Intae Moon (@IntaeMoon) profile picture

Postdoc fellow at @HarvardDBMI | PhD at @MITEECS | @DanaFarber | Ex-Student researcher at @GoogleHealth | Research on machine learning and biomedicine

Cambridge, MA
Joined March 2020
I'm excited to share our panel sequencing-based cancer type prediction and visualization tool (itmoon7.github.io/onconpc/#/…), based on our recent @NatureMedicine publication (rdcu.be/dHVaq)! Learn more by checking out our short tutorial.
1
3
2
17
7,831
Using routinely collected EHR and NGS panel sequencing data at @DanaFarber, we provide unique insights into the genetics and prognostic characteristics of CUP tumor samples and demonstrate the potential for algorithm-guided treatment selection. Please see the manuscript for more!
201
SurvLatent ODE continued to outperform the Khorana score when focusing on the highest risk cancer types, pancreatic and stomach cancer. Within this cohort, SurvLatent ODE identified a low risk subpopulation comparable to the population baseline (shown as a dotted line).
1
As a clinical use case, we predicted time to Deep Vein Thrombosis (DVT), a life-threatening complication for patients with cancer. SurvLatent ODE outperforms the current clinical standard Khorana Risk scores, while providing clinically meaningful and interpretable latent states.
1