3D measurement of brain arteries: What can we learn?

ArticleDetailDownload PDF

Professor Tetsuya Takahashi of the Hiroshima University, and his colleagues Dr Naoyuki Kitamura and Mr Hiroki Yoshimura, are investigating the possibility that arteries elongate and kink with age. They are particularly interested in understanding whether this finding could explain some neurological disorders. To get more accurate measurements, the team have developed an innovative artery measuring tool that can trace arteries on images obtained by magnetic resonance angiography (MRA) in three dimensions. Their novel tool provides new insights on vessel structure and function.

Neurovascular compression (NVC) syndrome is caused by arteries putting direct pressure on the brain stem and cranial nerves (the nerves that emerge directly from the brain), and results in a variety of neurological disorders. The most common arteries responsible for the syndrome are the superior cerebellar artery, which supplies blood to the superior part of the cerebellum, pressure from which can cause long-term severe facial pain known as trigeminal neuralgia; and the anterior inferior cerebellar artery, or posterior inferior cerebellar artery (one of the three main arteries that supply blood to the cerebellum), which can cause facial spasm when putting pressure on the relative facial nerves. Magnetic resonance angiography (MRA) is the current standard diagnostic tool to confirm such contact between arteries and nerves, and in certain persistent cases, surgical decompression can potentially help to resolve the symptoms.

Neurological disorders can arise when arteries put direct pressure on the brain stem and cranial nerves – the nerves that emerge directly from the brain. My_box_pra/depositphoto.com

What happens to our arteries with age?

Ageing is a complicated process which influences body functions and organ structure, including blood vessels. It is well known that the prevalence of vascular disease increases with age. The changes that have been observed and studied in arteries are mostly related to arterial diameter, the thickness of the artery wall (atherosclerosis and stenosis), and the consistency of the arterial wall, including analysis of the different artery wall constituents: collagen, elastin, and smooth muscle cells. Far less attention has been paid to the changes that occur in relation to the length of the arteries and their possible consequences.

“Increases in artery length with age may cause the artery to twist and turn, potentially resulting in nerve compression.”

Professor Tetsuya Takahashi, Dr Naoyuki Kitamura, and Mr Hiroki Yoshimura, at Hiroshima University, set out to investigate this further. As Takahashi explains, ‘We wondered whether arteries elongate during ageing, and in particular around the brain stem where several nerves and arteries are coming and going together in a very narrow space’. Arteries that are in contact with nerves in that area can often sharply change their direction, so that when observed using MRA they have a hairpin-like appearance.

Magnetic resonance angiography (MRA) of a healthy subject showing normal appearance of bilateral vertebral arteries (yellow arrows). 1. medulla oblongata 2. cerebellar hemisphere A case of hypoglossal nerve palsy. The distorted left vertebral artery (yellow arrows) applies pressure to the hypoglossal nerve (red arrow). 1. medulla oblongata 2. cerebellar hemisphere 3. hypoglossal nerve

The team are working on the hypothesis that possible increases in artery length with age may cause the artery to twist and turn, potentially resulting in nerve compression.

A three-dimensional approach

With the aim of taking more accurate measurements, the Hiroshima International University team used an arterial-length measuring tool developed by L Pixel Corporation (Tokyo, Japan). Their tool was applied to 3D-images of the arteries obtained by MR examination. The software was designed to automatically track the centre of each artery’s cross-section along the long axis of the artery, mark each branching point of the arteries around the brain stem, and finally measure the length of each section between branching points.

Their method has the advantage of three-dimensional imaging and measurement, compared to the 2D measuring methods used previously.

A new view

The study involved 1,000 consecutive patients over 40 years of age who underwent head MRA and magnetic resonance imaging (MRI) examination as a follow-up after cerebral infarction, cerebral haemorrhage, cerebral aneurysm, persistent headache, or as a brain screening for vascular disease, a practice very common in Japan. The average patient age was 66.8 ± 12.9 years, and the proportion of women in the population studied was 55.2%.

The researchers measured the length of the vertebrobasilar arteries (the arteries that provide blood to the brain) and their branches. Their newly developed software allowed the team to make more accurate measurements and comparisons of the lengths of the arteries between various age groups. The researchers also evaluated the involvement of the different arterial sections in brain stem compression in all 1,000 patients. The presence of arterial compression of the brain stem was studied and confirmed by observing and documenting the loss of symmetry of the brain stem in the axial MRA images. The arteries responsible for the compression were also identified on specially processed MRA images.

“The team found that changes in the arterial route, rather than an increase in arterial length, are more critical for vascular compression.”

Arteries get longer with age

Out of the 1,000 patient MRAs studied, 138 showed signs of brain stem compression, but none of them had any symptoms caused by cranial nerve compression. The team also found a small but significant elongation in cerebral arteries with age: an increase of 1.5% to 2.6% per decade of life. In addition, it was shown that arteries compressed the brain stem more frequently in the elderly than in the younger population. However, the researchers believe that the higher frequency of arterial compression of the brain stem in older years might not be due to the elongation of the arteries. Instead, they suggest that the compression could be related to changes in the arterial route by kinking – something that takes place over time.


When classified by age, the presence of brain stem compression was 1.1%, 2.2%, 3.1%, 4.1%, and 3.3% for subjects in their 40s, 50s, 60s, 70s, and 80s, respectively; however, the differences were not statistically significant. The proportion of the presence of brain stem compression was shown to be higher in the cases with arterial elongation than in those without.

Takahashi’s findings suggest that elongated arteries could be at least partially responsible for the brain stem compression observed in their patients. Although the number of patients with brain stem compression tended to be higher among older adults, the differences in arterial length between the different age categories were small. Notably, the researchers found that changes in the arterial route were more critical for vascular compression of the brain stem and the cranial nerves, rather than the increased length of the artery. It is the kinking of arteries and not their elongation that is thought to be the primary cause of nerve compression by the arteries.

Future research

The team is now looking to upgrade the LOOKREC system (as depicted in the image below) into an AI diagnostic tool. Their aim is to develop an innovative AI system capable of generating a radiological MRI/MRA report with minimal radiologist involvement. To achieve this goal, the team will incorporate the growth technology of AI to their diagnostic and measurement routines in addition to the preceding teleradiology service (the transmission of radiological images from one location to another for the purposes of sharing them). This will enable them to automatically identify and mark the section of the artery that is being investigated and eventually measure the length of each section between branching points. Such a development will help to provide more accuracy in the measurement of the elongation of the arteries. Excitingly, it offers a novel method to precisely identify kinking of arteries with a view to early diagnosis and treatment of nerve compression.

A conceptual image of the LOOKREC service built on Google cloud platform, developed by MNES Inc. The authorised user can upload and download medical images regardless of time and place.
Could your tool be applied to other conditions, such as vascular brain disease?
Theoretically, our method is applicable to other long tracts such as the nerves or urinary tract. However, no pathological condition has been reported in which the reduction or increase in the length of these tracts is pathogenic.




Research Objectives

The Hiroshima University research team investigate the idea that arteries elongate with age, and whether this makes them turn, or kink.


MNES Inc, Kasumi clinic


Tetsuya Takahashi , MD, PhD, is a professor at the Department of Rehabilitation, Hiroshima University. Professor Takahashi has been involved in diagnosing, treating, and researching a diverse array of neurological disorders at the Department of Clinical Neuroscience and Therapeutics of Hiroshima University.

Naoyuki Kitamura , MD, is a radiologist and used to be a classmate of Professor Takahashi at Hiroshima University. He founded NMES Inc in 2000 and Kasumi clinic in 2015. He is a pioneer in the field of teleradiology.

Mr Hiroki Yoshimura is a medical student at Hiroshima University. He and his colleagues are working under Dr Kitamura at the LAIME laboratory. LAIME is an informal group of more than 50 Japanese university students developing python code for artificial intelligence, with support from MNES Inc. The team have successfully participated in Kaggle competitions.

555–36 Kurosegakuendai
Higashihiroshima City
Hiroshima 739–2695, Japan

E: t-taka@hirokoku-u.ac.jp
T: +81 823 70 4574
W: mnes.life/products

Related posts.


Leave a Comment

Your email address will not be published. Required fields are marked *

Share this article.

Share on facebook
Share on twitter
Share on whatsapp
Share on linkedin
Share on reddit
Share on email