Cookie Preferences
By clicking, you agree to store cookies on your device to enhance navigation, analyze usage, and support marketing. More Info
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
June 17, 2026
For centuries, we’ve called the eyes the "windows to the soul," but for modern neurologists, they are quite literally a window into the brain. The retina and the central nervous system share the same embryonic origins, developing from the same neural tissue in the womb. Because of this deep biological connection, the back of your eye acts as a non-invasive map of your brain's health, displaying a complex web of nerves and blood vessels that can (theoretically!) mirror certain neurodevelopmental conditions.
Recently, a buzz rippled through the mental health community when a study published in partnership with Seoul National University Bundang Hospital claimed a massive breakthrough. Researchers developed an Artificial Intelligence (AI) model that could screen children for Attention-Deficit/Hyperactivity Disorder (ADHD) using nothing more than a simple retinal photograph. The study, which prospectively recruited children from Severance Hospital and Eunpyeong St. Mary’s Hospital, produced results that were staggering: the AI reportedly achieved an accuracy rate of 96.9%!
In the world of medical testing, scientists use a metric called AUROC (Area Under the Receiver Operating Characteristic) to measure how well a test works.
An AUROC of 96.9% is a near-perfect score, suggesting a tool is ready for immediate, real-world deployment. While headlines promised a revolution in mental health screening, a deeper look into this research and the study’s design has exposed that this 96.9% AUROC was more likely evidence of a flawed methodology rather than a biological reality.
To build their screening tool, researchers analyzed over 1,100 retinal images using a digital pipeline called AutoMorph and a machine-learning model known as XGBoost. The AI was trained to hunt for physical signals of the "Dopamine Connection." Dopamine is the primary neurotransmitter involved in ADHD, but it is also essential to the eye. It regulates synaptic formation, retinal blood flow, and vascular endothelial regulation. Because dopamine dysregulation influences how blood vessels grow and remodel, the study hypothesized that an ADHD brain would leave a unique "fingerprint" on the retinal vasculature, resulting in denser, thicker vessel structures.
On paper, the logic was sound: use AI to spot the subtle vascular remodeling caused by dopaminergic shifts. But a closer look at the investigation revealed that the AI wasn't just spotting ADHD; it was over-indexing on technical noise.
The most significant "smoking gun" flagged by critics is a massive temporal mismatch. In other words, there was a severe disparity in the timeframes and conditions under which the retinal images for the two comparison groups were collected. For an AI to learn a biological condition, it must compare groups under identical technical conditions. Instead, this study created a time-traveling dataset:
A scientific study is only as reliable as its control group. The control in any experiment acts as a baseline against which the study group is compared. In this case, the control group should be composed of children without any neurodevelopmental disorders, or of “typically developing” children.
In this study, the control group wasn't composed of healthy children from the community. Instead, they were patients visiting a tertiary ophthalmology clinic. Children visiting a specialist eye hospital are rarely "typical." They are there because they have symptomatic eye issues. This introduced a massive selection bias involving three major confounders:
When training AI, you must never allow the "test questions" to leak into the "study material." The researchers, however, committed a fundamental violation of machine learning hygiene known as Eye-to-Eye Data Leakage. The study split the data by the eye rather than by the participant.
Human eyes are highly correlated; the left eye is a near-mirror of the right. If a child's left eye was used for training and their right eye was used for testing, the AI was effectively "cheating." Instead of learning the general traits of ADHD, the model was potentially memorizing individuals. This error artificially balloons accuracy metrics.
The true test of medical AI is diagnostic specificity, or differential diagnosis. This refers to the ability to tell one condition apart from another. While the model claimed 96.9% accuracy against a flawed control group, its performance collapsed when faced with real-world complexity.
When the researchers asked the AI to differentiate between ADHD and Autism Spectrum Disorder (ASD), the accuracy plummeted to a poor 63% AUROC. In real-world clinical settings, an accuracy of 63% is dangerously close to a 50% coin flip. Since ADHD frequently co-occurs with ASD, anxiety, or intellectual disabilities, an AI that cannot handle these "clinical differentials" is functionally useless in a doctor's office. The failure at this stage proves the model was likely detecting technical quirks of the dataset rather than a unique biological marker for ADHD.
To move from the lab to the clinic, we must establish a foundation built on rigor rather than high-speed data scraping. Moving forward, we must demand these 3 Pillars of Trusted Medical AI :
The dream of a quick eye scan to diagnose ADHD is not dead, but it must be rescued from "fast science" shortcuts and buzzy headlines.
Choi H, Hong J, Kang HG, Park MH, Ha S, Lee J, Yoon S, Kim D, Park YR, Cheon KA. Retinal fundus imaging as biomarker for ADHD using machine learning for screening and visual attention stratification. NPJ Digit Med. 2025 Mar 17;8(1):164. doi: 10.1038/s41746-025-01547-9. PMID: 40097590; PMCID: PMC11914053.
.png)
The Neuroeconomic Perspective
Neuroeconomics combines neuroscience, psychology, and economics to understand how people make decisions. Neuroeconomic studies suggest that brain regions responsible for evaluating risk and reward, including the prefrontal cortex and dopamine pathways, function differently in individuals with ADHD. These insights are crucial for developing more tailored interventions. For example, understanding how ADHD affects reward processing might inform strategies that help individuals resist impulsive choices or increase motivation for delayed rewards.
Understanding Decision-Making in ADHD
We know that decision-making is a sophisticated process involving various cognitive procedures. It’s not just about choosing between options but also about how to weigh risks, rewards, and potential future outcomes; Attention, motivation, and cognitive control are core to this process. For individuals with ADHD, however, this neural framework is affected by impairments in attention and impulse control, often resulting in “delay discounting”—the tendency to prefer smaller, immediate rewards over larger, delayed ones.
This propensity for impulsive decisions is more than a personal challenge; it has broader societal and economic implications. Previous studies have shown that these tendencies in ADHD can lead to issues in academics, work, finances, and personal relationships, emphasizing the need for targeted support and interventions.
Implications and Future Directions
This review highlights a need for continued research to bridge the gaps in understanding how ADHD-specific cognitive deficits influence decision-making. Viewing ADHD through a neuroeconomic lens clarifies how cognitive and neural differences affect decision-making, often leading to impulsive choices with economic and social impacts. This perspective opens doors to more effective interventions, improving decision-making for individuals with ADHD. Future policies informed by this approach could enhance support and reduce associated societal costs.
Typically, clinicians rely on both subjective and objective observations, patient interviews and questionnaires, as well as reports from family and (in the case of children) parents and teachers, in order to diagnose ADHD.
A group of researchers are aiming to find a diagnostic test that is purely objective and utilizes recent technological advancements. The method they developed involves analyzing videos of children in outpatient settings, focusing on their movements. The study included 96 children, half of whom had ADHD and half who did not.
This new method could potentially provide a more objective way to diagnose ADHD, reducing the reliance on subjective observations and reports. It can help doctors make more accurate diagnoses, ensuring that those who need help get it and that those who don't aren't misdiagnosed.
Background:
Our understanding of Attention-deficit/hyperactivity disorder (ADHD) has grown and evolved considerably since it first appeared in the DSM-II as “Hyperkinetic Reaction of Childhood.” This study aimed to find the disorder’s placement within the modern psychopathology classification systems like the Hierarchical Taxonomy Of Psychopathology (HiTOP).
The HiTOP model aims to address limitations of traditional classification systems for mental illness, such as the DSM-5 and ICD-10, by organizing psychopathology according to evidence from research on observable patterns of mental health problems.. Is ADHD best categorized under externalizing conditions, neurodevelopmental disorders, or something else entirely? A recent study by Zheyue Peng, Kasey Stanton, Beatriz Dominguez-Alvarez, and Ashley L. Watts takes a closer look at this question using a symptom-focused approach.
The Study:
Traditionally, ADHD has been associated with externalizing behaviors, such as impulsivity and hyperactivity, or with neurodevelopmental traits, like cognitive delays. However, this study challenges the idea of placing ADHD into a single category. Instead, it maps ADHD symptoms across three major psychopathology spectra: externalizing, neurodevelopmental, and internalizing.
The findings reveal that ADHD symptoms don’t fit neatly into one box. For example, symptoms like impulsivity, poor school performance, and low perseverance were strongly associated with externalizing behaviors. On the other hand, cognitive disengagement (e.g., daydreaming, blank staring) and immaturity were closely linked to neurodevelopmental challenges. Interestingly, cognitive disengagement also showed ties to internalizing symptoms, such as anxiety or depression.
This research underscores the complexity of ADHD. Rather than treating ADHD as a single, unitary construct, the study advocates for a symptom-based approach to better understand and treat individuals. By acknowledging that ADHD symptoms relate to multiple psychopathology spectra, clinicians and researchers can move toward more nuanced classification systems and targeted interventions.
Conclusion:
Ultimately, this study highlights the need for modern systems to move beyond rigid categories and adopt a more flexible, symptom-focused framework for understanding ADHD’s place in psychopathology.
For centuries, we’ve called the eyes the "windows to the soul," but for modern neurologists, they are quite literally a window into the brain. The retina and the central nervous system share the same embryonic origins, developing from the same neural tissue in the womb. Because of this deep biological connection, the back of your eye acts as a non-invasive map of your brain's health, displaying a complex web of nerves and blood vessels that can (theoretically!) mirror certain neurodevelopmental conditions.
Recently, a buzz rippled through the mental health community when a study published in partnership with Seoul National University Bundang Hospital claimed a massive breakthrough. Researchers developed an Artificial Intelligence (AI) model that could screen children for Attention-Deficit/Hyperactivity Disorder (ADHD) using nothing more than a simple retinal photograph. The study, which prospectively recruited children from Severance Hospital and Eunpyeong St. Mary’s Hospital, produced results that were staggering: the AI reportedly achieved an accuracy rate of 96.9%!
In the world of medical testing, scientists use a metric called AUROC (Area Under the Receiver Operating Characteristic) to measure how well a test works.
An AUROC of 96.9% is a near-perfect score, suggesting a tool is ready for immediate, real-world deployment. While headlines promised a revolution in mental health screening, a deeper look into this research and the study’s design has exposed that this 96.9% AUROC was more likely evidence of a flawed methodology rather than a biological reality.
To build their screening tool, researchers analyzed over 1,100 retinal images using a digital pipeline called AutoMorph and a machine-learning model known as XGBoost. The AI was trained to hunt for physical signals of the "Dopamine Connection." Dopamine is the primary neurotransmitter involved in ADHD, but it is also essential to the eye. It regulates synaptic formation, retinal blood flow, and vascular endothelial regulation. Because dopamine dysregulation influences how blood vessels grow and remodel, the study hypothesized that an ADHD brain would leave a unique "fingerprint" on the retinal vasculature, resulting in denser, thicker vessel structures.
On paper, the logic was sound: use AI to spot the subtle vascular remodeling caused by dopaminergic shifts. But a closer look at the investigation revealed that the AI wasn't just spotting ADHD; it was over-indexing on technical noise.
The most significant "smoking gun" flagged by critics is a massive temporal mismatch. In other words, there was a severe disparity in the timeframes and conditions under which the retinal images for the two comparison groups were collected. For an AI to learn a biological condition, it must compare groups under identical technical conditions. Instead, this study created a time-traveling dataset:
A scientific study is only as reliable as its control group. The control in any experiment acts as a baseline against which the study group is compared. In this case, the control group should be composed of children without any neurodevelopmental disorders, or of “typically developing” children.
In this study, the control group wasn't composed of healthy children from the community. Instead, they were patients visiting a tertiary ophthalmology clinic. Children visiting a specialist eye hospital are rarely "typical." They are there because they have symptomatic eye issues. This introduced a massive selection bias involving three major confounders:
When training AI, you must never allow the "test questions" to leak into the "study material." The researchers, however, committed a fundamental violation of machine learning hygiene known as Eye-to-Eye Data Leakage. The study split the data by the eye rather than by the participant.
Human eyes are highly correlated; the left eye is a near-mirror of the right. If a child's left eye was used for training and their right eye was used for testing, the AI was effectively "cheating." Instead of learning the general traits of ADHD, the model was potentially memorizing individuals. This error artificially balloons accuracy metrics.
The true test of medical AI is diagnostic specificity, or differential diagnosis. This refers to the ability to tell one condition apart from another. While the model claimed 96.9% accuracy against a flawed control group, its performance collapsed when faced with real-world complexity.
When the researchers asked the AI to differentiate between ADHD and Autism Spectrum Disorder (ASD), the accuracy plummeted to a poor 63% AUROC. In real-world clinical settings, an accuracy of 63% is dangerously close to a 50% coin flip. Since ADHD frequently co-occurs with ASD, anxiety, or intellectual disabilities, an AI that cannot handle these "clinical differentials" is functionally useless in a doctor's office. The failure at this stage proves the model was likely detecting technical quirks of the dataset rather than a unique biological marker for ADHD.
To move from the lab to the clinic, we must establish a foundation built on rigor rather than high-speed data scraping. Moving forward, we must demand these 3 Pillars of Trusted Medical AI :
The dream of a quick eye scan to diagnose ADHD is not dead, but it must be rescued from "fast science" shortcuts and buzzy headlines.
Background:
One of the more persistent concerns among parents of children with ADHD is whether stimulant medications will stunt their child's growth. A large Israeli cohort study now offers some of the most rigorous reassurance to date, and its methodology sets it apart from earlier research.
The question has long been complicated by a more fundamental uncertainty: do growth differences in children with ADHD stem from the condition itself, from stimulant treatment, or from factors present before any medication is ever prescribed? Without a clear answer, clinicians and families have faced a genuine dilemma when weighing the benefits of stimulant therapy against potential long-term physical costs.
Most previous studies compounded this difficulty by comparing group-average heights, which ignores the crucial variable of genetic potential. A child who is short relative to the general population may simply have short parents. Failing to account for this introduces systematic bias and can make medications appear more harmful than they are.
The Study:
The Israeli research team addressed this directly. Using health records from a nationwide provider, they assembled a retrospective cohort of children born between 1995 and 2003, following them through 2023. This amount of time was long enough for all participants to have reached adult stature (defined as 17 or older for females, 19 or older for males). Their sample included 5,671 children with untreated ADHD, 11,846 who received stimulant treatment, and 47,258 non-ADHD controls. Children who took stimulants for only one to two months, or who had chronic medical conditions requiring long-term medication, were excluded to avoid confounding the results.
Crucially, adult height was evaluated not against population norms but against each individual's expected height, calculated from parental heights using the Tanner-Goldstein-Whitehouse method, a standard approach for estimating genetic height potential via mid-parental height.
When the researchers compared adult heights across the three groups using analysis of variance (ANOVA), they did find statistically significant differences. But statistical significance, particularly in studies with tens of thousands of participants, does not automatically translate into clinical significance. The effect sizes were consistently very small, and the absolute differences were under one centimeter, which is a margin considered clinically negligible.
Their conclusion is measured but clear: after accounting for genetic growth potential, neither an ADHD diagnosis nor stimulant treatment was associated with meaningful reductions in adult height. The findings, they argue, support prioritizing behavioral and functional outcomes when making treatment decisions, since the risk of clinically significant height loss appears to be minimal.
The Take-Away:
For families navigating ADHD treatment, the practical implication is significant: concerns about permanent growth suppression, while understandable, should not be the primary driver of whether or how long a child receives stimulant therapy.
A recent meta-analysis examined how well cognitive behavioral therapy (CBT) improves not just symptoms, but everyday functioning and quality of life in adults with ADHD.
The Background:
ADHD in adults affects far more than attention or impulsivity. It often disrupts key areas of life:
These broad impacts highlight a key issue: reducing symptoms does not automatically translate into better day-to-day functioning.
CBT is a structured, skills-based therapy that helps people:
While both medication (especially stimulants) and CBT improve core ADHD symptoms, CBT is particularly aimed at improving real-world functioning.
The Study:
The researchers analyzed studies involving adults diagnosed with ADHD (or showing clinically significant symptoms). They included:
They focused specifically on outcomes beyond symptoms:
The Results:
1. Strongest Effects: Occupational functioning
CBT showed consistently strong improvements in work-related functioning compared to control groups, both immediately after treatment and at follow-up. This was the most robust finding across domains.
2. Moderate Improvement: Global Functional Impairment
CBT led to moderate improvements in overall daily functioning, with some evidence that gains persist over time. In studies tracking individuals over time, improvements were even stronger at follow-up.
3. Modest Gains: Social Relationships
CBT produced small to moderate improvements in social functioning. Benefits were present both after treatment and at follow-up, but were less pronounced than in work-related outcomes.
4. Limited Effects: Academic Functioning
There were moderate short-term gains when CBT was compared to control groups, but these did not persist at follow-up. Within-subject studies showed only small improvements overall.
5. Modest and Inconsistent Effects: Quality of Life
Improvements in quality of life were small when compared to control groups and often did not last. However, studies tracking individuals over time showed moderate improvements, suggesting some benefit that may not always show up clearly in between-group comparisons.
Overall, the findings suggest:
One notable nuance: CBT did not always outperform other active treatments (like medication or other therapies). This suggests that while CBT is effective, its benefits may partly overlap with broader therapeutic or support effects rather than relying on a single, unique mechanism.
The Take-Away:
CBT is a valuable, evidence-based treatment for adults with ADHD, especially for improving work functioning and overall daily life management. However, its impact on relationships, academic outcomes, and quality of life is more limited and less consistent, pointing to the need for more targeted or combined approaches in those areas.
_logo%201.svg)
