A new study uses AI on brain scans to predict depression. The findings are modest, but the implications go beyond the hospital.
[read the full story...]Brain scans, depression, and AI: small signals, big questions
No Responses »
Internet-delivered CBT for depression: real-world evidence shows similar benefits to face-to-face therapy
This Finnish study of 5,834 healthcare records found therapist-guided internet CBT showed similar depression improvements to face-to-face therapy, providing real-world evidence beyond selective RCT populations.
[read the full story...]Inside the diagnostic grey zone: using machine learning to separate bipolar and major depression
High misdiagnosis rates between bipolar and major depressive disorder cause real harm to patients and services. This new neuroimaging study tested whether brain connectivity and machine learning could do a better job of telling the two apart, with interesting but limited results.
[read the full story...]Spotting the storm before it breaks: mapping the prodrome of severe mental illness
People with severe mental illness often face years of poor health before diagnosis. A new study uses machine learning and clinical notes to map the early warning networks of symptoms that could help us intervene earlier.
[read the full story...]Predicting antidepressant response using artificial intelligence
Holly Fraser discusses new findings on whether and how we can predict antidepressant response using artificial intelligence.
[read the full story...]Genetic risk for schizophrenia is associated with changes in heart structure and function
Nadine Parker and Ole Andreassen summarise a recent UK population-based cohort study, which looks at the impact of polygenic risk for schizophrenia on cardiac structure and function in over 32,000 people.
[read the full story...]Stratified care versus stepped care for depression: which is more effective?
Sarah Watts reviews a cluster randomised clinical trial investigating the effectiveness of stratified care compared to stepped care for depression, which has implications for IAPT services.
[read the full story...]Mental health stigma and online social support for bipolar disorder: what can we learn from Twitter?
Charlotte Walker explores an online ethnography study that explores how Twitter users discuss mental illness, particularly bipolar disorder, and in what context; focusing specifically on the areas of stigma and social support.
[read the full story...]Predicting suicide attempts in adolescents: machine learning is powerful, but don’t forget Bayes’ rule
Derek de Beurs explores a recent study that uses longitudinal clinical data and machine learning to predict suicide attempts in adolescents.
[read the full story...]Can a machine learning approach help us predict what specific treatments work best for individuals with depression?
Marcus Munafo explores a recent study that uses a machine learning approach across two trials (STARD*D and CO-MED) to try and predict treatment outcomes (primarily focusing on the antidepressant citalopram) for depression.
[read the full story...]