With the help of artificial intelligence, scientists from the University of Texas at Austin and the Genome Center in New York determine the genes that shape the human skeleton from shoulder width to leg length.


At the heart of the study were 39,000 medical images from those included in Biobank in the UK. By applying the deep learning models to 31,221 full-body dual-energy X-ray absorptiometry images, the researchers were able to extract 23 different phenotypes obtained from images involving bone length and hip and shoulder width.

All skeletal proportions are highly hereditary.

They compared the measurements with each person's genetic sequence. In doing so, they found 145 points in the genome that control the proportions of the skeleton. The points are enriched with genes that regulate skeletal development, but also have a direct bearing on major musculoskeletal diseases.

OpenAI boss: Sorry, artificial intelligence will really take jobs

Studies have shown that people who have a higher ratio between hip width and height are more likely to develop osteoarthritis or have hip pain framar.bg.

When the ratio of the length of the femur to height is greater, then arthritis and other problems with the knees occur. Back pain is characteristic of people when there is a higher ratio between the length of the torso and its height.

"The disorders in question develop from the biomechanical strain on the joints throughout a person's life. The proportions of the skeleton affect everything. From our walk to the way we sit. This also determines the fact that they are risk factors for various diseases," said study co-author Aiskarest Kuhn.


The team's work presents a roadmap that shows the relationship between specific genes and the length of the skeleton of different parts of the body.

The results of the study are also important in terms of understanding evolution. For example, several genetic segments that control skeletal proportions overlap more than expected with regions of the genome that also exist in great apes and many vertebrates, but are very different in humans. This provides a genomic justification for differences in human skeletal anatomy.

Farewell AI, organoid intelligence (OI) looms

Last but not least, the study is essential in terms of the role of artificial intelligence. As emphasized by Assoc. Vagish Narasiman, leader of the multidisciplinary team: "Our research is a powerful demonstration of the impact of artificial intelligence in medicine. This is particularly true for analysing and qualitatively defining image data and merging all this information with health records and genetic data quickly and on a very large scale."

Phenotypic and polygenic risk analyses established specific connectivity between hip and knee osteoarthritis. And this is one of the leading causes of disability in adults in the United States, the researchers emphasize.

The study provides more genomic evidence of evolutionary change in hand-leg proportions and hip width in humans.

Billionaires Will Transplant Their Brains into Robots to Live Forever

Of particular importance is the fact that the study opens the veil of the evolutionary past on the one hand, and on the other opens a window into the future, when doctors will be able to more accurately predict the risks of developing diseases later in the life of their patients.

The study's findings were published in the journal Science under the title "The Genetic Architecture and Evolution of the Human Skeletal Form."

artificial intelligence