Computers, facial recognition, acromegaly and early diagnosis.

From J D Faccinetti – Pituitary World News Co-founder  –  Technology could have a profound effect in the way acromegaly is diagnosed and hopefully prompt primary care doctors, dentist and other primary care health professionals to consider it on their list of possibilities when looking a patients with some of the related symptoms.

We know that if found early, many of the most debilitating effects of the disease can be avoided and the patient can be potentially cured. Because it might be theoretically possible to recognize the condition by looking at photographs, now with the titanic advances in face recognition technology, a practical tool could be developed to aid in recognizing the disease.

Today, on average, acromegaly takes 6 to 10 years to diagnose from the onset of the first symptoms. So a general practitioner that is aided by this type of technology to suspect that there might be something not quite right is – from my humble patient point of view – an exciting possibility for the tens of thousands of people running around undiagnosed.

A recent study published in PubMed, the National Library of Medicine from the National Institutes of Health, (NIH), discusses a fascinating study that concluded that a computer model can sort photographs of patients with Acromegaly from photographs of normal subjects. We provided a link to the abstract here: Early diagnosis of acromegaly: computers vs. clinicians – Miller, et al.

Another related article published in the Journal of Clinical Endocrinology and Metabolism discusses how the typical changes of the face caused by acromegaly might be recognized by face classification.   Both articles concluded that classification accuracy by software is higher than by medical experts or general internists, particularly in patients with mild features of acromegaly. Here’s a link to the article A Novel Approach to the Detection of Acromegaly: Accuracy of Diagnosis by Automatic Face Classification – Schneider, Kosilek, et al. 


© 2015, Pituitary World News. All rights reserved.


  1. I am 27(F) and stumbled upon my acromegaly diagnosis, as many of us do. After addressing “the wrong” symptoms for almost 2 years, it was a birthday selfie with my fiancé and a reminiscent browsing of Facebook that brought my attention to a change in my smile and difference in my bite.

    I consulted an orthodontist who, after viewing my X-rays, insisted I see an endocrinologist. I was diagnosed on July 26th, 2015 and had transphenoidal surgery at UCSF on Sept 15th, 2015 to remove most of my pituitary tumor.

    Advances in facial recognition software are rapidly being implemented and improved by social media companies. I can post a photo of myself and 5 friends to Facebook, and in seconds, each individual’s name/profile is automatically associated and linked with their faces for my approval.

    What if Facebook could tell you “something’s wrong!”

    What if Facebook could have sent me a notification that I had acromegaly based on the growth of my jaw, the skin-tags on my neck or the massive proportional changes in my hands and feet?
    Wouldn’t that be something!

    I hope that advancements in facial recognition software and selfies alike will prove helpful to others in future diagnosis of acromegaly.


Please enter your comment!
Please enter your name here