About me
I am a radiation oncologist and Clinical Associate Professor in the Department of Radiation Oncology at Stanford University School of Medicine. My email address is mgens@stanford.edu. This is me:
Links
Selected research projects
Nnet-survival: A Scalable Discrete-Time Survival Model using Neural Networks
A survival / time-to-event model that can be used with neural networks. Implemented in Keras and PyTorch deep learning libraries.
I'm gratified that Nnet-survival has been used in many studies/papers, such as:
- Time-related survival prediction in molecular subtypes of breast cancer using time-to-event deep-learning-based models
- Integrating Multiomics Information in Deep Learning Architectures for Joint Actuarial Outcome Prediction in Non-Small Cell Lung Cancer Patients After Radiation Therapy
- Preoperative CT-based Deep Learning Model for Predicting Disease-Free Survival in Patients with Lung Adenocarcinomas
- Distant metastasis time to event analysis with CNNs in independent head and neck cancer cohorts
- Comparison of State-Of-The-Art Neural Network Survival Models With The Pooled Cohort Equations for Cardiovascular Disease Risk Prediction
- Improving decision making in the management of hospital readmissions using modern survival analysis techniques
- Forecasting End Strength in the US Army
Machine learning prognostic model for patients with metastatic cancer
A high-dimensional prognostic model using electronic medical record data that outperforms traditional models.
- Paper describing model creation, published in JNCI
- Paper comparing model performance to that of treating radiation oncologist (model did slightly better!), published in JAMIA
- Randomized study in which we implemented the model in medical oncology clinics aiming to increase advance care planning conversations, published in JNCI in 2024
- Code at GitHub
Other projects
TumorMap-Lung
An open imaging dataset of deidentified chest CT scans from patients with metastatic lung cancer, with 310 lung tumors segmented across 67 scans.
TrialFetch.com
A free web app where you enter free text information for a cancer patient, and the site shows matching clinical trials and highlights some of the most promising ones
Seneca-tutor
A large language model-based tutoring app that runs locally with a web interface using Flask and the Anthropic API, and keeps a memory of a lesson plan and what it has worked on with the student so far.