In many areas of society, from healthcare to big business, companies are turning to artificial intelligence (AI) to solve pressing problems. Within the medical sphere, researchers recently developed an automated, deep-learning (DL) AI resource to help measure cerebral ventricle volume on MRIs for pediatric patients. In fact, explains Scienmag, this new tool completes its job in just under a half-hour. By being able to evaluate and track this volume, researchers can better identify and treat patients with hydrocephalus, a rare condition characterized by fluid accumulation in the ventricles. Read the full study findings in the Journal of Neurosurgery.
So to first understand what this new tool is, it’s important to understand the concepts of deep-learning (DL) and artificial intelligence (AI). According to BuiltIn, AI is a:
wide-ranging branch of computer science concerned with building smart machines capable of performing tasks that typically require human intelligence.
Investopedia expands on this through its explanation of deep-learning as:
an artificial intelligence (AI) function that imitates the workings of the human brain in processing data and creating patterns for use in decision making. Deep learning is a subset of machine learning in artificial intelligence that has networks capable of learning unsupervised from data that is unstructured or unlabeled.
Thus, artificial intelligence allows us to create machines or technology that can perform certain tasks, which we normally believe can only be achieved through human intelligence and logic. Deep-learning builds upon AI by allowing the technology to learn by example and make decisions using patterns and algorithms.
DL-AI and Hydrocephalus
Typically, to diagnose hydrocephalus, doctors must use neuroimaging to discover any enlarged ventricles. Additionally, doctors track clinical symptoms. Next, hydrocephalus pressure is relieved through placing a shunt, allowing for excess fluid to drain away from the brain. Then, doctors track and monitor patients to determine the efficacy of the shunt. Throughout this experience, doctors pay close attention to ventricular volume, which allows for doctors to make the most informed decision regarding patient care. However, monitoring this can be invasive, difficult, and time-consuming. As a result, this deep-learning AI could offer a more efficient option to evaluate a patient’s health while also streamlining their care.
In developing the new model, researchers sourced data from 200 pediatric patients with acute obstructive hydrocephalus. For this study, pediatric patients were defined as those younger than 22 years old. From these patients, researchers chose sets of T2-weighted MRIs, as well as 3D T1-weighted MRIs. Researchers also chose similar MRIs from a control group.
Next, researchers created a deep-learning AI model which can calculate ventricular volume and also delineate ventricular borders on imagery. This process was compared to manual segmentation and volume calculation, as well as FreeSurfer software. Ultimately, the deep-learning AI process was found to be incredibly accurate compared to other models, as well as more efficient.
Admittedly, this research is only preliminary, and additional research needs to be done. However, the results thus far are incredibly promising and may highlight a new and more effective way to diagnose this condition.
Our body produces cerebrospinal fluid (CSF), and some is absorbed into the bloodstream. Typically, CSF prevents brain injury, removes waste, maintains pressure, and keeps the brain buoyant. However, when too much CSF is produced and not enough is absorbed, it results in hydrocephalus, a rare condition characterized by fluid accumulation in the brain’s ventricles. Causes of hydrocephalus include obstruction, poor absorption, or simple overproduction. In patients with hydrocephalus, the extra CSF puts excess pressure on the brain, causing damage. Symptoms vary, but may include:
- Nausea and vomiting
- Unusually large head, or rapid head size increase
- Bulging or tender soft spot on the head
- Poor muscle tone and coordination
- Failure to thrive
- Memory loss or confusion (adult patients)
- Vision impairment (adult patients)
- Loss of bladder control (adult patients)
- Changes in thinking or reasoning (adult patients)
- Loss of balance and coordination (adult patients)
Learn more about hydrocephalus.