Skip to main content
Top

17-06-2024 | Neuroblastoma

Evolutionary Strategies AI Addresses Multiple Technical Challenges in Deep Learning Deployment: Proof-of-Principle Demonstration for Neuroblastoma Brain Metastasis Detection

Authors: Subhanik Purkayastha, Hrithwik Shalu, David Gutman, Andrei Holodny, Shakeel Modak, Ellen Basu, Brian Kushner, Kim Kramer, Sofia Haque, Joseph N. Stember

Published in: Journal of Imaging Informatics in Medicine

Login to get access

Abstract

Two significant obstacles hinder the advancement of Radiology AI. The first is the challenge of overfitting, where small training data sets can result in unreliable outcomes. The second challenge is the need for more generalizability, the lack of which creates difficulties in implementing the technology across various institutions and practices. A recent innovation, deep neuroevolution (DNE), has been introduced to tackle the overfitting issue by training on small data sets and producing accurate predictions. However, the generalizability of DNE has yet to be proven. This paper strives to overcome this barrier by demonstrating that DNE can achieve satisfactory results in diverse external validation sets. The main innovation of the work is thus showing that DNE can generalize to varied outside data. Our example use case is predicting brain metastasis from neuroblastoma, emphasizing the importance of AI with limited data sets. Despite image collection and labeling advancements, rare diseases will always constrain data availability. We optimized a convolutional neural network (CNN) with DNE to demonstrate generalizability. We trained the CNN with 60 MRI images and tested it on a separate diverse collection of images from over 50 institutions. For comparison, we also trained with the more traditional stochastic gradient descent (SGD) method, with the two variants of (1) training from scratch and (2) transfer learning. Our results show that DNE demonstrates excellent generalizability with 97% accuracy on the heterogeneous testing set, while neither form of SGD could reach 60% accuracy. DNE’s ability to generalize from small training sets to external and diverse testing sets suggests that it or similar approaches may play an integral role in improving the clinical performance of AI.
Literature
1.
go back to reference Stember JN, Shalu H. (2022). Deep neuroevolution squeezes more out of small neural networks and small training sets: Sample application to MRI brain sequence classification. International Symposium on Intelligent Informatics 2022 Aug 31 (pp. 153–167). Singapore: Springer Nature Singapore. Stember JN, Shalu H. (2022). Deep neuroevolution squeezes more out of small neural networks and small training sets: Sample application to MRI brain sequence classification. International Symposium on Intelligent Informatics 2022 Aug 31 (pp. 153–167). Singapore: Springer Nature Singapore.
2.
go back to reference Stember JN, Young RJ, Shalu H. (2023). Direct evaluation of treatment response in brain metastatic disease with deep neuroevolution. Journal of Digital Imaging. 2023 Apr;36(2):536–46. Stember JN, Young RJ, Shalu H. (2023). Direct evaluation of treatment response in brain metastatic disease with deep neuroevolution. Journal of Digital Imaging. 2023 Apr;36(2):536–46.
7.
go back to reference Eche, T., Schwartz, L. H., Mokrane, F. Z., & Dercle, L. (2021). Toward generalizability in the deployment of artificial intelligence in radiology: Role of computation stress testing to overcome underspecification. Radiology: Artificial Intelligence, 3(6). https://doi.org/10.1148/ryai.2021210097. Eche, T., Schwartz, L. H., Mokrane, F. Z., & Dercle, L. (2021). Toward generalizability in the deployment of artificial intelligence in radiology: Role of computation stress testing to overcome underspecification. Radiology: Artificial Intelligence, 3(6). https://​doi.​org/​10.​1148/​ryai.​2021210097.
10.
go back to reference Appenzeller, A., Bartholomäus, S., Breitschwerdt, R., Claussen, C., Geisler, S., Hartz, T., ... & Zeissig, S. R. (2021, November). Towards distributed healthcare systems–virtual data pooling between cancer registries as backbone of care and research. In 2021 IEEE/ACS 18th International Conference on Computer Systems and Applications (AICCSA) (pp. 1–8). IEEE. https://doi.org/10.1109/AICCSA53542.2021.9686918. Appenzeller, A., Bartholomäus, S., Breitschwerdt, R., Claussen, C., Geisler, S., Hartz, T., ... & Zeissig, S. R. (2021, November). Towards distributed healthcare systems–virtual data pooling between cancer registries as backbone of care and research. In 2021 IEEE/ACS 18th International Conference on Computer Systems and Applications (AICCSA) (pp. 1–8). IEEE. https://​doi.​org/​10.​1109/​AICCSA53542.​2021.​9686918.
11.
12.
17.
go back to reference DuBois, S. G., Kalika, Y., Lukens, J. N., Brodeur, G. M., Seeger, R. C., Atkinson, J. B., ... & Matthay, K. K. (1999). Metastatic sites in stage IV and IVS neuroblastoma correlate with age, tumor biology, and survival. Journal of pediatric hematology/oncology, 21(3), 181–189. https://doi.org/10.1097/00043426-199905000-00005. DuBois, S. G., Kalika, Y., Lukens, J. N., Brodeur, G. M., Seeger, R. C., Atkinson, J. B., ... & Matthay, K. K. (1999). Metastatic sites in stage IV and IVS neuroblastoma correlate with age, tumor biology, and survival. Journal of pediatric hematology/oncology, 21(3), 181–189. https://​doi.​org/​10.​1097/​00043426-199905000-00005.
18.
go back to reference Matthay, K. K., Brisse, H., Couanet, D., Couturier, J., Bénard, J., Mosseri, V., ... & Michon, J. (2003). Central nervous system metastases in neuroblastoma: radiologic, clinical, and biologic features in 23 patients. Cancer, 98(1), 155–165. https://doi.org/10.1002/cncr.11448. Matthay, K. K., Brisse, H., Couanet, D., Couturier, J., Bénard, J., Mosseri, V., ... & Michon, J. (2003). Central nervous system metastases in neuroblastoma: radiologic, clinical, and biologic features in 23 patients. Cancer, 98(1), 155–165. https://​doi.​org/​10.​1002/​cncr.​11448.
19.
go back to reference Kramer, K., Kushner, B. H., Modak, S., Pandit-Taskar, N., Smith-Jones, P., Zanzonico, P., ... & Cheung, N. K. V. (2010). Compartmental intrathecal radioimmunotherapy: results for treatment for metastatic CNS neuroblastoma. Journal of neuro-oncology, 97(3), 409–418. https://doi.org/10.1007/s11060-009-0038-7. Kramer, K., Kushner, B. H., Modak, S., Pandit-Taskar, N., Smith-Jones, P., Zanzonico, P., ... & Cheung, N. K. V. (2010). Compartmental intrathecal radioimmunotherapy: results for treatment for metastatic CNS neuroblastoma. Journal of neuro-oncology, 97(3), 409–418. https://​doi.​org/​10.​1007/​s11060-009-0038-7.
Metadata
Title
Evolutionary Strategies AI Addresses Multiple Technical Challenges in Deep Learning Deployment: Proof-of-Principle Demonstration for Neuroblastoma Brain Metastasis Detection
Authors
Subhanik Purkayastha
Hrithwik Shalu
David Gutman
Andrei Holodny
Shakeel Modak
Ellen Basu
Brian Kushner
Kim Kramer
Sofia Haque
Joseph N. Stember
Publication date
17-06-2024
Publisher
Springer International Publishing
Published in
Journal of Imaging Informatics in Medicine
Print ISSN: 2948-2925
Electronic ISSN: 2948-2933
DOI
https://doi.org/10.1007/s10278-024-01165-z