Skip to main content
Search

University of Zurich

9 related articles

Subscribe to updates
  1. Nuclear-mass prediction steps out of the black box: an interpretable AI rediscovers forgotten Jaffe factorization and pushes WS4 error below 0.12 MeV

    Nuclear physics has long relied on phenomenological models to predict nuclear binding energies, while high-precision AI models are hard to trust when extrapolating. An interpretable neural network taking only proton and neutron numbers as input spontaneously forms a double-helix geometry in its internal representations, reproduces macroscopic terms such as the liquid-drop model in a principal-component decomposition, and rediscovers the local Jaffe factorization proposed in 1969 and since overlooked. Applied as a local correction to the WS4 model, this factorization brings the root-mean-square binding-energy error below 0.12 MeV without relying on a black box.

  2. medRxiv

    Therapy sessions can now run in silico: raising therapist empathy prompts lifts every motivational-interviewing integrity score

    Psychotherapy research is constrained by costly trials and clinical features that are hard to manipulate at random. A preregistered proof-of-concept study paired therapist and patient agents built on a locally deployed Llama-3.3-70B-Instruct to generate 512 automated motivational interviewing sessions, manipulating only the therapist prompts for empathy, affirmation and alliance building. Among 480 sessions retained for analysis, all four motivational interviewing treatment integrity global ratings were higher under high common factors, with adjusted mean differences of 0.54 to 0.95 points.

  3. bioRxiv

    Olive repeats get a genome-wide read: three plant DNA models agree on which STR loci shift their embeddings

    Olive short tandem repeats (STRs, stretches whose repeat-unit copy number varies) had been read mainly through small microsatellite panels, leaving genome-wide variation and interpretation open. Across 212 Iranian and Mediterranean olive accessions, the study catalogued 117,119 polymorphic STR loci and passed natural allele series through three plant genomic language models, AgroNT, PlantCAD2 and GPN, measuring how far embeddings move as repeat length departs from the reference allele. On 62,847 loci, the three models shared 7,417 embedding-sensitive loci, and at most tested loci STR responses exceeded length-matched nonrepetitive indels.

  4. Epicardial and pericardial fat are distinguished as two functionally distinct cardiac depots, with deep-learning segmentation and standardization advancing their imaging quantification toward cardiovascular risk stratification

    This review systematically examines the anatomy, physiology, and pathophysiological roles of epicardial adipose tissue (EAT) and pericardial adipose tissue (PAT), compares CT, MRI, and echocardiography for quantifying these depots, notes that deep-learning segmentation (with Dice coefficients reported from 0.82 to 0.98) improves accuracy and reproducibility, and summarizes conflicting imaging evidence on EAT's link to cardiovascular outcomes alongside findings that statins, GLP-1 receptor agonists, SGLT2 inhibitors, and lifestyle interventions can reduce EAT volume.

  5. bioRxiv

    After a 2011 marine heatwave wiped out seagrass, Shark Bay bottlenose dolphin adult survival fell from 0.99 to 0.85 and the two gulf populations declined by 39% and 36%

    Using 20 years (2003-2023) of dolphin photo-identification data, the study fitted a Bayesian Hidden Markov Model to estimate annual, age-class-specific survival probabilities, abundance and recruitment rate for Indo-Pacific bottlenose dolphins (Tursiops aduncus) in the western (WSB) and eastern (ESB) gulfs of Shark Bay, and related them to seagrass loss following the 2011 marine heatwave, finding adult survival fell from 0.99 (0.98-1.00) pre-MHW to 0.85 (0.79-0.90) post-MHW in WSB and from 0.96 (0.92-0.98) to 0.89 (0.85-0.

  6. SpaCEy links tissue spatial patterns to clinical outcomes with an explainable graph neural network, improving prediction and surfacing spatial markers in lung and breast cancer cohorts

    The authors present SpaCEy (Spatial Clinical Explainability), an explainable graph neural network that builds each tissue sample into a spatial graph via Delaunay triangulation, with nodes as single-cell protein-marker abundances and no predefined cell-type or anatomical-region inputs, learns predictive embeddings with a GNN, and uses a GNNExplainer-style explainer to output edge masks that are aggregated over k-hop neighbourhoods into node importance, thereby localising contiguous outcome-associated spatial regions and key proteins; it predicted progression in a 416-patient lung adenocarcinoma cohort (accuracy 0.68, F1 0.68, AUC 0.62, versus Ali et al. 0.61/0.59/0.57 and SPACE-GM 0.55/0.52/0.

  7. Stroke

    Responsible AI for Acute Stroke Management: A Review of Explainability and Fairness

    This review synthesizes the current literature on explainable and fair artificial intelligence in acute stroke management, noting that AI already performs strongly on tasks such as large vessel occlusion detection, Alberta Stroke Program Early Computed Tomography Score scoring, and functional outcome prediction, while post hoc explainability methods remain approximate and rarely formally tested, fairness evaluation remains uncommon due to limited demographic metadata, regulatory constraints, and the absence of stroke-specific fairness criteria, explainability and fairness remain largely disconnected, and generalizability is affected by dataset partitioning and reporting practices, leading to a call for unified evaluation frameworks that jointly assess explainability, fairness, and generali

  8. Pain Management

    Determination of Candidate Predictors for Chiropractic Treatment Outcome of Spinal Pain Using a Machine Learning Framework for Small Datasets

    In this prospective cohort study, 96 patients with spinal pain completed an extensive questionnaire covering pain, mood, sleep, lifestyle, and treatment expectations at baseline and at 1 and 3 months, with binary recovery outcomes defined at 3 months for pain intensity, disability, quality of life, and Patient Global Impression of Change; a three-step machine learning framework combining SHAP-based candidate feature selection, Leave-One-Out Cross-Validation with permutation testing, and generalization testing yielded AUC values of 0.93-0.99 in LOOCV and 0.62-0.

  9. bioRxiv

    Cortico-subcortical multi-head self-attention as a substrate for cognitive performance

    The work proposes that cortico-thalamic circuits are well suited to implement multi-head self- and cross-attention: layer 2/3 pyramidal cells maintain a recurrent key-value memory while layer 5 pyramidal cells decode the memory retrieved by an incoming query, the computation of keys, values and queries maps onto core and matrix thalamo-cortical projections distributed across the micro- and macro-columns of a cortical area, one cortical area forms an attention head and cortex a multi-head self-attention network, the same thalamo-cortical microcircuit also calculates sensory prediction errors guiding gradient-based synaptic plasticity, and a reward-prediction error gates via basal ganglia the cortical output and the re-activation of hippocampal memories, with the trained network aligning wit

You’re all caught up