📄papers

i've published two peer-reviewed papers so far. the short version of each, for anyone who'd rather skim than download the pdf.

DeepCOVNet Model for COVID-19 Detection Using Chest X-Ray Images

Wireless Personal Communications, Springer · April 10, 2023

a deep-learning model for detecting covid-19 from chest x-rays — not to replace a radiologist, but to triage fast when radiologists are overwhelmed. the paper covers the architecture, the augmentations that actually moved the needle, and where the model fails (it does fail — any honest covid-classification paper owes you that section).

Evaluating the Performance of Machine Learning Models for Diabetes Prediction with Feature Selection and Missing Values Handling

IJMIT · June 26, 2023

a comparison study of several ml models for diabetes prediction, with a deliberate focus on the parts most papers skip: how missing values are filled, and which feature-selection strategy you use. the punchline is that careful preprocessing moves performance more than the choice of "fancy" model does. the applied version lives as diabetes prediction.

both papers are googlable by title. if you'd like me to walk you through either, reach out on twitter.