AI Can Now Predict Weight Gain in Patients with Schizophrenia and Depression
Table of Contents
By dr. Michael Lee, world Today News – November 2, 2023
In a stunning breakthrough that bridges the gap between neuroscience and mental healthcare, researchers have developed an artificial intelligence model capable of predicting future weight gain in individuals diagnosed with schizophrenia and depression.The system doesn’t rely on conventional metrics like diet or exercise, but rather analyzes MRI scans of the brain.
How the AI “Oracle” Works
The research,conducted by an international team,began with a interesting premise: can an AI learn to estimate a person’s weight simply by looking at their brain? The first step involved training a machine learning model using MRI images from healthy individuals. The AI was tasked with independently determining each person’s weight based solely on the brain scans. “And our algorithm does it quite well,” explains a psychiatrist involved in the study, based in Munich.
But the real innovation came when the researchers applied the model to the brain scans of patients with mental health conditions. “In these cases, our prognosis model made systematic errors,” says researcher Dr. Koutsouleris. “It incorrectly determined the weight of the corresponding patients.”
Specifically,in patients with schizophrenia,the AI consistently overestimated their weight. This wasn’t a flaw in the AI itself, but a reflection of neurological differences. Individuals with schizophrenia often have smaller volumes in certain brain regions – notably the anterior cerebral cortex,a key area involved in the reward system and,crucially,appetite regulation.The AI, having been trained on healthy brains, incorrectly associated smaller volume in this region with higher weight.
“This system largely controls our eating behavior,” Dr. Koutsouleris clarifies. “And our prediction model had previously learned from healthy people: less volume in these brain regions means higher weight.” The crucial point is that schizophrenia patients exhibit these smaller brain volumes early in their diagnosis, but don’t necessarily have a higher Body Mass Index (BMI) at that time.
The Predictive Power of the “BMI Gap”
The researchers didn’t stop there. They tracked the patients’ actual BMI over a year following their initial diagnosis and the AI’s weight assessment. The results were striking. “And then we see that those patients for whom our AI model had misjudged a BMI that was too high are actually gaining a lot of weight,” Dr. Koutsouleris reports. This pattern was notably pronounced in schizophrenia patients, but also observed in those with depression.
The key takeaway? The difference between the AI’s estimated BMI and the patient’s actual BMI – what the researchers call the “BMI gap” – is a powerful predictor of future weight gain. This suggests that neurological factors, detectable through brain scans, play a notable role in weight management for individuals with these mental health conditions.
This research opens up exciting possibilities for proactive intervention and personalized treatment plans, perhaps helping clinicians address weight gain before it becomes a serious health concern for their patients. Further research is underway to refine the model and explore its potential applications in other mental health conditions.
