J Pathol Transl Med.  2021 Mar;55(2):118-124. 10.4132/jptm.2020.12.22.

Deep learning for computer-assisted diagnosis of hereditary diffuse gastric cancer

• Rasmussen SA
• Arnason T
• Huang WY

• KMID: 2513897
• DOI: http://doi.org/10.4132/jptm.2020.12.22

Abstract

Background
Patients with hereditary diffuse gastric cancer often undergo prophylactic gastrectomy to minimize cancer risk. Because intramucosal poorly cohesive carcinomas in this setting are typically not grossly visible, many pathologists assess the entire gastrectomy specimen microscopically. With 150 or more slides per case, this is a major time burden for pathologists. This study utilizes deep learning methods to analyze digitized slides and detect regions of carcinoma.
Methods
Prophylactic gastrectomy specimens from seven patients with germline CDH1 mutations were analyzed (five for training/validation and two for testing, with a total of 133 tumor foci). All hematoxylin and eosin slides containing cancer foci were digitally scanned, and patches of size 256×256 pixels were randomly extracted from regions of cancer as well as from regions of normal background tissue, resulting in 15,851 images for training/validation and 970 images for testing. A model with DenseNet-169 architecture was trained for 150 epochs, then evaluated on images from the test set. External validation was conducted on 814 images scanned at an outside institution.
Results
On individual patches, the trained model achieved a receiver operating characteristic (ROC) area under the curve (AUC) of 0.9986. This enabled it to maintain a sensitivity of 90% with a false-positive rate of less than 0.1%. On the external validation dataset, the model achieved a similar ROC AUC of 0.9984. On whole slide images, the network detected 100% of tumor foci and correctly eliminated an average of 99.9% of the non-cancer slide area from consideration.
Conclusions
Overall, our model shows encouraging progress towards computer-assisted diagnosis of hereditary diffuse gastric cancer.

Keyword

Machine learning; Pathology; Computer-assisted diagnosis; Stomach neoplasms; Deep learning

Figure

• 

  Fig. 1 A representative example of a manually annotated tumor region.

• 

  Fig. 2 Distribution of image patches into training, validation, and test data.

• 

  Fig. 3 Receiver operating characteristic curve for classification of individual patches from test data. The area under the curve is 0.9986.

• 

  Fig. 4 Examples of 256 × 256 pixel patches correctly classified as cancer (A–D) or normal (E–H) by the trained model.

• 

  Fig. 5 A portion of a whole slide image analyzed by the trained model. Panels on the right show close ups of correctly identified tumors. Panels on the left show false-positive patches.

Reference

References

1. van der Post RS, Vogelaar IP, Carneiro F, et al. Hereditary diffuse gastric cancer: updated clinical guidelines with an emphasis on germline CDH1 mutation carriers. J Med Genet. 2015; 52:361–74.

2. Rajpurkar P, Irvin J, Zhu K, et al. CheXnet: radiologist-level pneumonia detection on chest X-rays with deep learning. Preprint at: https://arxiv.org/abs/1711.05225 . 2017.

3. Gulshan V, Peng L, Coram M, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA. 2016; 316:2402–10.

4. Campanella G, Hanna MG, Geneslaw L, et al. Clinical-grade computational pathology using weakly supervised deep learning on whole slide images. Nat Med. 2019; 25:1301–9.

5. Liu Y, Gadepalli K, Norouzi M, et al. Detecting cancer metastases on gigapixel pathology images. Preprint at: https://arxiv.org/abs/1703.02442 . 2017.

6. Li C, Wang X, Liu W, Latecki LJ. DeepMitosis: mitosis detection via deep detection, verification and segmentation networks. Med Image Anal. 2018; 45:121–33.

7. Steiner DF, MacDonald R, Liu Y, et al. Impact of deep learning assistance on the histopathologic review of lymph nodes for metastatic breast cancer. Am J Surg Pathol. 2018; 42:1636–46.

8. Huang G, Liu Z, Van Der Maaten L, Weinberger KQ. Densely connected convolutional networks. Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit. 2017; 2017:4700–8.

9. Bianco S, Cadene R, Celona L, Napoletano P. Benchmark analysis of representative deep neural network architectures. IEEE Access. 2018; 6:64270–7.

10. Mormont R, Geurts P, Maree R. Comparison of deep transfer learning strategies for digital pathology. Proc Comput Soc IEEE Conf Comput Vis Pattern Recognit. 2018; 2018:2375–84.

11. Bankhead P, Loughrey MB, Fernandez JA, et al. QuPath: Open source software for digital pathology image analysis. Sci Rep. 2017; 7:16878.

12. Abadi M, Agarwal A, Barham P, et al. Tensorflow: large-scale machine learning on heterogeneous distributed systems [Internet]. Mountain View: TensorFlow;2015. [cited 2020 Jun 9]. Available from: http://tensorflow.org .

13. Hanna MG, Reuter VE, Samboy J, et al. Implementation of digital pathology offers clinical and operational increase in efficiency and cost savings. Arch Pathol Lab Med. 2019; 143:1545–55.