Projects

Last updated on Jul 28, 2022

An app-based solution to polypharmacy management using data integration and patient empowerment. A winner in MedHacks 2021, this project proposed initial product design and proof of concept for an app aimed to mitigate overprescribing and aid in medication management.

Last updated on Jul 28, 2022

MindPsych

A digital mental health solution for psychiatric diagnostics and treatment through data integration. MindPsych provides a patient data integration app as well as a provider portal allowing for optimal patient-provider interactions.

Last updated on Feb 8, 2022

Sentence Generation and Classification with Variational Autoencoder and BERT

A conditional variational autoencoder uses a sentence premise and lass label to generate a hypothesis specific to that class type. Then, a BERT model is used to classify the results in order to evaluate the style transfer to generating the sentences.

Last updated on Feb 8, 2022

Image Captioning using an LSTM Network

Created a deep learning network for automatic image captioning of the COCO dataset using a ResNet 50 CNN encoder and a LSTM RNN decoder.

Last updated on Jan 21, 2022

Convolutional Networks for Semantic Segmentation

Built various deep CNN architectures and trained them to predict the segmentation masks of the images in the India Driving Dataset as a semantic segmentation task.

Last updated on Jan 21, 2022

Multi-Layer Neural Network Classification of Fashion MNIST Dataset

Implemented a multi-layer neural network from scratch to classify images of fashion products using the Zalando’s Fashion MNIST Dataset in order to identify strategies that optimize performance.

Last updated on Jan 21, 2022

Logistic and Softmax Regression to Predict Vehicle Types from Images

Trained logistic and softmax regression models built from scratch to predict vehicle types from 200x300 pixel images from the Comprehensive Cars dataset.

Last updated on Feb 8, 2022

Steam Video Game Review Recommender System

Given video game reviews from Steam, two prediction tasks were completed. Given a user’s history, one predicted whether a user would play a new game and the other predicted how long the user would play a game.

Last updated on Jan 31, 2022

California Traffic Records Collision Severity Prediction

Using the California Statewide Integrated Traffic Records System dataset from 2001-2020, we look at various features and models to classify between collision severity.

Last updated on Jan 24, 2022

Classifying Evoked Potentials from Spinal Cord Microelectrode Array Data

Applied PCA and LDA classification on spinal cord microelectrode array data to classify stimulus and non-stimulus conditions, as well as location and amplitude of stimulus.

Last updated on Jan 24, 2022

Spike-Timing Dependent Plasticity in a Neural Network

Created a network model with synapses that updated according to the Hebbian learning rule (STDP) using Brian2.

Last updated on Feb 8, 2022

Revealing Location-Specific Variation and Drug Transport Specificity in the Allen Brain Atlas

Analyzed 7k gene RNA-seq dataset for a single clinically healthy patientfrom Allen Brain Atlas using ICA, PCA, clustering, and classification to predict brain regions (98.7% accuracy for 3 regions, 67.1% accuracy for 10 regions) to see if variations in gene expression can be used to predict between regions at different resolutions and if that data can be used to determine localization of drug behavior based on chemical similarity to metabolites transported by transporters genetically expressed.