MD candidate · Tbilisi, Georgia
Elene Kapanadze. I combine clinical training with computational neuroscience, aiming at objective, biomarker-based diagnosis in psychiatry.
A live feed of new papers from Swiss institutions on the topics I work on. It updates itself every time the page loads, straight from Europe PMC, so it never goes out of date.
Papers matching the filters (the current year is partial).
Most frequent last authors, usually the group leaders, among the latest 100 matching papers.
Computational neuroscience models you can test yourself. Move the sliders and watch the neuron respond.
This model describes how a spike (action potential) arises from sodium and potassium channels opening and closing. Increase the input current and watch the neuron switch from silence to repetitive firing.
Synapses change strength depending on spike timing. If the presynaptic neuron fires just before the postsynaptic one, the connection strengthens; if it fires just after, the connection weakens. Move the timing difference and see the change in synaptic weight.
Coming next: integrate-and-fire networks, EEG signal explorer, and a PhD navigator for Georgian students.
Clinical training, a first publication, and a change of direction toward the brain.
I am a sixth-year medical student at Petre Shotadze Tbilisi Medical Academy in Georgia. My first publication was a microbiology study of antibiotic resistance patterns in Staphylococcus aureus, where I did the laboratory work, the statistics and the data visualization.
That taught me how to run a study from bench to manuscript. My interest has since moved to computational and systems neuroscience, and to neuropsychiatry in particular: how signals from the brain, such as EEG and imaging, can support more objective diagnosis of psychiatric conditions.
The research radar and the lab on this page are built to be useful, not decorative: the radar tracks new work from Swiss labs as it is published, and the lab explains models I study, in English and in Georgian, where good neuroscience material is scarce.
Six years of medical training. I know what the diagnostic problem looks like at the bedside, which is the problem the models are meant to serve.
I designed, ran and published a laboratory study, with statistical analysis in SPSS and data visualization.
Coursework in computational neuroscience, and an independent reanalysis of published ENIGMA-ADHD neuroimaging data, built in Python and published on GitHub Pages.
One peer-reviewed paper, and an independent computational project built on published data.
A public health-oriented study of antibiotic resistance. My contributions: bacterial culture, disk diffusion susceptibility testing, SPSS statistical analysis, data visualization and manuscript drafting.
Reproduced case-control cortical thickness effect sizes across 34 regions from published ENIGMA-ADHD summary statistics (Hoogman et al., 2019). Re-applied Benjamini–Hochberg FDR correction, compared adult versus adolescent cohorts, and built interactive visualizations. Python (pandas, statsmodels, Plotly), built with AI coding assistance.
Open the project page →