Sparkle Russell-PuleriCheXplain: A minimum viable healthcare product prototype that explains how a multi-class deep…by Sparkle Russell-Puleri and Ravi S. Rāmphal6 min read·May 10, 2022----
Sparkle Russell-PuleriinTowards Data SciencePredicting 30-day Readmission risk for ICU patients using a multi-task deep learning frameworkUnsupervised clustering of patient sub-populations for multitask learning using electronic health records13 min read·May 21, 2020----
Sparkle Russell-PuleriinTowards Data SciencePredicting future medical diagnoses with RNNs using Fast AI API from scratchFull pytorch implementation Doctor AI paper using Electronic Health Records8 min read·Apr 21, 2019--2--2
Sparkle Russell-PuleriinTowards Data ScienceUsing Electronic Health Records to predict future diagnosis codes with Gated Recurrent UnitsBackground: Detailed review of Doctor AI: Predicting Clinical Events via Recurrent Neural Nets (Choi et.al 2016)8 min read·Mar 12, 2019--1--1
Sparkle Russell-PuleriGated Recurrent Units explained with matrices: Part 2 Training and Loss Functionby: Sparkle Russell-Puleri and Dorian Puleri4 min read·Mar 6, 2019--1--1
Sparkle Russell-PuleriinTowards Data ScienceGated Recurrent Units explained using Matrices: Part 1by: Sparkle Russell-Puleri and Dorian Puleri11 min read·Feb 24, 2019--11--11
