Length-scale study in deep learning prediction for non-small cell lung cancer brain metastasis
Abstract
Deep learning-assisted digital pathology has demonstrated the potential to profoundly impact clinical practice, even surpassing human pathologists in performance. However, as deep neural network (DNN) architectures grow in size and complexity, their explainability decreases, posing challenges in interpreting pathology features for broader clinical insights into physiological diseases. To better assess the interpretability of digital microscopic images and guide future microscopic system design, we developed a novel method to study the predictive feature length-scale that underpins a DNN’s predictive power. We applied this method to analyze a DNN’s capability in predicting brain metastasis from early-stage non-small-cell lung cancer biopsy slides. This study quantifies DNN’s attention for brain metastasis prediction, targeting features at both the cellular scale and tissue scale in H&E-stained histological whole slide images. At the cellular scale, the predictive power of DNNs progressively increases with higher resolution and significantly decreases when the resolvable feature length exceeds 5 microns. Additionally, DNN uses more macro-scale features associated with tissue architecture and is optimized when assessing visual fields greater than 41 microns. Our study computes the length-scale requirements for optimal DNN learning on digital whole-slide microscopic images, holding the promise to guide future optical microscope designs in pathology applications and facilitating downstream deep learning analysis.
Copyright and License
This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material.
Funding
This study was supported by U01CA233363 from the National Cancer Institute (RJC) and by the Washington University in St. Louis School of Medicine Personalized Medicine Initiative (RJC). HZ, SL, SM, and CY are also supported by Sensing to Intelligence (S2I) (Grant No. 13520296) and Heritage Research Institute for the Advancement of Medicine and Science at Caltech (Grant No. HMRI-15-09-01).
Data Availability
The data is available at CaltechData https://doi.org/10.22002/dw66e-mbs82.
Code Availability
The code is available at Github https://github.com/hwzhou2020/NSCLC_ResNet and https://github.com/hwzhou2020/NSCLC_length_scale.
Files
Name | Size | Download all |
---|---|---|
md5:fde81c17ad03f2d7ce6a94e174d39eaf
|
3.2 MB | Preview Download |
md5:43e2cae5585447e299eadcf6dddccc76
|
1.6 MB | Preview Download |
Additional details
- National Cancer Institute
- U01CA233363
- Washington University in St. Louis School of Medicine Personalized Medicine Initiative
- Sensing to Intelligence
- 13520296
- Heritage Research Institute for the Advancement of Medicine and Science at Caltech
- HMRI-15-09-01
- Accepted
-
2024-09-17Accepted
- Available
-
2024-09-27Published online
- Publication Status
- Published