When it comes to improving the health and lives of our children, any investment we make pays back in a myriad of ways.
One way, for example, is taking the AI technology behind such recent innovations as self-driving cars and applying it to medical technology has great potential, both in improving preventative medicine, the ability of professionals to recognize abuse, and aid in research that will help improve the quality of life for children with disabilities.
AII Can Help Contextualize Medical History to Discover Child Abuse
Part of the duty of many medical professionals, including some nurses, is to report to authorities if they suspect that a child is being abused. However it can be very difficult to identify what constitutes a pattern of abuse and what doesn’t.
There are certain types of bone fractures common to child abuse victims, however a single incident isn’t enough to raise legitimate suspicion. It takes a history of similar injuries to constitute and abusive pattern, and even then other circumstances such as bone disease could be responsible. Medical professionals must be extremely careful when they question caregivers about injuries and about deciding whether or not to make a report.
This is one specific case where AI can be extremely useful. By teaching an algorithm to identify patterns over a long medical history that might include different treatment facilities and medical professionals, detection of child abuse and the accuracy of reports can be much improved. It likely wouldn’t — and shouldn’t — replace the judgement of medical professionals when examining the data, but it can help flag cases worthy of investigation.
Safeguarding Vulnerable Populations
Access to preventative care is a key component of a healthy population. However, the radiation and cost of scans often cause people to hesitate before receiving them. People in rural areas also struggle to access the facilities required to ensure adequate medical care. In the same way, children, a vulnerable section of the population, are positioned to suffer more acutely from lacking preventative care.
Among many growing concerns in the healthcare industry is access to care for people who live in rural and remote areas. Emergencies are much more difficult to deal with when access to facilities and technology is limited by distance. As the world becomes more interconnected, however, opportunities have opened up to connect people with specialists remotely, transfer information quickly, and update treatment information across multiple healthcare providers simultaneously.
Teleradiology is part of the growing telehealth trend, aided by smart devices and patient self-reporting. Machine learning is a key part of the process that collects and analyzes the data from patients who self report, as well as aggregating data from different treatment centers and specialists. It can also be used to better analyze the specific needs of rural communities based on the health issues that people from those locations are experiencing. Chatbots based on machine learning are one telehealth solution, able to provide answers to common questions, receive self-reporting from patients, and connect them with specialists remotely.
From a government standpoint, such technology would, ideally, allow them to allocate healthcare resources more effectively for matters of public health or for departments such as Veterans’ Affairs.
Reducing the Dangers of Care
Treatments and diagnosis methods that then contribute to long-term health issues have been a concern for a long while. Patients and their families sometimes have to make difficult decisions, especially when children need care that is potentially harmful.
The radiation involved in scans such as the CT scan, for example, can cause long-term harm if a patient needs enough scans over their lifetime. Parents might rightly be concerned about exposing their children to radiation to examine a complex bone fracture. Scientists have been working on this issue, and in 2012 developed a scan that can be done with a much smaller radiation dose.
AI and Understanding: Helping Improve Quality of Life for People With Disabilities
Preventative care and quality of life care are similar but distinct concepts — both help people avoid emergency care, and decrease the burdens on patients. Preventative care often focuses on detecting and diagnosing early, and taking steps to ensure that healthcare issues don’t happen. Quality of life care accepts that a condition is irreversible, focusing instead on treating patients with dignity and enabling them to live with disabilities with as much independence, self-determination and comfort as possible.
Researchers have been experimenting with combining machine learning algorithms and MRI brain scans in order to detect autism in children earlier in their lives. Insights into how infants experience the world and how their senses develop have given scientists a baseline. Machine learning can compare imaging results to what we know about infant brains in order to help us understand which children experience the world in fundamentally different ways, and how.
These are early-stage findings, and while useful for scientists, it may be some time before there is a reliable system in place — backed by larger studies — to be of real use for families. The research, however, shows promising results that could, one day, allow earlier detection of lifelong disabilities like autism. Earlier detection means earlier treatment, which has the potential to significantly improve the life of people with disabilities.
Radiology is Producing Safer, More Accurate Tools
The more effective detection and diagnosis methods become, the more we can help everyone — but especially members of vulnerable populations. Children with limited access to care, with disabilities, with terminal illnesses, and who suffer from abuse, will all benefit from technology that draws better connections between results, and from diagnosis techniques that are less dangerous.
About the author
Brooke Faulkner is a mom and writer in the beautiful Pacific Northwest. She loves researching the current state of medicine and sharing her findings with other families. You can find more of her writing on twitter or at contently.