Our Mission
PeopleBeatingCancer supports an evidence-based integrative approach to cancer care. For most newly diagnosed patients, FDA-approved therapies form the foundation of treatment, while evidence-based complementary therapies may help reduce side effects and improve survivorship.
Click the orange button to the right to learn more.
- You are here:
- Home »
- Blog »
- Marfan Syndrome »
- Can Artificial Intelligence Recognize Marfan Syndrome From a Face? What the Research Shows
Can Artificial Intelligence Recognize Marfan Syndrome From a Face? What the Research Shows
Can Artificial Intelligence Recognize Marfan Syndrome From a Face? What the Research Shows. Artificial intelligence may soon help physicians recognize Marfan syndrome from an ordinary facial photograph. Although AI cannot diagnose Marfan syndrome by itself, new research suggests that computer vision can identify characteristic facial features with remarkable accuracy. Early recognition could lead to faster diagnosis and potentially life-saving monitoring of the aorta.
I think it is difficult at best, dangerous at worst to think that facial recognition of Marfan Syndrome can be identified 98.5% of the time, as the study linked below professes. Let me explain my thinking.
What Facial Features Are Common in Marfan Syndrome?
Examples
- long narrow face
- deep-set eyes
- down-slanting eyes
- underdeveloped cheekbones
- recessed jaw
- elongated skull
These facial features have long been part of the clinical assessment of Marfan syndrome, although they are not sufficiently sensitive to diagnose the condition on their own.
How Accurate Is AI?
| Study | Accuracy |
|---|---|
| Heliyon 2024 | 98.5% |
| False positives | 0% |
| False negatives | 2% |
AI Does Not Replace a Medical Diagnosis
The diagnosis still depends upon
- family history
- genetic testing
- Ghent criteria
- echocardiogram
- eye examination
- skeletal measurements
In my experience as a long-term cancer survivor who reads and writes about all things health, I have learned that a formal diagnosis of just about any serious condition has consequences. Those consequences can be pro or con, positive or negative. In our society, a health condition leads, inevitably, to treatment. And treatment, all forms of treatment, have risks.
For example, any/all cancers that are diagnosed at stage 0, or pre-cancer, think DCIS, Barrett’s Esophagus or colon polyps. may not benefit from treatment. Treatment can cause side effects.
The excerpt below sums up my thinking about Marfan Syndrome in general.
My point, therefore, is that if a person is diagnosed or identified as having “Marfan Syndrome,” he/she then starts down the path of treatment. In my experience as a person who was recently diagnosed with Marfan Syndrome, the therapy path was heart surgery.
A geneticist at the Cleveland Clinic told me, over a Zoom call, that I have Marfan Syndrome. The geneticist then told me to meet with a surgeon who would try to convince me to undergo heart surgery. There was no evaluation of where I am on the “mild to severe” Marfan spectrum.
My father is another example of my thinking. While my dad was never diagnosed as having Marfan Syndrome, he did exhibit many of the typical symptoms of MS.
Dad developed scoliosis in his thirties. His feet were a mess. Dad had surgery to fix a heart valve in his seventies. Dad died of a heart attack at the age of 89.
Where are I and my dad and I on the Marfan spectrum? I would argue that undergoing open-heart surgery at the age of 64 (me) without knowing were I am on the spectrum would be a mistake. Treatment would expose me to a host of possible side effects.
Facial recognition of Marfan Syndrome, for both my dad and me, would be problematic.
For the record, I have an echocardiogram annually to keep an eye on my heart health. I keep a close eye on my aortic root and desending aorta with the understanding that Marfan Syndrome can cause serious heart issues that I have to be aware of.
Further, my diet is heart healthy. My daily exercise is moderate. I take a number of heart healthy supplements. The key metrics illustrated by my annual echo have not budged in years.
Can marfan syndrome be identified through facial recognition?
Marfan syndrome cannot be definitively diagnosed through facial recognition alone. However, certain facial features are commonly associated with Marfan syndrome and might be identified through a detailed physical examination, which could contribute to a clinical diagnosis. Some of these features include:
I am not any sort of medical professional. I am a long-term cancer survivor whose FDA-approved standard-of-care conventional therapies resulted in many, many short, long-term and late-stage side effects while providing no improvement in overall survival.
I reached complete remission after I underwent a non-FDA-approved, non-conventional therapy. But that’s another long story…
Are you somewhere on the Marfan Syndrome spectrum? How do you feel about facial recognition of marfan syndrome?
Thanks,
David Emerson
- Cancer Survivor
- Cancer Coach
- Director PeopleBeatingCancer
External authority links
Pilot study exploring artificial intelligence for facial-image-based diagnosis of Marfan syndrome
“Highlights
-
We explore Artificial Intelligence for detection of Marfan Disease (MFS) from simple facial images.
-
AI proves extremely effective at detection of MFS.
-
Overall accuracy is 98.5 % (0 false positives, 2 % false negatives).
-
Clinical usefulness is anticipated.
Background
Marfan Syndrome (MFS), a genetic disorder impacting connective tissue, manifests in a wide array of phenotypes which can affect numerous bodily systems, especially the thoracic aorta. The syndrome often presents distinct facial features that potentially allow for diagnostic clinical recognition. Herein, we explore the potential of Artificial Intelligence (AI) in diagnosing Marfan syndrome from ordinary facial images, as assessed by overall accuracy, F1 score, and area under the ROC curve…
Results
Overall accuracy was 98.5 % (0 % false positive, 2 % false negative). F1 score was 97 % for Marfan facies and 99 % for non-Marfan facies. Area under the ROC curve was 100 %.
Conclusion
An Artificial Intelligence (AI) program was able to distinguish Marfan from non-Marfan facial images (from ordinary on-line photographs) with an extremely high degree of accuracy. Clinical usefulness of this program is anticipated. However, due to the limited and preliminary nature of this work, this should be viewed as only a pilot study.”
To learn more:
Connective Tissue / Marfan Hub
- Marfan Syndrome – Need to Know
- Marfan Syndrome and Aortic Dilation: How to Reduce Cardiovascular Stress
- Marfan Syndrome and Exercise
To learn more about Marfan syndrome-
- Living a Long Life With Marfan Syndrome
- Ascending Aortic Aneurysm
- Can Supplements Help Support Connective Tissue Health?
- Athletes Living With Marfan Syndrome
- Marfan Bone Health
- Blood Pressure Stabilization
- Aortic Dilation
- Exercise with Marfan Syndrome
- Diet for Marfan Syndrome
- Resveratrol and Aortic Growth
- Cardiovascular toxicity after cancer
- About David Emerson
Cardiovascular & Survivorship
- Chemotherapy-induced cardiomyopathy
- Exercise during chemotherapy
- Sleep therapies for cancer survivors
Integrative Therapy Content
- Curcumin and cancer
- Omega-3 fatty acids and cancer
- Anti-inflammatory diet content
- Microbiome and cancer side effects
Chronic Disease / Inflammation
Supportive Care