Synthetic patient and interview transcript creator: an essential tool for LLMs in mental health
Warner, A., LeDue, J., Cao, Y., Tham, J., & Murphy, T. H. (2025). Synthetic patient and interview transcript creator: an essential tool for LLMs in mental health. Frontiers in Digital Health, 7. doi:10.3389/fdgth.2025.1625444
Developing high-quality training data is essential for tailoring large language models (LLMs) to specialized applications like mental health. To address privacy and legal constraints associated with real patient data, we designed a synthetic patient and interview generation framework that can be tailored to regional patient demographics. This system employs two locally run instances of Llama 3.3:70B: one as the interviewer and the other as the patient. These models produce contextually rich interview transcripts, structured by a customizable question bank, with lexical diversity similar to normal human conversation. Demographic distributions of generated patient profiles were not significantly different from real-world population data and exhibited expected variability. Additionally, for the patient profiles we assessed LLM metrics and found an average Distinct-1 score of 0.8 (max = 1) indicating diverse word usage. By integrating detailed patient generation with dynamic interviewing, the framework produces synthetic datasets that may aid the adoption and deployment of LLMs in mental health settings.
Brain-age prediction: Systematic evaluation of site effects, and sample age range and size
Yu Y, Cui HQ, Haas SS, et al. Brain-age prediction: Systematic evaluation of site effects, and sample age range and size. Hum Brain Mapp. 2024;45(10):e26768.
doi:10.1002/hbm.26768
Structural neuroimaging data have been used to compute an estimate of the biological age of the brain (brain-age) which has been associated with other biologically and behaviorally meaningful measures of brain development and aging. The ongoing research interest in brain-age has highlighted the need for robust and publicly available brain-age models pre-trained on data from large samples of healthy individuals. To address this need we have previously released a developmental brain-age model. Here we expand this work to develop, empirically validate, and disseminate a pre-trained brain-age model to cover most of the human lifespan. To achieve this, we selected the best-performing model after systematically examining the impact of seven site harmonization strategies, age range, and sample size on brain-age prediction in a discovery sample of brain morphometric measures from 35,683 healthy individuals (age range: 5-90 years; 53.59% female). The pre-trained models were tested for cross-dataset generalizability in an independent sample comprising 2101 healthy individuals (age range: 8-80 years; 55.35% female) and for longitudinal consistency in a further sample comprising 377 healthy individuals (age range: 9-25 years; 49.87% female). This empirical examination yielded the following findings: (1) the accuracy of age prediction from morphometry data was higher when no site harmonization was applied; (2) dividing the discovery sample into two age-bins (5-40 and 40-90 years) provided a better balance between model accuracy and explained age variance than other alternatives; (3) model accuracy for brain-age prediction plateaued at a sample size exceeding 1600 participants. These findings have been incorporated into CentileBrain (https://centilebrain.org/#/brainAGE2), an open-science, web-based platform for individualized neuroimaging metrics.
Normative modelling of brain morphometry across the lifespan with CentileBrain: algorithm benchmarking and model optimisation
Ge R, Yu Y, Qi YX, et al. Normative modelling of brain morphometry across the lifespan with CentileBrain: algorithm benchmarking and model
optimisation. Lancet Digit Health. 2024;6(3):e211-e221. doi:10.1016/S2589-7500(23)00250-9
The value of normative models in research and clinical practice relies on their robustness and a systematic comparison of different modelling algorithms and parameters; however, this has not been done to date. We aimed to identify the optimal approach for normative modelling of brain morphometric data through systematic empirical benchmarking, by quantifying the accuracy of different algorithms and identifying parameters that optimised model performance. We developed this framework with regional morphometric data from 37 407 healthy individuals (53% female and 47% male; aged 3-90 years) from 87 datasets from Europe, Australia, the USA, South Africa, and east Asia following a comparative evaluation of eight algorithms and multiple covariate combinations pertaining to image acquisition and quality, parcellation software versions, global neuroimaging measures, and longitudinal stability. The multivariate fractional polynomial regression (MFPR) emerged as the preferred algorithm, optimised with non-linear polynomials for age and linear effects of global measures as covariates. The MFPR models showed excellent accuracy across the lifespan and within distinct age-bins and longitudinal stability over a 2-year period. The performance of all MFPR models plateaued at sample sizes exceeding 3000 study participants. This model can inform about the biological and behavioural implications of deviations from typical age-related neuroanatomical changes and support future study designs. The model and scripts described here are freely available through CentileBrain.
Normative Modeling of Brain Morphometry in Clinical High Risk for Psychosis
ENIGMA Clinical High Risk for Psychosis Working Group, Haas SS, Ge R, et al. Normative Modeling of Brain Morphometry in Clinical High Risk for Psychosis. JAMA Psychiatry. 2024;81(1):77-88. doi:10.1001/jamapsychiatry.2023.3850
Importance: The lack of robust neuroanatomical markers of psychosis risk has been traditionally attributed to heterogeneity. A complementary hypothesis is that variation in neuroanatomical measures in individuals at psychosis risk may be nested within the range observed in healthy individuals.
Objective: To quantify deviations from the normative range of neuroanatomical variation in individuals at clinical high risk for psychosis (CHR-P) and evaluate their overlap with healthy variation and their association with positive symptoms, cognition, and conversion to a psychotic disorder.
Design, setting, and participants: This case-control study used clinical-, IQ-, and neuroimaging software (FreeSurfer)-derived regional measures of cortical thickness (CT), cortical surface area (SA), and subcortical volume (SV) from 1340 individuals with CHR-P and 1237 healthy individuals pooled from 29 international sites participating in the Enhancing Neuroimaging Genetics Through Meta-analysis (ENIGMA) Clinical High Risk for Psychosis Working Group. Healthy individuals and individuals with CHR-P were matched on age and sex within each recruitment site. Data were analyzed between September 1, 2021, and November 30, 2022.
Main outcomes and measures: For each regional morphometric measure, deviation scores were computed as z scores indexing the degree of deviation from their normative means from a healthy reference population. Average deviation scores (ADS) were also calculated for regional CT, SA, and SV measures and globally across all measures. Regression analyses quantified the association of deviation scores with clinical severity and cognition, and 2-proportion z tests identified case-control differences in the proportion of individuals with infranormal (z < -1.96) or supranormal (z > 1.96) scores.
Results: Among 1340 individuals with CHR-P, 709 (52.91%) were male, and the mean (SD) age was 20.75 (4.74) years. Among 1237 healthy individuals, 684 (55.30%) were male, and the mean (SD) age was 22.32 (4.95) years. Individuals with CHR-P and healthy individuals overlapped in the distributions of the observed values, regional z scores, and all ADS values. For any given region, the proportion of individuals with CHR-P who had infranormal or supranormal values was low (up to 153 individuals [<11.42%]) and similar to that of healthy individuals (<115 individuals [<9.30%]). Individuals with CHR-P who converted to a psychotic disorder had a higher percentage of infranormal values in temporal regions compared with those who did not convert (7.01% vs 1.38%) and healthy individuals (5.10% vs 0.89%). In the CHR-P group, only the ADS SA was associated with positive symptoms (β = -0.08; 95% CI, -0.13 to -0.02; P = .02 for false discovery rate) and IQ (β = 0.09; 95% CI, 0.02-0.15; P = .02 for false discovery rate).
Conclusions and relevance: In this case-control study, findings suggest that macroscale neuromorphometric measures may not provide an adequate explanation of psychosis risk.
A three-dimensional virtual mouse generates synthetic training data for behavioral analysis
Bolaños, L.A., Xiao, D., Ford, N.L. et al. A three-dimensional virtual mouse generates synthetic training data for behavioral analysis. Nat Methods 18, 378–381 (2021). https://doi.org/10.1038/s41592-021-01103-9
We developed a three-dimensional (3D) synthetic animated mouse based on computed tomography scans that is actuated using animation and semirandom, joint-constrained movements to generate synthetic behavioral data with ground-truth label locations. Image-domain translation produced realistic synthetic videos used to train two-dimensional (2D) and 3D pose estimation models with accuracy similar to typical manual training datasets. The outputs from the 3D model-based pose estimation yielded better definition of behavioral clusters than 2D videos and may facilitate automated ethological classification. The model can be a time savings as annotation markers used for training pose-estimation tools such as DeepLabCut can be built into the model and transferred as coordinates without a need for hand labelling. See potential applications for autism research published in Spectrum News.
Towards a Visualizable, De-identified Synthetic Biomarker of Human Movement Disorders
J Parkinsons Dis. 2022 Aug 27;1(-1):2085-2096. doi: 10.3233/JPD-223351.
A number of opportunities are made available by advancements in computer vision that will enable digitization of human form, movements, and will represent them synthetically in 3D. Representing human movements within synthetic body models will potentially pave the way towards objective standardized digital movement disorder diagnosis and building sharable open-source datasets from such processed videos. Such digital movement capturing methods will be important for both machine learning-based diagnosis and computer vision-aided clinical assessment. It would also supplement face-to-face clinical visits and be used for longitudinal monitoring and remote diagnosis. Towards a Visualizable, De-identified Synthetic Biomarker of Human Movement Disorders..