Software Developer in Chicago, United States

I created what is now the state-of-the-art deep learning architecture for analyzing CT scans. Computer-implemented methods and apparatuses for anomaly detection in volumetric images are provided. A two-dimensional convolutional neural network (CNN) is used to encode slices within a volumetric image, such as a CT scan. The CNN may be trained using an output layer that is subsequently omitted during the use of the CNN as an encoder. The CNN encoder output is applied to a recurrent neural network (RNN), such as a long short-term memory network. The RNN may output various indications of the presence, probability, and/or location of anomalies within the volumetric image.

I developed an artificial intelligence software to automatically detect lung nodules which are often missed by radiologists and can portend cancer in CT scans. The link attached leads to a video demonstrating the AI results and the integration into the radiology workflow.

I deployed and used a Spark cluster to predict the genre of songs in The Echo Nest’s Million Song Dataset using data on volume, tempo, pitch, and “danceability”. I also wrote the code to train models using Spark’s MLlib.

I was the primary investigator and first author of an international study on using AI to detect COVID-19, published in a peer-reviewed medical journal.

Abstract
Purpose: In the context of the COVID-19 pandemic, rapid triage of cases and exclusion of other pathologies with artificial intelligence (AI) can assist over-stretched radiology departments.
We aim to validate three open-source AI models on an external test set.

Approach:
We tested three open-source deep learning models, COVID-Net, COVIDNet-S-GEO, and CheXNet, for their ability to detect COVID-19 pneumonia and to determine its severity using 129 chest x-rays from two different vendors. Results: All three models detected COVID-19 pneumonia. Only the COVID-19 Net-S-GEO and CheXNet models performed well on severity scoring; COVID-Net only performed well at either task on images taken with a Philips machine (AUC 0.735) and not an Agfa machine (AUC 0.598).

Conclusions:
Chest x-ray triage using existing machine learning models for COVID-19 pneumonia can be successfully implemented using open-source AI models. Evaluation of the model using local x-ray machines and protocols is highly recommended before implementation to avoid vendor or protocol-dependent bias.

Abstract: Securities and Exchange Commission (SEC) regulates US financial markets. One requirement for securities market participants is to provide disclosure to the public. SEC’s EDGAR (Electronic Data Gathering,
Analysis and Retrieval) database aggregates and disseminates the public disclosure data. There are more than 100 types of forms that market participants fill out and electronically file with EDGAR. The project is focused on a very important filing type, the 10­K. Publicly traded companies disclose comprehensive information about the company operations regularly. This project demonstrates the data collection, manipulation, and analysis (sentiment based on NLP) of the 10­K filings leveraging the Hadoop data processing framework for rapid data analysis.