastatic cancer over 20 years is 'high' in this context.
Mechanism
Prostate tissue is poorly vascularized — it uptakes very little FDG, making PET/CT 'almost entirely useless' for prostate cancer. DWI instead directly detects cellular density, which correlates with Gleason grade and aggressiveness. Combined with the 4Kscore (which incorporates PSA, free PSA, intact PSA, hK2, age, DRE, and biopsy history), the pathway substantially reduces both over-biopsy and under-diagnosis.
Diffusion-weighted image of the prostate coupled with the more advanced molecular tests the 4k as an example of a blood test when combined have totally revolutionized the way we think about prostate cancer — so we now have a blood test that produces much better resolution than just a PSA but more importantly we have this MRI and even in a practice as small as mine I have had two patients for whom PSA is high, 4k comes back high... we run them through this MRI and the answer is nope that's totally fine.
Breast imaging protocol: mammography plus density-stratified add-on imaging
WhatStep 1: Mammography for all women at appropriate screening age (40+ or 50+ depending on jurisdiction). Step 2: Determine breast density from the mammogram report — dense vs. fatty. Step 3: If tissue is dense, add breast ultrasound or breast MRI (ideally DWI-MRI). Mammography-only surveillance in a dense-breasted woman has ~55% sensitivity — inadequate for reliable clearance.
WhenScreening intervals: either annual or biennial (academic debate). Serial comparisons markedly improve sensitivity in dense tissue — subtle temporal change is more detectable than any single image.
DoseMammogram radiation dose: ~0.05 mSv (negligible). Breast MRI adds the full MRI protocol time with no radiation. Ultrasound: no radiation, but operator-dependent and limited to single 2D slices swept through volume.
For whomAll women of screening age. Women with dense breast tissue should specifically request or demand additional imaging. Younger premenopausal women are more likely to have dense tissue. BRCA carriers and high-risk women have separate more aggressive protocols (not detailed in this episode).
WhySensitivity/specificity trade-off is breast-density dependent. In fatty breasts: sensitivity ~95%+, specificity high — mammography is reliable. In dense breasts: sensitivity ~55%, meaning 45 of 100 women with cancer get a false negative on mammography alone.
CaveatsMRI has its own blind spots: small calcifications (often the first sign of ductal carcinoma in situ) are poorly visualized on MRI but well seen on mammogram. Mammography and MRI are complementary, not interchangeable. MRI alone cannot replace mammography.
Attariwala explains the biological mechanism: 'breast tissue is composed of normal subcutaneous tissue which is mainly fat and as well as glandular tissue and so when women are in their childbearing age it's almost all glandular tissue to produce milk for eventually feeding a baby and as women then go through menopause that glandular tissue can invariably involute.' Why some women retain glandular tissue through menopause is unknown. The DWI data from UCSF and Memorial Sloan Kettering now shows DWI sensitivity equivalent to contrast-enhanced breast MRI — important because contrast MRI is burdensome and gadolinium carries its own risk profile, while DWI needs no injection.
Mechanism
X-ray/mammography physics: low-energy photons pass freely through fat (low density) but are absorbed by glandular tissue and calcifications (higher density). Dense glandular tissue absorbs the photon beam, masking any tumor behind it — the same physical phenomenon that makes bones appear white on standard x-rays.
A single mammogram on a dense woman it's sensitivity is about fifty five percent — it's actually quite poor — whereas on a woman who actually has fatty tissue it's very high, it can actually be over ninety five percent.
Also said
“Many states and I think they're actually over 38 and possibly soon to become a federal law in United States is going to require that the very first line on a mammogram report is going to be that the woman's breast tissue is dense limiting mammographic sensitivity... to actually kind of find out — is this good enough.”— Actionable: women should read the first line of their mammogram report for the density disclosure.
“Basically any DWI with MRI you actually are as sensitive as giving a contrast injection breast MRI — well but that's diffusion done right and the problem is out of the box it's not always done right.”— Caveat on DWI for breast: requires expert calibration, not standard off-the-shelf DWI.
Incidental brain aneurysm: when to screen and what to do when found
WhatMRA (magnetic resonance angiography) of the brain can be added to any brain MRI without contrast injection or additional radiation. When an intracranial aneurysm is found incidentally: evaluate size and location, consider elective treatment (coiling or surgical clipping), initiate family screening given genetic clustering.
WhenStrong family history of intracranial or other vascular aneurysm deaths, especially in relatives who died young. Can be bundled with whole-body MRI. The MRA sequence uses the iron in flowing arterial blood as the signal — no gadolinium needed.
DoseMRA adds time to a brain MRI; in Attariwala's protocol it is included in the whole-body 55-minute scan. Elective treatment: coiling is endovascular (lower risk), clipping is surgical. Risk of elective treatment far lower than rupture mortality (93-95%).
For whomAnyone with a first- or second-degree relative who died of intracranial aneurysm rupture. Attia argues health-conscious adults with resources should consider it as part of comprehensive screening.
WhyRupture is nearly always fatal — 93-95% mortality. Elective detection allows intervention at a time and place of choosing. Familial clustering is strong: when one aneurysm is found, relatives should be screened.
CaveatsPublished prevalence (~0.5-0.8%) means ~995-997 per 1,000 will have a normal scan — important for framing in shared decision-making. US insurance may not cover MRA without clear family history indication; Attia fought for six months to get one patient covered, ultimately denied.
Attia recounts a patient with two first/second-degree relatives dead from aneurysms in different vascular territories: 'Our team did a bit of work on this and actually found evidence that there really was potentially a genetic component here... the woman just paid out of pocket for it... nine thousand dollars. And it was negative.' The cash pay cost in the US for MRA is approximately $9,000 at time of recording. Attariwala's clinic in Vancouver charges substantially less (implied but not stated). The scan can be done simultaneously with the brain component of a whole-body protocol.
Mechanism
MRA technique: exciting arterial blood flowing in a specific direction (toward the brain) using directional gradient pulses — the iron in hemoglobin generates signal that differs from stationary tissue, producing a vascular map. SWI (susceptibility-weighted imaging) differentiates oxygenated arterial from deoxygenated venous blood using their different magnetic susceptibility.
The mortality of a ruptured aneurysm is over 93 to 95 percent so most people don't make it whereas when you do find them earlier all sorts of options such as coiling where you can actually treat it or clipping — and so that's actually one of the real powers of being able to kind of see what's going on without any injection or anything like that.
Also said
“When we actually scanned a thousand people we actually found eight intracranial brain aneurysms — so point eight percent. That's higher than I would have guessed.”— Empirical prevalence in a self-selected screened population; higher than commonly cited population estimates.
Radiation dose minimization: choosing the lowest-radiation modality when clinically equivalent
WhatFor any diagnostic imaging order, the first question should be: do I need anatomic information, functional information, or both? Second: can the clinical question be answered by a radiation-free modality (MRI or ultrasound) before resorting to CT or nuclear medicine? CT should not be the default for organs or questions where MRI provides equivalent or superior information.
WhenEvery imaging decision, especially in patients under 40, female patients, and anyone with anticipated cumulative radiation exposure (oncology follow-up, trauma protocols, annual CT scans).
DoseRadiation reference points: chest x-ray ~0.1 mSv; mammogram ~0.05 mSv; abdominal CT ~8-10 mSv; whole-body PET/CT ~30-40 mSv (close to 80% of 50 mSv/year NRC recommended limit). MRI/ultrasound: zero radiation.
For whomAll patients, but priority emphasis: children, adolescent females, women of childbearing age, patients expected to need repeated surveillance imaging.
WhyDNA damage from ionizing radiation is cumulative and dose-dependent. Risk of cancer induction is highest in young females. The principle is ALARA — as low as reasonably achievable. The FDA recommendation of 50 mSv/year is a regulatory limit, not a target.
CaveatsThere are clinical situations where the information urgency and CT speed are irreplaceable (trauma, acute PE, acute stroke). The goal is to minimize unnecessary exposure, not to refuse CT when it is genuinely the best tool.
Attariwala notes that background radiation varies significantly by geography and altitude: 'higher you go actually get more cosmic radiation... pilots for example actually when they wear glasses they actually have to block the UV radiation... their route matters.' A transatlantic flight via the North Pole exposes passengers to significantly more radiation than the extra distance would suggest because of reduced ozone shielding. An online calculator is available for pilots/flight attendants to calculate their dose (NRC requirement). Denver residents at elevation get ~6-7 mSv/year background vs ~2-3 mSv at sea level.
Mechanism
Ionizing radiation damages DNA, and the probability of cancer induction is proportional to dose. Young females are most vulnerable because egg DNA is frozen from embryonic development — the same DNA that began meiosis in utero has been sitting without repair for the entire lifetime of the individual, making it more susceptible to double-strand breaks than continuously regenerating sperm.
We're actually starting to see that a lot of patients individuals are like look I know about the potential damaging effects of x-ray radiation therefore I don't want a lot of x-rays CT type scans.
Also said
“The younger you are the greater the risk of cancer induction from CT scanners... and the sex as well matters so females are actually more sensitive to radiation than men.”— Two key modifiers of radiation risk that clinical imaging decisions should account for.
False positive triage: the two-step rule before biopsy
WhatWhen an imaging study returns an uncertain finding ('shadow,' asymmetric tissue, unusual density), the protocol should be: (1) obtain dedicated imaging at the highest-quality center available for that organ/question, not generic follow-up imaging at the same center; (2) only proceed to biopsy when both anatomic AND functional imaging (DWI or equivalent) show concordant abnormality. One positive alone is insufficient for biopsy decision.
WhenAny incidental finding that is not clearly benign. Especially important for pancreatic shadows (where biopsy via ERCP carries severe complication risk), breast asymmetry (where biopsy is lower-risk but emotional toll is high), and prostate (where 4Kscore + MRI should precede biopsy).
DoseSecond-opinion imaging: days to weeks depending on access. The emotional toll of an uncertain finding is substantial — expediting the second imaging reduces harm.
For whomAnyone with an incidental finding on any imaging study. Particularly relevant for organs where biopsy complication risk is high (pancreas) or where tissue is difficult to access.
WhyAttia describes the 'call I'm most afraid of' as a pancreatic shadow — because the biopsy pathway (ERCP) carries serious complication risk, and the famous Stanford case (ERCP complication cascade leading to death in a woman without cancer) illustrates the harm of premature biopsy.
CaveatsThis protocol requires access to expert imaging — not universally available. The practical alternative for patients without access to centers like Attariwala's is to seek referral to academic medical centers with organ-specific imaging expertise.
Attariwala describes the pancreas specifically: 'when we're looking at organs particularly the pancreas or any of the visceral solid organs we're looking about seven different filters looking at different ways top to bottom front to back to really be able to see what's going on.' The pancreatic duct is visible on MRI — duct obstruction or dilation is often the first sign of pancreatic cancer. The false-positive protocol from his own cohort: male with asymmetric breast tissue (biopsy: normal male glandular tissue) and woman with seatbelt-injury scar (biopsy: fibrous scar tissue with trapped fluid). Both cases would have been resolved by better history-taking before biopsy.
The harm of a false positive — the harm is that we see something that turns out in the long run to not be cancer but in the process of going down the diagnostic pathway to get there you are either physically harmed by something we do subsequently for example another biopsy... and of course the emotional toll it takes on you.
What's new
Personal practice updates, fresh positions, predictions
7 items
DWI/DWIBS: water-motion imaging as a whole-body 'lump detector'
~second half
Diffusion-weighted imaging with background subtraction (DWIBS) measures water displacement at two points separated by ~60 microseconds. Where water cannot move freely — because tightly packed cancer cells restrict it — the tissue appears abnormal. This maps onto the classical surgical exam: cancer tissue is distinctly firmer than surrounding normal tissue.
Why this matters: Brings the binary yes/no logic of nuclear medicine (PET) into the radiation-free MRI platform. One plus one equals three: functional information layered onto exquisite anatomic detail.
Background
Standard MRI is purely anatomic. DWI derives from Fick's law of diffusion — restriction of normal stochastic water motion is the functional signal. Attariwala spent a decade custom-tuning hardware and sequences to make whole-body DWI clinically actionable.
Attariwala explains the parallel to PET/CT: 'just like the PET CT where that famous 1+1=3 is exactly what we're doing — and the beauty of it is there's absolutely no radiation.' The rotating DWIBS display he built is adapted from nuclear medicine PET/CT workflow — a transparent-body view where areas of high cellular density appear as dark spots against background. Currently this visualization software is proprietary and not commercially available. Standardization of DWI signal intensity across scanners (the QIBA/RSNA effort) is underway organ by organ — liver, prostate, and breast — but not yet complete, meaning off-the-shelf DWI cannot be assumed equivalent to what Attariwala produces.
What DWI or diffusion-weighted imaging does is basically like you tell women for breast cancer feel for a lump and basically when you're feeling for a lump that's a hard spot and so the reason it's hard is because you have this increased cellular density and so with that increased cellular density that's where water is restricted from moving.
Also said
“We're actually looking at water at two points in time... within about 60 microseconds of water motion and so by doing that what happens is that you first look at water at one point in time and then you look at it at the second point in time and if it hasn't moved it's because it's not allowed to move — it's effectively trapped between walls.”— Precise mechanism of the DWI signal: restricted diffusion = cellular density = potential malignancy.
“We're actually combining the anatomic and functional and that's where just like the PET CT where that famous 1+1 equals 3 is exactly what we're doing and the beauty of it is there's absolutely no radiation.”— Frames DWI MRI as a radiation-free replacement/complement for PET/CT.
Mammography sensitivity is ~55% in dense breasts — close to coin-flip
~second half, slice 2
In women with dense (glandular) breast tissue — common in premenopausal women and in a fraction of postmenopausal women for unknown reasons — mammography sensitivity drops to approximately 55%, because low-energy x-ray cannot penetrate glandular tissue adequately. By contrast, in fatty breasts sensitivity can exceed 95%. Women who do not know their breast density may be receiving false reassurance from a 'normal' mammogram.
Why this matters: Many women and their physicians treat a negative mammogram as a reliable clearance. If breast tissue is dense, it is roughly as informative as a coin flip for cancer detection.
Background
More than 38 US states had enacted laws requiring breast density notification at time of recording; federal legislation was pending. Density is reported on the mammogram — but rarely explained to patients in terms of what it means for test accuracy.
Attariwala explains the biology: premenopausal breast tissue is primarily glandular (for lactation), and in some women that glandular component persists through and after menopause rather than involuting to fat. 'Mammograms are very very good at shining through fat' — they see calcifications clearly because dense calcium absorbs x-ray photons; but glandular tissue itself absorbs the low-energy x-ray and becomes opaque, masking tumors behind it. When tissue is dense, the next steps are breast ultrasound or breast MRI, the latter now with emerging DWI data from UCSF and Memorial Sloan Kettering showing DWI sensitivity equivalent to contrast-enhanced breast MRI.
A single mammogram on a dense woman it's sensitivity is about fifty five percent — it's actually quite poor — whereas on a woman who actually has fatty tissue it's very high, it can actually be over ninety five percent.
Also said
“Many states and I think they're actually over 38 and possibly soon to become a federal law in United States is going to require that the very first line on a mammogram report is going to be that the woman's breast tissue is dense limiting mammographic sensitivity or the breast tissue is almost entirely fat.”— Policy context: mandatory density disclosure is now (or soon) required — patients should check their reports.
Prostate MRI with DWI is replacing biopsy in Australia and parts of Europe
~slice 4
In Australia and across Scandinavia and the UK, DWI MRI of the prostate has become standard-of-care before or instead of biopsy. PSA elevation alone triggers biopsy in the traditional pathway; MRI allows differentiation between high-density lesions (suspicious) and normal-density tissue, reducing unnecessary biopsies.
Why this matters: The US and Canada are behind. In countries with single-payer systems, MRI has been cost-justified because it eliminates a large fraction of unnecessary biopsies, which themselves carry bleeding, infection, and psychological harm.
Background
PSA can be elevated by three distinct causes: prostate cancer, prostatitis (inflammation), and benign prostatic hyperplasia. Without imaging, a high PSA alone is a poor predictor of true cancer. The 4Kscore adds specificity over PSA alone.
Attariwala: 'In Australia in Europe particularly UK Scandinavia it's actually the de facto standard — basically almost all men are actually getting screened with MRI.' The argument is that all men either die with or from prostate cancer, and the clinical imperative is to separate slow indolent tumors (watchful waiting) from aggressive ones requiring intervention. MRI with DWI gives the spatial resolution to see if a lesion is growing between scans, making watchful waiting protocol more rigorous without repeated biopsies. Attia describes two patients with high PSA and high 4Kscore — 16–20% estimated risk of metastatic cancer over 20 years — who had clearly normal MRIs, allowing avoidance of biopsy.
Diffusion-weighted image of the prostate coupled with the more advanced molecular tests the 4k as an example of a blood test when combined have totally revolutionized the way we think about prostate cancer so we now have a blood test that produces much better resolution than just a PSA but more importantly we have this MRI.
Also said
“I have had two patients for whom PSA is high, 4k comes back high so these are patients who now have a 20% chance maybe 16 18 20% chance of having cancer in their prostate or having metastatic cancer over the next two decades... we run them through this MRI and the answer is nope that's totally fine.”— Clinical use case: MRI negative result avoids biopsy in high-PSA patients.
MRI quality is almost entirely a function of hardware tuning, not Tesla strength
~slice 3
Hospitals brag about 3T or 4T magnets, but the 1.5T scanner at Attariwala's clinic produces whole-body DWI images superior to dedicated prostate MRIs done at 3T or 4T centers. The reason: 150+ tunable parameters per sequence, most left at vendor defaults in clinical settings. A 3T magnet has a ~15 cm wavelength (about the width of the human head), creating penetration and homogeneity issues in the abdomen; 1.5T has a ~30 cm wavelength and penetrates large-body regions more evenly.
Why this matters: Patients fly across the country and pay large sums for '3T MRI' assuming higher Tesla = better imaging. This is often false, especially for whole-body DWI.
Background
The parallel is the McLaren MP4/4 F1 car: a 1.5-liter engine producing 1,200 HP. Size is irrelevant if you do not know how to engineer for the objective. Attariwala spent ~10 years with MRI physicists at academic centers iteratively tuning sequences.
Attariwala purchased the hardware with ~50 custom options the vendor had 'never seen' — a complete educated guess at what combination would enable whole-body connectivity from head to foot on a single continuous table pass. He tested protocols on himself as a subject, using his nuclear medicine training to specify the functional signal he wanted and his radiology training to specify the anatomic resolution. The resulting signal-to-noise ratio enables isotropic (1x1x1 mm cube) DWI of the brain and near-isotropic in the body — resolution that commercially available systems 'not yet' match.
Part of it is it's like the biggest problem with me is that I'm an engineer in medicine... I kind of started the back end and say okay what do I want to know, what I want to see and how do I make it work and so I kind of work backwards and then this is where you put your engineering or physics hat on.
Also said
“With a 3 Tesla magnet one of the things that actually commonly happens is that when you look at the wavelength of a 3 Tesla... the 3 Tesla wavelength is roughly 15 centimeters — so it's the width of your head — 1.5 is roughly 30 centimeters so it's the width of mostly players' shoulders. So as a result you actually start to get a lot more penetration with a lower field magnet.”— The physics reason a 1.5T outperforms 3T for whole-body work.
“There are roughly about a hundred and fifty parameters per t1 t2 fat saturation sequence that you can actually adjust to make it work the way you need it and most of the time what commonly happens almost everywhere is that you'll have the vendor will actually come in and set the standard parameters and from there on it's kind of like let's make it as simple as possible, push a button — and that's kind of not what I do.”— Why default-configuration MRI scanners produce inferior images regardless of Tesla rating.
Intracranial aneurysm prevalence ~0.8% in a screened population
~slice 3
In a cohort of 1,000 individuals who received whole-body MRI including a magnetic resonance angiogram (MRA) of the brain vessels, Attariwala found 8 intracranial aneurysms — a 0.8% prevalence, higher than published population estimates (which may undercount because autopsy data is incomplete). Ruptured aneurysm mortality is 93–95%; elective treatment of incidental aneurysms is comparatively low-risk.
Why this matters: The argument for incidental aneurysm screening: the condition is silent, treatment when incidental is well-tolerated, and rupture is almost uniformly fatal. An MRA of the brain adds minimal time to a whole-body scan.
Background
Published population estimates are generally lower than 0.8%; Attariwala hypothesizes the difference reflects survival bias in the autopsy-derived data. He also observes strong familial clustering — when one aneurysm is found, family members present for screening.
Attia describes a patient with a strong family history of aneurysm deaths — mother and an additional relative — in different vascular territories, suggesting a connective tissue or genetic basis. Attia's team spent six months fighting her US insurer to cover an MRA before the woman paid out of pocket — the cash price was $9,000, and the scan was negative. Attariwala's coiling and clipping options provide effective elective treatment: 'when you do find them earlier all sorts of options such as coiling where you can actually treat it or clipping and so that's actually one of the real powers of being able to kind of see what's going on without any injection or anything like that you can see the exquisite detail of the arteries.'
When we actually scanned a thousand people we actually found eight intracranial brain aneurysms — so point eight percent. That's higher than I would have guessed.
Also said
“The mortality of a ruptured aneurysm is over 93 to 95 percent so most people don't make it whereas when you do find them earlier all sorts of options such as coiling where you can actually treat it or clipping.”— The clinical asymmetry justifying incidental screening: silent and treatable vs. ruptured and nearly universally fatal.
Radiation dosimetry: age and sex matter more than most patients realize
~slice 1
Radiation sensitivity is highest in young females — specifically around age 12 when egg DNA that was 'frozen' during embryonic development begins releasing into circulation. The Columbia landmark paper that forced radiology to reassess CT radiation risk found that women are more sensitive than men at every age, and younger patients carry higher cancer-induction risk per millisievert than older patients.
Why this matters: Most radiation discussions focus on cumulative dose (50 mSv/year NRC limit) without weighting for age or sex. The clinical implication: defer CT in young women whenever ultrasound or MRI is feasible.
Background
Understanding of radiation-induced cancer induction largely derives from Hiroshima/Nagasaki follow-up cohorts and the Chernobyl/Fukushima exposures. The Columbia paper quantified cancer induction rates from CT-specific dose patterns.
Attariwala explains the egg biology: 'the egg was actually produced during embryonic stage and as that DNA is effectively frozen in time... around twelve is the worst time for females.' The contrast with male germ cells: sperm are continuously produced with fresh DNA, so the dose-per-sperm is much lower and the DNA 'age' is always recent. This biological asymmetry is why pediatric and adolescent radiology protocols use ALARA (as low as reasonably achievable) principles most aggressively in young girls.
Females are actually more sensitive to radiation than men... the egg was actually produced during embryonic stage and as that DNA is effectively frozen in time... twelve is sort of the worst time for females.
PET/CT has three major blind spots where MRI excels
~slice 4
FDG-PET/CT is unreliable in the brain (high baseline glucose uptake masks lesions), kidneys (glucose excreted there accumulates and obscures), and prostate (poorly vascularized, low glucose uptake). MRI with DWI is superior in all three organs. PET/CT retains advantage in high-metabolic tumors like lymphoma or very aggressive carcinomas outside those organs.
Why this matters: Clinicians often default to whole-body PET/CT for cancer staging without accounting for these anatomic blind spots. A well-executed DWI MRI addresses the three largest gaps.
The brain in particular is a glucose hog — that's all it can use unless you're in ketosis and then the other problem is that the glucose is then excreted by the kidneys and so the kidneys now become difficult to see because they're actually full of glucose... and in the prostate, prostate is very very poorly perfused and as a result it doesn't get a lot of glucose coming to it and so that is actually almost entirely useless with FDG.
Recommendations
Products, supplements, and tools mentioned in the episode
The only known clinic performing whole-body DWI/DWIBS at the resolution and technical calibration described in the episode. Attia has sent numerous patients and has himself been scanned multiple times.
Attariwala started AIM as a private MRI clinic to have a machine he could 'play with' as an engineer. The evolved product, PreNuVu, is positioned as 'putting the power of preventive medicine into patients' hands.' The whole-body scan takes ~55 minutes, uses a custom-configured 1.5T magnet with ~50 non-standard hardware options, and produces near-isotropic DWI resolution that commercially available systems cannot match.
Aim is a private MRI company and we basically put in the MRI machine so I could play with it... we've actually now kind of moved it into PreNuVu and what PreNuVu is about is actually to basically sort of put the power of preventive medicine into patients' hands.
4Kscore blood test for prostate cancer risk stratification
Tool
Blood test that provides a probability of aggressive or metastatic prostate cancer over a 20-year horizon, incorporating PSA, free PSA, intact PSA, hK2, age, DRE result, and prior biopsy history. Attia uses this as the step between elevated PSA and ordering an MRI.
Attia describes the 4Kscore as producing 'much better resolution than just a PSA' because it quantifies a specific risk (metastatic cancer over 20 years) rather than just detecting a biochemical marker. In two of his patients, a 4Kscore showing 16-20% risk triggered an MRI that came back clearly negative — allowing biopsy avoidance.
vs alternatives
PSA alone is non-specific: elevated by prostatitis, BPH, and cancer equally. 4Kscore adds specificity but still requires imaging to localize. DWI MRI provides the spatial answer PSA and 4Kscore cannot.
The 4k as an example of a blood test — when combined have totally revolutionized the way we think about prostate cancer. So we now have a blood test that produces much better resolution than just a PSA.
NRC / IAEA radiation dose calculator for flight crew and travelers
Tool
Online calculator for estimating radiation dose from flight routes, altitude, and geography. FAA/NRC require pilots and flight attendants to calculate their occupational radiation dose. Attia notes he has calculated his own East-West Coast travel doses and found them low, but that Polar routes dramatically increase exposure.
Ozone at polar latitudes is thinner, removing the primary shield against cosmic radiation. A North Pole-crossing flight exposes passengers to significantly more radiation than the extra distance would suggest — a non-linear increase. The NRC annual limit is 50 mSv. Background radiation at sea level is approximately 2-3 mSv/year; at Denver altitude approximately 6-7 mSv/year.
There's actually a calculator available online for people to actually determine how much radiation they're getting from exposure and it's actually required that pilots and flight attendants calculate their dose.
Breast density self-check: reading the first line of your mammogram report
Practice
In 38+ US states (and pending federal law at time of recording), mammogram reports must state breast tissue density in the first line. Women should check this: 'dense tissue limiting mammographic sensitivity' triggers the need for additional imaging (ultrasound or MRI); 'almost entirely fat' means mammography is reliable.
The density notification law was passed precisely because women and many physicians did not know to ask. Attariwala emphasizes that a negative mammogram in a woman with dense breasts is not reassuring — it may simply mean the mammogram could not see through the tissue. The add-on modality decision: breast ultrasound is widely available but limited to 2D slices and operator-dependent; breast MRI (ideally DWI-calibrated) is more sensitive but less available.
Many states and I think they're actually over 38 and possibly soon to become a federal law in United States is going to require that the very first line on a mammogram report is going to be that the woman's breast tissue is dense limiting mammographic sensitivity or the breast tissue is almost entirely fat in which case mammograms are helpful — because that actually will really allow women to determine was this test good enough.
Lines worth pulling out — contrarian, specific, or perfectly phrased
6 items
The mortality of a ruptured aneurysm is over 93 to 95 percent so most people don't make it whereas when you do find them earlier all sorts of options such as coiling where you can actually treat it or clipping.
Starkest clinical asymmetry in the episode: silent and treatable versus ruptured and nearly universally fatal. The core argument for preventive aneurysm screening.
A single mammogram on a dense woman it's sensitivity is about fifty five percent — it's actually quite poor — whereas on a woman who actually has fatty tissue it's very high, it can actually be over ninety five percent.
The single most clinically actionable statistic in the episode. Coin-flip accuracy for the patients most likely to receive false reassurance.
You can make a test that is a hundred percent sensitive if you're willing to have zero percent specificity — you could send a little card that says you have cancer and you show it to every single person you meet... you will never have a false negative.
The sharpest possible explanation of sensitivity vs. specificity: memorable, precise, and non-mathematical.
The prostate — prostate is very very poorly perfused and as a result it doesn't get a lot of glucose coming to it and so that is actually almost entirely useless with FDG.
PET/CT has three major clinical blind spots — brain, kidney, prostate — and this is the most clinically consequential for men. DWI MRI resolves all three.
There are roughly about a hundred and fifty parameters per t1 t2 fat saturation sequence that you can actually adjust to make it work the way you need it and most of the time what commonly happens almost everywhere is that you'll have the vendor come in and set the standard parameters — push a button — and that's kind of not what I do.
Explains why 'I got an MRI' tells you almost nothing about image quality. Vendor-default configuration leaves most diagnostic potential on the table.
When we actually scanned a thousand people we actually found eight intracranial brain aneurysms — so point eight percent. That's higher than I would have guessed.
Empirical data from a real screened cohort — more compelling than population estimates — and from a host who routinely cites the rarity of benefit in preventive screening.
Sign in to share feedback
Tell us if this brief hit the mark or missed it — feedback feeds back into the next iteration of the prompt.
Reading is free for everyone. A free account adds the personal layer: save protocols, follow experts, and see how the other experts weigh in on this same topic.
Educational summary of the cited expert source — not medical advice. Open the source recording linked above and consult a qualified physician before acting on any protocol.
