Skip to content
View ArunAryal10's full-sized avatar
0 followers · 2 following

Highlights

  • Pro

Block or report ArunAryal10

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ArunAryal10/README.md

Arun Aryal

PhD candidate in Computational, Cognitive & Network Neuroscience at the Cole Neurocognition Lab, Rutgers University–Newark. I build models of how brain networks give rise to cognition and behavior, and apply that same modeling toolkit to health and medical data.

Research: functional connectivity, activity-flow modeling, neural decoding, task-fMRI Interests: Health-AI, computational neuroscience, medical imaging, machine learning for biology Currently: seeking a Summer 2027 internship in the Health-AI / neuroscience space

LinkedIn

Selected work

Project What it is
Activity-Flow Decoding + MaxT Permutation Testing Task-fMRI: does functional connectivity + activation structure support above-chance somatomotor decoding? (36 subjects, HPC/SLURM)
Brain-Age CNN from MRI Convolutional net regressing brain age from grey-matter MRI (Pearson R 0.90, MAE 2.5 yrs)
Breast-Cancer Subtyping from Gene Expression ML + deep learning classifying molecular subtypes from 54k-feature expression data
RNN Working Memory (Delayed Match-to-Sample) From-scratch PyTorch RNN that holds a stimulus across a delay — modeling prefrontal working memory
CNN Image Classification Compact PyTorch CNN for image classification (Animal Faces)
MannChill Full-stack (FastAPI + React) Health-AI app modeling allostatic load, with wearable integration
Hodgkin–Huxley · Integrate-and-Fire Biophysical neuron models built from scratch in MATLAB

Toolbox

Python PyTorch TensorFlow scikit-learn MATLAB NumPy pandas

Neuroimaging (fMRI, nibabel, SPM12) · functional connectivity · activity-flow modeling · deep learning (PyTorch/Keras) · statistical inference & permutation testing · HPC/SLURM

Contact: LinkedIn

Pinned Loading

  1. Hackathon2026 Hackathon2026 Public

    MannChill — a full-stack (FastAPI + React) Health-AI app measuring allostatic load for the Nepali diaspora, with WHOOP and voice-journal integration.

    JavaScript

  2. RNN-Delayed-Match-to-Sample-PyTorch RNN-Delayed-Match-to-Sample-PyTorch Public

    From-scratch PyTorch RNN trained on a delayed match-to-sample working-memory task, with delay-period decoding of the maintained stimulus.

    Python

  3. CNN-Image-Classification-PyTorch CNN-Image-Classification-PyTorch Public

    Compact PyTorch CNN for image classification on the Animal Faces (AFHQ) dataset, with a reproducible training pipeline.

    Python

  4. CNN-Architecture-for-MRI-based-brain-age-prediction CNN-Architecture-for-MRI-based-brain-age-prediction Public

    A CNN that predicts brain age from SPM12-preprocessed grey-matter MRI (Pearson R 0.90, MAE 2.5 years).

    Jupyter Notebook

  5. Task-Functional-connectivity-analysis-in-one-shot-learning Task-Functional-connectivity-analysis-in-one-shot-learning Public

    Task-fMRI functional-connectivity and activity-flow decoding of a one-shot learning task, with MaxT permutation testing for family-wise-error control (36 subjects, HPC/SLURM).

    Jupyter Notebook

  6. ML-DL-models-to-predict-breast-cancer-using-gene-expression-data ML-DL-models-to-predict-breast-cancer-using-gene-expression-data Public

    Classifying breast-cancer molecular subtypes from 54k-feature gene-expression data with scikit-learn and a Keras deep net (~92% test accuracy).

    Jupyter Notebook 1