⏫ AI Detects ADHD in Adults From Medical Records

Researchers in Sweden have developed an AI tool that can identify signs of attention deficit hyperactivity disorder in adults by analyzing their medical records.

The model was trained on the medical histories of nearly 8,000 patients, including people with confirmed ADHD and patients with other psychiatric conditions.

🔍 The system looks for subtle patterns in a person's medical history. For example, it may flag birth complications or substance use-related diagnoses—the latter can sometimes reflect attempts to cope with undiagnosed ADHD symptoms.

The tool performed best when predicting ADHD 6 months before a formal diagnosis. In a test group of 800 people, split evenly between patients with and without ADHD, the model correctly identified 80% of ADHD cases and correctly ruled it out in 77% of non-ADHD cases.

✅ There are still important limitations. The model struggles to distinguish ADHD from depression and anxiety disorders because their symptoms can overlap. This means it would need further refinement before being used in real clinical settings, where patients often have multiple overlapping diagnoses.

Can AI be trusted in psychiatry?

❤️ — Yes, algorithms can help
🔥 — No, only human doctors

@hiaimediaen