Skip to content

AI tool automatically identifies different types of brain injury

May 29, 2020

This is a guest post by Steven Baylis, Partner at Lime Solicitors: 

 

The consequences of a head injury can be very subtle, as well as catastrophic. The brain is divided into numerous areas, each responsible for different functions. For example, the temporal lobe is responsible for processing auditory information, and damage to this area of the brain can result in hearing loss. This same part of the brain is also associated with memory and emotion. In the initial stages following a traumatic brain injury the focus may well be on life saving, with measures such as the need to reduce intra-cranial pressure being crucially important. In the initial stages of treatment a person can be unconscious, so AI may be able to play an important role in providing a prognosis to assist with early life saving intervention. As areas of damage to the brain may be very small, the intervention of AI may assist radiographers in identifying areas of damage. This may reduce pressures on radiographers and improve the speed of initial investigations and in turn the speed at which effective treatment can be delivered.

 

The algorithms employed with AI do seem capable of predicting quite accurately the likelihood of a patient dying. This will not necessarily assist with providing a prognosis for those patients who survive a traumatic brain injury. The effects of what objectively look similar brain injuries can result in differing symptoms and different levels of recovery. It may be the case that the very small detail into which AI can explore may be able to accurately predict what symptoms will be suffered; however, this has yet to be fully tested. A further complication is what is known as ‘brain plasticity’, which is effectively the ability of the brain to re-wire itself. This may vary from individual to individual, so this again may dilute the ability of AI to provide an accurate prognosis. All this potential for variation will probably mean that neuro-psychological testing will continue to play a crucial part in identifying which areas of cognitive function have been adversely affected by a brain injury. This in turn will impact on the treatment and rehabilitation that is appropriate. Whilst AI might be capable of giving some general guidance it remains to be seen how accurately it will be capable of predicting all the potentially subtle effects of a traumatic brain injury and whether this might remove or reduce the need for individually administered assessments. As algorithms become more sophisticated and more research is undertaken, it will be interesting to see what developments occur. With the recent interest in head injuries in sport  more research is being undertaken all the time, which should only help with advancing the accuracy of AI in terms of providing an accurate prognosis for patients suffering from a traumatic brain injury.

 

The earlier a prognosis can be determined, then the sooner targeted treatment can be administered, which can only be beneficial to patients. It is too early to determine how much input AI will be able to have in terms of reducing the input of medical experts such as neurosurgeons and neurologists. I would be surprised, given the often unique and subtly variable presentation of symptoms whether medical expert input at present can be significantly reduced. Many more trials will need to be undertaken to determine the accuracy of the algorithms. If there is a subsequent error than I would anticipate the threshold for negligence being the same as it is currently for medical experts. A doctor is not negligent if they acted in accordance with a responsible body of opinion. This raises the interesting prospect of how could AI, based on the algorithmic approach not be acting in accordance with a responsible body of opinion? It will be very interesting to see how the impact of AI develops in this area of medicine.

No comments yet

What are you thinking?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: