Maya Trutschl, an 18-year-old from Shreveport, won the $25,000 Davidson Fellows Scholarship for engineering a system to detect pressure injuries before they become visible. Her solution uses machine learning and thermal cameras that cost under $130 to predict risk and alert nurses when intervention is needed. Bedsores contribute to approximately 60,000 deaths annually in the US, highlighting the significance of early detection systems like Maya’s.
Written by the local model on 2026-09-16,
using this article's own text rather than the other coverage of the
same event (that is the story summary below).
At an age when most teenagers might have their minds on crushes, fashion or the latest album releases, Shreveport resident Maya Trutschl was thinking about a problem in hospital ICUs: how to stop bedsores before they appear.
uncertain
they → have → bedsores
In the ICU, patients lie motionless for hours.
asserted
patients → lie → hours
Their skin may show no signs of damage.
uncertain
skin → show → damage
But beneath the surface, tissue is already breaking down.
asserted
tissue → break → surface
By the time a nurse notices the wound, the invisible injury has already begun.
asserted
injury → notice → wound
The 18-year-old decided to fix this.
asserted
old → decide → this
And she did!
asserted
she → do → ?
The teenager has just won a $25,000 Davidson Fellows Scholarship for engineering a system that detects pressure injuries before they become visible.
asserted
they → win → injuries
“I am profoundly honoured to be recognised as a Davidson Fellow and to become part of such an inspiring community of curious, dedicated young people
asserted
I → honour → people
,” Maya Trutschl said in a statement.
asserted
Trutschl → say → statement
“This achievement motivates me to continue pursuing research, and I hope it encourages other students to chase even the smallest questions that fascinate them.
asserted
that → motivate → them
”
The silent injury nobody talks about
Bedsores may not sound like a pressing problem, but pressure injuries contribute to an estimated 60,000 deaths in the US each year.
uncertain
injuries → talk → US
This toll is more than the lives most cancers claim.
uncertain
cancers → claim → lives
About 2.5 million Americans develop bedsores each year, according to the Cleveland Clinic.
uncertain
Americans → develop → Clinic
They contribute to roughly 60,000 deaths and cost the healthcare system over $26 billion.
asserted
They → contribute → billion
And unfortunately, the patients who suffer them most cannot advocate for themselves.
asserted
who → suffer → themselves
This is not just limited to ICU patients, but also to people at home with limited mobility and homebound people cared for by family members with no medical training.
asserted
This → limit → training
To prevent bedsores, patients are repositioned on a fixed schedule.
asserted
patients → prevent → schedule
Even a small delay or missed schedule can increase the risk of developing a bedsore.
asserted
delay → miss → bedsore
By the time visible signs appear, subdermal tissue will have already begun to deteriorate, and the window to prevent permanent damage starts closing.
Using machine learning and thermal cameras
Maya Trutschl built a solution for this problem, and it addresses two problems.
asserted
it → appear → problems
A computer programme predicts the risk of pressure injuries, while a thermal camera monitors the patient’s position and alerts the nurse when intervention is required.
asserted
intervention → predict → nurse
First, she built a machine learning model using the MIMIC-IV dataset, a real-world collection of ICU patient records.
asserted
she → build → records
The dataset was severely imbalanced, containing about 3,900 pressure ulcer cases against 90,000 negative outcomes.
asserted
dataset → contain → outcomes
She combined SMOTE oversampling with undersampling and tested 10 classification models to prevent her predictions from ignoring the rare but critical cases.
asserted
She → combine → cases
She set up a thermal-sensing camera that costs less than $130.
asserted
that → set → 130
This camera automatically monitors whether a patient stays in one position for too long.
asserted
patient → monitor → long
When movement is needed, it alerts nurses.
asserted
it → need → nurses
In busy units, this alert has a significant impact.
asserted
alert → have → impact
“I tested it on real patients in a hospital ICU for more than 150 hours, where it tracked patient positioning with more than 99% accuracy.
asserted
it → test → accuracy
In short, my project catches an invisible injury before it starts, protecting patients and freeing nurses to spend their time where it's needed most,” Trutschl said.
asserted
Trutschl → catch → time
Maya Trutschl will attend the Massachusetts Institute of Technology this autumn.
asserted
Trutschl → attend → Technology
She plans to study aerospace engineering or computer science.
asserted
She → plan → engineering
She wants to continue building intelligent systems that solve ‘meaningful problems’.
asserted
that → want → problems
She secured the first spot in Embedded Systems at the Regeneron International Science and Engineering Fair and has been recognised as a 2026 Regeneron Science Talent Search Scholar and National STEM Festival Champion.
asserted
She → secure → Scholar
The teen has also published her research at an international conference in Copenhagen.
asserted
teen → publish → Copenhagen
While in high school, she founded a Girls Who Code club to bring computer science education to local middle school girls.
asserted
Who → found → girls
She also leads The Unity Network, an international research mentoring organisation that provides high school students with free resources, mentorship and opportunities to prepare for science competitions.
asserted
that → lead → competitions
She likes to bake in her free time.
asserted
She → like → time
“I bake everything from cookies to full cakes.
asserted
I → bake → cakes
I love that it's a way to wind down and to be creative.”
asserted
it → love → ?
The Aporia
