Two medical applications of deep neural networks
Abstract/Contents
- Abstract
- Many life threatening conditions can be cured if treated early enough. Further, many life threatening conditions are obvious to a trained medical expert. Recent advances in deep learning can be leveraged to distill some of this highly valuable medical expertise. A novel large scale dataset of labeled skin lesion images is collected from the internet. An ImageNet pretrained model is fine-tuned on the dataset to achieve board-certified dermatologist level performance at detecting skin cancer. Separately, a neural net is trained to detect ICD codes from biological waveforms such as electrocardiogram and photoplethysmogram. With this net, users can be reliably placed into lower and higher risk groups. Users can be triaged with an unreliable test to improve the effective sensitivity of a limited reliable test. A semantic embedding of ICD codes self-organized by organ (heart, liver, brain, etc.) is demonstrated. The neural net learned this semantic embedding of ICD codes as a byproduct of being trained to detect ICD codes from raw waveforms
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2021; ©2021 |
| Publication date | 2021; 2021 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Kuprel, Brett Ryan |
|---|---|
| Degree supervisor | Thrun, Sebastian, 1967- |
| Thesis advisor | Thrun, Sebastian, 1967- |
| Thesis advisor | Pauly, John (John M.) |
| Thesis advisor | Girod, Bernd |
| Degree committee member | Pauly, John (John M.) |
| Degree committee member | Girod, Bernd |
| Associated with | Stanford University, Department of Electrical Engineering |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Brett Kuprel |
|---|---|
| Note | Submitted to the Department of Electrical Engineering |
| Thesis | Thesis Ph.D. Stanford University 2021 |
| Location | https://purl.stanford.edu/gb534fn0007 |
Access conditions
- Copyright
- © 2021 by Brett Ryan Kuprel
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...